Innovation Funding Database
Choose Your Area of Innovation:
Advanced Materials & Manufacturing
Aerospace & Spacetech
Agtech & Foodtech
Artificial Intelligence & Machines Learning
Biotech
Cleantech & Climatetech
Cybersecurity
Defensetech & Dual-Use Tech
eXtended Reality
Healthtech
Medtech
Other Tech
Quantum & Photonics
Robotics & Autonomous Systems
NGA SBIR OSW26BZ06-DV033: Agentic AI Based Cognitive Radar for GEOINT Mission
Deadline: October 21, 2026
Funding Award Size: $1.5m
Description: Complete guide to NGA SBIR Direct to Phase II topic OSW26BZ06-DV033, agentic AI based cognitive radar for GEOINT. Up to $1.5M over 24 months. TS/SCI required. Closes October 21, 2026.
Quick Answer
OSW26BZ06-DV033 is a Direct to Phase II SBIR topic from the National Geospatial-Intelligence Agency under the 2026 SBIR Broad Agency Announcement, Release 6. Phase I proposals are not accepted. NGA wants a radar that senses its own operating environment and decides, without a human in the loop, what waveform to transmit: which frequency, which bandwidth, which pulse repetition frequency. The point is resilience, keeping GEOINT collection working when an adversary is actively trying to deny it. The award must not exceed $1,500,000 for up to a 24-month period of performance, and the technical volume runs to 40 pages. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
Before anything technical, understand what makes an NGA SBIR different from every other component in this cycle. NGA is a member of the U.S. Intelligence Community, and this work is classified. The instructions state that all contractor personnel shall possess a current Top Secret personnel security clearance and be eligible for favorable NGA adjudication for Sensitive Compartmented Information access. Contractors are subject to counterintelligence-scope polygraph examination. Your Key Personnel section must list the clearance level held by each person. Your Facilities section must state your Facility Clearance Level, safeguarding level, CAGE code, and physical addresses. And NGA adds an evaluation criterion beyond the standard ones: your ability to perform controlled work, meaning CUI and classified, is part of your technical evaluation score.
If your company does not hold a facility clearance and cleared staff, or does not have a credible plan and a sponsor for getting them, this topic is not accessible to you regardless of how good your radar work is. That is the honest first filter.
Two other structural notes. The offeror shall not propose option periods. And the technical volume is 40 pages, twenty for feasibility and twenty for the Phase II proposal, which is by far the most generous page allowance in the 2026 cycle and the largest feasibility allowance of any component.
Topic At a Glance
Topic number: OSW26BZ06-DV033
Title: Agentic AI Based Cognitive Radar for GEOINT Mission
Agency: National Geospatial-Intelligence Agency (NGA), a Department of War combat support agency and a member of the U.S. Intelligence Community
Solicitation: NGA 2026 SBIR Broad Agency Announcement, Release 6, Proposal Submission Instructions
Program type: Direct to Phase II only. This topic is accepting Direct to Phase II proposals only
Award: must not exceed $1,500,000
Period of performance: up to 24 months
Options: the offeror shall not propose option periods
Technical volume: 40 pages maximum, consisting of Part 1 Phase I Justification at 20 pages maximum and Part 2 Phase II Technical Proposal at 20 pages maximum. Pages in excess will not be considered. Number all pages consecutively
Component Technology Priority Areas: Integrated Sensing and Cyber, Trusted AI and Autonomy, Integrated Network Systems-of-Systems
Projected CMMC level requirement: Level 2 (Self)
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic. Note that the work itself is classified, which imposes far stricter personnel and facility requirements than an ITAR notice would
Clearance environment: all contractor personnel shall possess a current Top Secret personnel security clearance and be eligible for favorable NGA adjudication for SCI access. Contractors are subject to counterintelligence-scope polygraph as requested
Added evaluation criterion: NGA will evaluate a vendor's ability to perform controlled work, meaning CUI and classified, as part of the technical evaluation score
Section 508: the Commercialization Strategy shall address Section 508 compliance per NGA Instruction 8400.4 and Section 508 of the Rehabilitation Act, with an outline of how compliance will be achieved
Company Commercialization Report: information contained in the CCR will not be considered by NGA during proposal evaluations
Percentage of Work: NGA will not accept any deviation to the POW requirements
Technical and Business Assistance: up to $50,000 per Phase II project, in addition to the cost ceiling and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5
Contract type: NGA typically provides a firm fixed price contract within 180 days of the proposal due date, at the discretion of the Contracting Officer
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026
Submission portal: DSIP at dodsbirsttr.mil
Keywords: agentic AI, cognitive radar, synthetic aperture radar
The Feasibility Bar, Which Is the First Thing to Check
This topic is accepting Direct to Phase II proposals only. To qualify for a Phase II award, proposers must submit Feasibility Documentation showing a developed concept for agentic AI based cognitive radar systems, and provide details on the components of the systems and subsystems.
The performer must demonstrate capability of the agentic AI based radar system with the detection of interference signals and mitigating these by transmitting a different signal.
Two quantitative improvements are stated. The signal-to-interference-and-noise ratio should improve from below 70 percent to 95 percent. And after mitigation, image quality and resolution should improve from below 70 percent detection and classification to 95 percent detection and classification.
Reading the stated metrics carefully
The second metric is unambiguous and it is the one to build your evidence around. Detection and classification performance rising from below 70 percent to 95 percent after interference mitigation is a clean, measurable claim: run your system against an interference scenario without mitigation, measure detection and classification rates, then run it with the agentic mitigation loop engaged and measure again.
The first metric is worded in a way that deserves a question. Signal-to-interference-and-noise ratio is conventionally expressed in decibels, not as a percentage, and "SINR below 70 percent to 95 percent" does not map onto standard usage. It may be intended as a normalized or relative improvement figure, or as a percentage of some reference SINR, or the percentage may be a drafting artifact carried from the detection and classification metric that follows it.
Send that question to DSIP Topic Q&A before it closes on October 7, and in the meantime report your SINR results in decibels with the interference conditions fully specified, alongside whatever normalized figure best corresponds to the stated 70 to 95 range. Stating both, and saying plainly why, is the defensible approach. Do not silently reinterpret the requirement.
The three things your feasibility documentation must establish
A developed concept for an agentic AI based cognitive radar system. Not a research proposal, a concept that exists.
Details on the components of the systems and subsystems. This is a request for architecture: what senses the environment, what performs the cognition, what synthesizes and transmits the waveform, and how they connect.
Demonstrated capability detecting interference and mitigating it by transmitting a different signal. This is the closed loop, and it is the part that separates cognitive radar from adaptive signal processing. Sense, decide, change what you transmit, measure the improvement.
The restriction that will disqualify some proposers
NGA applies the same hard constraint several other components use in this cycle, and it is stated in the Direct to Phase II guidelines rather than in the topic.
Feasibility documentation cannot be based upon or logically extend from any prior or ongoing federally funded SBIR or STTR work. Work submitted within the feasibility documentation must have been substantially performed by the proposer or the principal investigator. If technology in the feasibility documentation is subject to intellectual property, the proposer must either own the IP or must have obtained license rights to such technology prior to proposal submission, to enable it and its subcontractors to legally carry out the proposed work.
This is a real risk on a cognitive radar topic, because much of the U.S. work in this area has been funded through SBIR and STTR programs across the services. Audit the provenance of every result you intend to cite. Internally funded work, privately funded work, non-SBIR government contract work, and your own published academic work under other funding are cleaner ground.
If the proposer fails to demonstrate technical merit and feasibility equivalent to the Phase I level as described in the associated topic, the related Phase II proposal will not be evaluated.
Twenty pages is a lot of room, so use it
Most components in this cycle allow five pages of feasibility justification. NGA allows twenty. That changes the nature of the document: you can present measured data with full experimental conditions, architecture diagrams at the subsystem level, the interference scenarios you tested against, the decision logic of your cognition layer, and the before-and-after performance comparisons in detail. A five-page summary submitted into a twenty-page allowance reads as though you did not have the material.
What NGA Is Actually Looking For
The objective
Develop an agentic AI based cognitive radar system for automatic selection of radar parameters, waveforms, and exploitation for resilient GEOINT applications.
Note the third item. Parameters and waveforms are the transmit side. Exploitation is the processing side. The topic is asking for automatic selection across both, which means the cognition loop extends past waveform choice into how the returns are processed.
The concept, in NGA's words
Agentic AI enables autonomous control of expert systems for desired information gathering or completing a task. Compared to general AI systems, an agentic AI system can make its decision by sensing the environment autonomously without relying on human input. Hence agentic AI is desired for many expert systems such as self-driving cars and many other systems.
In this research, NGA is seeking development of an agentic AI based cognitive radar system for a next generation, resilient radar system.
Once initiated, an agentic AI enabled radar will sense the environment for operating conditions such as interference, clutter, and signal-to-noise ratio. Based on this information, the system will learn, meaning the cognition process, to transmit optimum waveforms with desired frequency, bandwidth, pulse repetition frequency, and other critical parameters.
This will enable future GEOINT capabilities to be resilient in the presence of asymmetric operating environments that adversaries may impose to significantly reduce our GEOINT mission.
What the description tells you to address
Three environmental conditions are named explicitly: interference, clutter, and signal-to-noise ratio. Your sensing layer should address all three, not just interference, even though interference is what the feasibility metric measures.
Four transmit parameters are named: frequency, bandwidth, pulse repetition frequency, and other critical parameters. Say which parameters your system controls and which it does not, and why.
The phrase "once initiated" is worth noticing. It implies human initiation followed by autonomous operation, which is a specific autonomy boundary. Define it: what does a human set up, what does the system decide thereafter, and what does the system do when conditions exceed its competence.
"Asymmetric operating environments that adversaries may impose" is the threat framing. This is a contested-spectrum problem, not a general radar optimization problem. An adversary observes what you transmit and responds. That adversarial dynamic is what makes a learning system valuable and it is also what makes it exploitable, so address the counter-adaptation question: what happens when the adversary learns your adaptation policy.
The GEOINT context, which shapes the application
NGA describes GEOINT as the exploitation and analysis of imagery and geospatial information to describe, assess, and visually depict physical features and geographically referenced activities on the Earth, consisting of imagery, imagery intelligence, and geospatial information. NGA manages the National System for Geospatial-Intelligence.
NGA Research supports the NSG and the National Security Strategy in three broad areas: Foundational GEOINT, Advanced Phenomenologies, and Analytic Technologies. A cognitive radar topic sits naturally in Advanced Phenomenologies.
The keyword list for this topic includes synthetic aperture radar, and the feasibility metric refers to image quality and resolution, so the imaging application is where NGA's interest lies. Frame your work in imaging terms, meaning resolution, image quality, and detection and classification performance on imagery, rather than purely in tracking or detection-range terms.
Phase II Scope
Prototype the agentic AI based cognitive radar system in a modeling and simulation environment. Develop software systems to select various parameters of the radar systems and operating environment. Analyze performance of the system in a testing environment.
What this scope does and does not include
Phase II is modeling, simulation, and software. There is no hardware build, no field demonstration, and no over-the-air requirement. Phase III is where the hardware happens: develop and build the hardware for the agentic AI based cognitive radar system for technology transition.
That is an unusually software-focused Phase II for a radar topic and it has consequences worth understanding.
It lowers the capital cost substantially. No radio frequency front end, no antenna, no anechoic chamber or range time. Your $1,500,000 goes almost entirely to labor, computing, and software.
It raises the burden on fidelity. If your only evidence is simulation, the credibility of your results rests entirely on whether the simulation environment is believable. What clutter model, what interference model, what channel model, what platform geometry, what validation of the simulation against measured data. A reviewer at NGA Research will probe this, and a proposal that presents impressive numbers from an unvalidated simulator is weaker than one presenting modest numbers from a simulator anchored to real measurements.
It makes the "testing environment" of the third sentence important. The topic distinguishes the modeling and simulation environment where you prototype from the testing environment where you analyze performance. Say what each is and how they differ, because that distinction is where independent evaluation of your system happens.
Three things worth proposing beyond the literal scope
The software system that selects parameters is named as a deliverable, and it is the artifact that transitions. Treat it as a product with an interface, a configuration model, and documentation, not as research code.
Explainability is not mentioned in the topic but is implied by the Trusted AI and Autonomy Component Technology Priority Area. An autonomous system choosing transmit parameters in a contested environment needs to be able to say why it chose them, both for operator trust and for post-mission analysis. Addressing it is a differentiator.
Competence boundaries and failure behavior. Define what the system does when the environment is outside its training distribution, when its own performance estimate degrades, or when its adaptation makes things worse. In an intelligence collection context, a system that recognizes it is failing and says so is more valuable than one that confidently continues.
The Security Requirements, Which Govern Everything
This is the longest section of the NGA instructions and it is the part that most determines who can actually win. Treat it as a scoped, costed part of your proposal rather than a compliance appendix, because NGA has made it a scored evaluation factor.
The added evaluation criterion
In addition to the DoW Program evaluation criteria, NGA will evaluate a vendor's ability to perform controlled work, meaning CUI and classified, as part of their technical evaluation score.
Read that plainly. Your security posture is scored technical content. Two proposals with identical radar approaches will separate on this.
Note also that NGA may use Systems Engineering and Technical Assistance support to assist with the programmatics of executing the evaluation process, but will not use SETA support to evaluate proposals.
Personnel security
All Contractor personnel shall possess a current Top Secret Personnel Security Clearance and be eligible for favorable NGA adjudication for SCI access. The government will be responsible for verifying security clearances and SCI eligibility in accordance with SEAD 7 on reciprocity of background investigations and national security adjudications. If needed, NGA will sponsor SCI accesses at its sole discretion.
Contractor personnel performing Top Secret/Sensitive Compartmented Information work on the contract are required to have active TS/SCI clearances for access to TS/SCI facilities, when performing duties within TS/SCI environments, and for access to TS/SCI computer systems.
Contractors are subject to a counterintelligence-scope polygraph examination as requested by NGA. As a condition of employment and assignment, contractors who have not successfully completed polygraph testing within the last five years must immediately schedule a polygraph examination and must complete the process, in no more than three test sessions, within 90 days.
Cleared Contractor personnel will be enrolled into the Director of National Intelligence Continuous Evaluation System throughout the lifecycle of the contract while at NGA, in accordance with SEAD 6. They must self-report security issues, foreign contacts, and other applicable information in accordance with SEAD 3 and NGA Instruction 5205.3. Cleared contractors will also be required to submit electronic fingerprints and be enrolled into NGA's Report of Arrest and Prosecution Background program, which supports continuous evaluation.
Cleared Contractor personnel designated as Tier 3 Privileged Users, or holding enhanced access through a Special Access Program or Controlled Access Program, will be required to participate in NGA's annual Security Financial Disclosure Program.
Some personnel may need to obtain access to Special Access Program, Controlled Access Program, or Alternate Compensatory Control Measures information as required for the execution of NGA's sensitive program missions. Access will not be granted without written permission from the Contracting Officer's Representative.
Defense Counterintelligence and Security Agency remains the Cognizant Security Agency for individuals performing work with collateral Secret or Top Secret clearances, and SEAD 3 reporting continues through the contractor's Facility Security Officer to DCSA via the Defense Information Security System.
What that means practically for a small business
The polygraph provision is the one most likely to surprise a first-time NGA bidder. Anyone not polygraphed in the last five years must schedule immediately and complete within 90 days, in no more than three sessions. That is a real constraint on staffing a project team, and it applies to people you may currently consider cleared.
The Top Secret requirement applies to "all Contractor personnel," which is broader than "personnel accessing classified material." Read together with the unclassified provisions below, there is a path for uncleared staff on the unclassified portion of the work, but it requires written COR approval.
Your Key Personnel section must list the clearance level held by each person, because, as the instructions put it, we will be operating in a classified environment. Do not omit this. A Key Personnel section without clearance levels is a visible gap on a scored criterion.
Physical security and facilities
All classified work performed at a non-NGA facility must be approved by the COR. Any classified work performed at contractor sites must be performed in either an NGA accredited Sensitive Compartmented Information Facility, an Other Government Agency SCIF, or an approved secure collateral space that has a Memorandum of Agreement, Memorandum of Understanding, Joint Use Agreement, or Co-Use Agreement with NGA for this contract. SCIF or secure facility accreditation shall be at least the level commensurate with the level of safeguarding required by the contract DD Form 254.
Accreditation follows the IC Technical Specifications for Construction and Management of Sensitive Compartmented Information Facilities Version 1.5.1, the IC Tech Specs for ICD and ICS 705. The NGA accrediting official or designee conducts periodic security inspections and reviews based on threat, facility modifications, sensitivity of programs, past security performance, or at least every five years. Technical Surveillance Countermeasures activities in NGA accredited SCIFs will only be conducted by NGA TSCM teams.
Contractor personnel are forbidden from bringing prohibited or unauthorized items into any NGA installation or other secure facility covered under this contract. These items include weapons, cell phones, cameras, two-way pagers, laptops, recording devices of any kind, flash drives, or any other removable media. Exceptions may be granted by NGA Security upon request, and personnel must bring documentation showing approval prior to entering with the items. Violations may subject the contractor and its personnel to civil or criminal liability.
Your Facilities and Equipment section must identify your Facility Clearance Level, the safeguarding level for each facility, the CAGE code for each facility, and the associated physical addresses. It must also identify how Controlled Unclassified Information will be protected in accordance with NIST SP 800-171 and provide your Supplier Performance Risk System score.
That SPRS score requirement is worth flagging. It is a number a reviewer can look at immediately, and a low or absent score is a visible weakness on the scored controlled-work criterion. Check your SPRS entry before you submit.
The unclassified path, which matters for team building
If applicable, and after discussion with the COR and written COR approval, uncleared Contractor personnel are authorized to work on this contract at the unclassified level, with access up to DoD Controlled Unclassified Information at the contractor site without the requirement of a security clearance.
Any Contractor personnel working with CUI information must receive a favorable HSPD-12 or HSPD-12 Tier 1 adjudication prior to accessing CUI. Personnel requiring access to CUI for 60 days or less must receive a favorable HSPD-12 adjudication. Personnel requiring CUI access for more than 60 days must receive a favorable HSPD-12 Tier 1 adjudication. NGA will sponsor the HSPD-12 and HSPD-12 Tier 1 background investigation for contractor personnel as needed.
Uncleared Contractor personnel visiting NGA facilities or other sites may receive an appropriate visitor badge and be escorted as appropriate, returning the badge at the end of each visit day.
This is the provision that makes a mixed team possible. Your machine learning engineers may be able to work the unclassified algorithm development while cleared staff handle anything requiring access to classified data or environments. But note the gating: written COR approval, and HSPD-12 adjudication before CUI access.
Foreign nationals
Foreign nationals are not permitted to perform unclassified work under the terms of this contract without prior written approval from the Contracting Officer or the COR. Should NGA identify the use of unauthorized personnel, the CO may direct the contractor, at its own expense, to remove and replace any unauthorized contractors, subcontractors, or other personnel performing on the contract, at NGA's discretion and without prejudice to its rights under any other contract provision, including termination for default.
Note the scope. This prohibition applies to unclassified work, which is stricter than most ITAR-restricted topics, where foreign national participation is disclosed and evaluated case by case. Here the default is not permitted, and approval must be obtained in writing in advance. If your machine learning team includes foreign nationals, resolve this before you propose.
Information security and cybersecurity
NGA has sole authority to determine whether and to what extent protected information will be provided. Access to classified information will be pursuant to the security requirements in the DD254. The contractor shall not access, download, print, or further disseminate any classified information outside the execution of defined contract requirements without written COR permission.
The contractor will comply with all applicable NGA, DoD, and IC information security policies, including the Consolidated NGA Security Classification Guide, for marking, handling, processing, storing, and safeguarding classified and unclassified material, and will give document markings the lowest possible security classification to maximize dissemination while maintaining confidentiality and integrity.
Access to CUI will be pursuant to DFARS clause 252.204-7012, which the prime shall include, including paragraph (m), in subcontracts for operationally critical support or involving covered defense information, without alteration except to identify the parties. FAR clause 52.204-21 must similarly flow down.
Contractor personnel shall not release any unclassified information pertaining to any part of this contract or any related program, in any medium, unless the COR has given prior written approval or in performance of a project scoped and negotiated by NGA. That constrains publication and marketing, and it is worth understanding before you plan a commercialization campaign around this work.
At minimum the contractor must implement 10 U.S.C. Sections 391 and 393 along with NIST SP 800-171. If using an external cloud service provider to store, process, or transmit covered defense information, the contractor shall require and ensure the provider meets security requirements equivalent to the FedRAMP Moderate baseline.
On discovering a cyber incident affecting a covered contractor information system or covered defense information, the contractor shall notify the DoD Cyber Crime Center and the COR in writing within 72 hours of incident discovery, and conduct a review for evidence of compromise. If malicious software is discovered and isolated, the contractor shall submit it to DC3 per COR instructions, and shall not send it to the COR.
For a machine learning program, the FedRAMP Moderate cloud requirement is the provision most likely to affect your technical plan and your budget. If your training pipeline assumes a commercial cloud environment, confirm that environment meets the equivalence standard, because retrofitting compute infrastructure mid-program is expensive.
Insider threat
The contractor will establish and maintain an insider threat program to gather, integrate, and report relevant and available information indicative of a potential or actual insider threat, consistent with Executive Order 13587 and the Presidential Memorandum on National Insider Threat Policy and Minimum Standards for Executive Branch Insider Threat Programs.
As soon as practicable, the contractor shall report to the COR events that may affect the eligibility of the entity or an employee for access to classified information, events indicating an insider threat, events affecting proper safeguarding, and events indicating classified information has been or is suspected to be lost or compromised.
Contract clauses to be aware of
The instructions reproduce or reference a set of clauses that will apply. FAR 52.204-7 System for Award Management, requiring registration at offer and continuously through final payment. FAR 52.204-27 Prohibition on a ByteDance Covered Application, meaning TikTok and successor applications.
NGA-specific clauses include 5X52.209-9003 Protection of Information and Nondisclosure Agreements, 5X52.227-9000 Unauthorized Use of NGA Name, Seal and Initials, 5X52.237-9001 Contractor Identification, 5X52.37-9000 Contractor Employee Data for Access to NGA Facilities or Sensitive Systems, which requires initial and timely updates to NGA's Human Capital Management System for all personnel with access to NGA facilities or sensitive systems and requires all employees to attend in-processing and out-processing briefings, and 5X52.246-9000 Contractor Compliance with all applicable NGA and U.S. Government installation regulations, directives, instructions, rules, policies and procedures.
The Unauthorized Use of NGA Name, Seal and Initials clause is worth reading before you plan any marketing around an award. So is the prohibition on releasing unclassified information about the contract without written COR approval. Together they substantially constrain how you can talk publicly about this work, which is a real consideration for a company whose commercialization strategy depends on visibility.
There is also an Inspector General cooperation requirement. The contractor must report to the NGA Inspector General, DoD IG, or Intelligence Community IG any and all possible violations of federal law or illegal intelligence activities related to the contract by individuals charging directly or indirectly to it. The IG has access to any such individual whose testimony is needed and direct access to all related records. Failure to cooperate is grounds for administrative action. Contractors must make employees aware of the NGA IG Hotline, and the requirement is supported by FAR 52.203-13 and NGA Instruction 7410.1.
Badging is governed too. NGA IC badges will only be issued to contractors providing direct charge support on an active TS/SCI NGA contract, even when seated at corporate locations outside NGA facilities. Badges must be used at least once during a one-month period at an NGA government facility or may be suspended or terminated for lack of activity, and they expire at the end of the supported contract. Badges will not be issued to contractors who do not need facility access. On departure, the contractor must return all NGA badges, Common Access Cards, hangtags, and other government furnished property no later than four business days from the date of departure.
Section 508 Compliance, Which Is a Proposal Requirement
This requirement is easy to miss because it sits inside the Commercialization Strategy instruction, and it is unusual among the components in this cycle.
The contractor shall ensure that all systems, hardware, software, software engineering, and information technology associated with this effort is made in a manner that is accessible for people with disabilities as directed in NGA Instruction 8400.4 and Section 508 of the Rehabilitation Act of 1973 as amended in 1998.
Specifically, all Information and Communications Technology associated with this contract may use the Web Content Accessibility Guidelines 2.1 to comply with Section 508, or use alternative designs or technologies which result in substantially equivalent or greater access to and use of the product for people with disabilities.
Furthermore, the contractor shall pursue human centered design and usability guidelines to ensure that all services associated with this topic area are accessible by as many users as possible and to drive modernization, innovation, and enhance mission support.
As part of the proposal, the offeror should include an outline of specifically how Section 508 compliance will be achieved in the design of the ICT product. The Phase II proposal should provide an explicit, detailed description of the approach, indicate what is planned, how and where the work will be carried out, a schedule of major events, how the solution will be Section 508 compliant, and the final product to be delivered. If a determination is made that a Section 508 exception request is justified, the rationale for the exception request must be made and submitted as part of the proposal.
How to handle it on this topic
A cognitive radar control system is mostly autonomous software, but the Phase II deliverable includes software systems to select radar parameters and to analyze performance, which implies an operator or analyst interface. Any interface is Information and Communications Technology and falls within scope.
The practical answer is short and specific: name WCAG 2.1 as your standard, describe how your interface elements will conform, whether keyboard navigation, screen reader compatibility, contrast, and non-color-dependent status indication, and say when in the schedule accessibility review happens. If some component genuinely cannot comply, the instructions give you an exception path but require the rationale in the proposal, so use it deliberately rather than staying silent.
Silence is the failure mode here. The instructions say the offeror should include an outline of specifically how compliance will be achieved, and this is a requirement other components in this cycle do not impose, so it is exactly the kind of thing a proposal reused from another agency will omit.
Funding, Cost Structure, and NGA Mechanics
The award
The Phase II amount must not exceed $1,500,000 for up to a 24-month period of performance. Costs must be separated and clearly identified on the Proposal Cover Sheet in Volume 1 and in Volume 3.
Unless otherwise stated in the individual topic announcement, NGA Phase II awards are capped at $1,500,000 each over a maximum 24-month period of performance, and Phase I awards are capped at $150,000 each over a maximum six-month period of performance.
NGA caps sequential Phase II contracts, meaning those proposed near the completion of the initial Phase II or Direct to Phase II contract, at the then-current Small Business Administration "without seeking SBA approval" ceiling over a maximum 24-month period of performance.
No option periods
The offeror shall not propose option periods.
That is a flat prohibition and it distinguishes NGA from most components in this cycle, several of which structure awards as a base plus options. Build a single 24-month effort. Do not carry an option structure over from another agency's proposal.
Contract type and timing
NGA typically provides a firm fixed price contract for its awards within 180 days of the proposal due date. The type of contract is at the discretion of the Contracting Officer.
One hundred eighty days from October 21, 2026 is roughly mid-April 2027. Plan your cash flow and staffing accordingly, and note that firm fixed price on a research effort means your cost estimate needs margin for the simulation fidelity work, which is the least predictable part of this scope.
Percentage of Work, with no exceptions
Review the updated Percentage of Work calculation details included in the DoW Program. NGA will not accept any deviation to the POW requirements.
On this topic the temptation is to subcontract the radar modeling to a specialist firm or the machine learning to a university. Model your POW before you assemble the team, and remember that every subcontractor also inherits the security requirements and the DFARS 252.204-7012 and FAR 52.204-21 flow-downs.
Technical and Business Assistance
Phase II awardees may request up to $50,000 per Phase II project. TABA funding is in addition to the Phase II cost ceiling and is not subject to profit or fee.
All requests for TABA must be completed using the SBIR/STTR TABA Request Form, and the completed form must be included in Volume 5 of the proposal submission in DSIP. OSW will not accept requests for TABA that do not utilize the form or that are not included as a submission document in Volume 5.
For an NGA topic, security and facility clearance consulting is the standout use if you are building that posture, followed by accessibility and Section 508 consulting, which is a stated proposal and performance requirement here and not one most small businesses staff internally.
The Company Commercialization Report is not evaluated
Completion of the CCR as Volume 4 is required, but information contained in the CCR will not be considered by NGA during proposal evaluations.
That differs from several other components in this cycle, where the CCR is explicitly scored. Complete it because it is required, but your effort belongs in the Commercialization Strategy inside the technical volume, which is where NGA actually reads your transition thinking and where the Section 508 requirement lives.
The technical volume structure
The technical volume consists of two parts, described in one place as Part A Feasibility Documentation and Part B Technical Proposal and in the volume list as Part 1 Phase I Justification and Part 2 Phase II Technical Proposal. Either way the structure is the same: each part is not to exceed twenty pages, for a technical volume maximum page count of 40 pages. The Government will not consider pages in excess of these limitations. Number all pages consecutively and follow the formatting requirements provided in the DoW SBIR Program.
The content of the technical volume follows the Technical Proposal Template provided in the DoW Program, with the additional items NGA specifies: Section 508 compliance in the Commercialization Strategy, clearance levels in Key Personnel, and CUI protection with SPRS score plus facility clearance level, safeguarding level, CAGE code, and addresses in Facilities and Equipment.
Evaluation, selection, and protests
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW Program Solicitation, plus the controlled-work criterion described above.
Proposing firms will be notified of selection or non-selection status within 90 days of the closing date of the topic, via email. Note that the NGA text says "for a Phase I award," which appears to be residual language given that this topic issues no Phase I award. Ninety days from October 21, 2026 is approximately January 19, 2027.
Refer to the DoW Program Solicitation for procedures to protest the announcement. As prescribed in FAR 33.106(b) and FAR 52.233-3, protests after award should be submitted to Patricia Hill at Patricia.D.Hill@nga.mil.
Questions
Specific questions pertaining to the administration of the NGA SBIR/STTR Program and these proposal preparation instructions should be directed to sbir@nga.mil.
The NGA instructions do not state that DSIP Topic Q&A is unavailable, so the standard DoW SBIR Program BAA Topic Q&A process applies and Topic Q&A closes to new questions two weeks before the topic closes, on October 7, 2026. Use it for the SINR metric question above all.
The References
Three, and unlike several components in this cycle they are substantive and well chosen. Together they define the intellectual lineage NGA expects you to know.
Haykin, "New generation of radar systems enabled with cognition," IEEE International Radar Conference, Arlington, 2010. This is the foundational cognitive radar paper. Haykin framed cognitive radar as a closed perception-action cycle with memory and attention, and if your architecture does not map onto that framing you should explain why.
Guerci, "Cognitive Radar: The Knowledge-Aided Fully Adaptive Approach," Artech House, 2010. This is the book, and its central idea, knowledge-aided fully adaptive processing, is the practical engineering realization of the concept. It is also the reference most directly relevant to the "exploitation" half of the objective, since knowledge-aided processing operates on the receive side.
Bell, Johnson, Smith, Baker, and Rangaswamy, "Cognitive radar for target tracking using a software defined radar system," 2015 IEEE Radar Conference. This is the experimental demonstration, on software defined radar hardware, and it is the closest of the three to what a Phase III hardware effort would look like.
The set tells you something. NGA cited the concept, the engineering framework, and the hardware demonstration, all from the pre-deep-learning era of cognitive radar. Your contribution is the agentic AI layer on top of that lineage, so position it explicitly: what does an agentic AI decision layer do that knowledge-aided fully adaptive processing did not, and why is that the right addition now. A proposal that reinvents cognitive radar without engaging Haykin and Guerci will look uninformed to a reviewer who chose those references.
Bring your own literature too, on the agentic side. The topic gives you no citations for agentic AI, reinforcement learning for waveform selection, or adversarial machine learning in contested spectrum, and the DoW Technical Proposal Template requires you to demonstrate awareness of the state of the art.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
DSIP Topic Q&A closes: October 7, 2026, two weeks before the topic closes, per the DoW SBIR Program BAA
Proposal deadline: October 21, 2026
Selection notification: within 90 days of the closing date, approximately January 19, 2027, via email
Contract award: NGA typically provides a firm fixed price contract within 180 days of the proposal due date, approximately mid-April 2027
Period of performance: up to 24 months
A working backward plan
Before September 23. Confirm your security posture honestly, because it is a scored criterion and a gating one. Facility Clearance Level, safeguarding level for each facility, CAGE codes, physical addresses, SPRS score, and NIST SP 800-171 status. Identify which personnel hold current Top Secret clearances and who has been polygraphed within five years, since anyone who has not must complete a polygraph in no more than three sessions within 90 days. Resolve any foreign national participation, remembering the prohibition extends to unclassified work absent prior written approval. Audit the funding provenance of every feasibility result, since work based upon or logically extending from prior federally funded SBIR or STTR work is excluded. Resolve intellectual property ownership or license rights. Confirm your compute environment meets FedRAMP Moderate equivalence if you will use a cloud provider for covered defense information. Read Haykin, Guerci, and Bell. Model your Percentage of Work before assembling a team. Send your SINR metric question to Topic Q&A early.
September 23 through October 5. Draft the 20-page feasibility documentation. Use the space: measured interference detection and mitigation results with full experimental conditions, the before-and-after detection and classification figures against the 70 to 95 percent target, SINR reported in decibels alongside any normalized figure, subsystem-level architecture, and the decision logic of your cognition layer. Draft the 20-page Phase II technical proposal against the DoW Technical Proposal Template, covering the modeling and simulation environment and its fidelity basis, the parameter selection software as a product, the testing environment and how it differs from the simulation environment, explainability, and competence boundaries. Include the Section 508 outline in the Commercialization Strategy, clearance levels in Key Personnel, and the full facility and CUI detail in Facilities and Equipment. Do not propose option periods.
October 6 through October 7. Submit remaining questions through DSIP Topic Q&A before it closes. The essential one is the SINR percentage metric. Administrative questions go to sbir@nga.mil.
October 8 through October 14. Build the cost volume, with costs separated and clearly identified on the Proposal Cover Sheet and in Volume 3, against the $1,500,000 and 24-month ceiling with no options. Price the simulation environment development and validation, the machine learning development and compute, any secure compute or facility costs, security administration including HSPD-12 sponsorship coordination and insider threat program maintenance, accessibility work, and the software productization effort. Remember firm fixed price is typical, so carry margin on the simulation fidelity work. Complete the SBIR/STTR TABA Request Form and place it in Volume 5.
October 15 through October 18. Complete Volume 4, the Company Commercialization Report, which is required though not evaluated by NGA. Assemble Volume 5 with the TABA form and any supporting documentation. Complete Volume 6 training and the Volume 7 foreign affiliations webform, remembering it must be the webform and will not be accepted as a PDF in Volume 5, and that no previous versions should be uploaded there. Run compliance: 20 plus 20 pages, consecutively numbered, no option periods proposed, clearance levels listed, SPRS score and facility details provided, Section 508 outline present.
October 19 through October 20. Submit and certify in DSIP.
Frequently Asked Questions
What is NGA SBIR topic OSW26BZ06-DV033?
OSW26BZ06-DV033 is a Direct to Phase II SBIR topic titled "Agentic AI Based Cognitive Radar for GEOINT Mission," released under the National Geospatial-Intelligence Agency 2026 SBIR Broad Agency Announcement, Release 6. The objective is to develop an agentic AI based cognitive radar system for automatic selection of radar parameters, waveforms, and exploitation for resilient GEOINT applications.
How much funding is available?
The Phase II amount must not exceed $1,500,000 for up to a 24-month period of performance. Phase II awardees may also request up to $50,000 in Technical and Business Assistance, in addition to the cost ceiling and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5.
When is the proposal deadline?
The topic opens September 23, 2026 and proposals are due October 21, 2026 through the Defense SBIR/STTR Innovation Portal at dodsbirsttr.mil.
Can I submit a Phase I proposal?
No. This topic is accepting Direct to Phase II proposals only.
Can I propose option periods?
No. The NGA instructions state that the offeror shall not propose option periods. Build a single effort within the 24-month ceiling.
How long can my technical volume be?
Forty pages total, consisting of a feasibility documentation part at 20 pages maximum and a technical proposal part at 20 pages maximum. Pages in excess will not be considered. Number all pages consecutively and follow the DoW SBIR Program formatting requirements. This is the most generous page allowance among the components in this cycle.
What must my feasibility documentation show?
A developed concept for an agentic AI based cognitive radar system, with details on the components of the systems and subsystems, and demonstrated capability detecting interference signals and mitigating them by transmitting a different signal. Two quantitative improvements are stated: signal-to-interference-and-noise ratio improving from below 70 percent to 95 percent, and image quality and resolution improving from below 70 percent detection and classification to 95 percent detection and classification after mitigation.
The SINR metric is expressed as a percentage. Is that right?
It is what the topic says, but signal-to-interference-and-noise ratio is conventionally expressed in decibels rather than as a percentage, so the stated "below 70 percent to 95 percent" does not map onto standard usage. Raise it through DSIP Topic Q&A before it closes on October 7. In the meantime report SINR in decibels with interference conditions fully specified, alongside whatever normalized figure corresponds to the stated range, and say why you are presenting both.
Can my feasibility evidence come from a prior SBIR award?
No. The NGA Direct to Phase II guidelines state that feasibility documentation cannot be based upon or logically extend from any prior or ongoing federally funded SBIR or STTR work, and that the work must have been substantially performed by the proposer or the Principal Investigator. This is a real risk on a cognitive radar topic, since much U.S. work in the area has been SBIR and STTR funded.
What happens if my feasibility documentation is inadequate?
If the proposer fails to demonstrate technical merit and feasibility equivalent to the Phase I level as described in the associated topic, the related Phase II proposal will not be evaluated.
Do I need a security clearance?
Yes. The instructions state that all Contractor personnel shall possess a current Top Secret Personnel Security Clearance and be eligible for favorable NGA adjudication for SCI access. Personnel performing TS/SCI work require active TS/SCI clearances for access to TS/SCI facilities, environments, and computer systems. NGA will sponsor SCI accesses at its sole discretion if needed.
Is a polygraph required?
Contractors are subject to a counterintelligence-scope polygraph examination as requested by NGA. As a condition of employment and assignment, contractors who have not successfully completed polygraph testing within the last five years must immediately schedule an examination and complete the process, in no more than three test sessions, within 90 days.
Can uncleared people work on this contract?
Yes, in a limited way. After discussion with the COR and with written COR approval, uncleared contractor personnel are authorized to work at the unclassified level with access up to DoD Controlled Unclassified Information at the contractor site without a security clearance. Personnel working with CUI must receive favorable HSPD-12 adjudication for access of 60 days or less, or HSPD-12 Tier 1 adjudication for more than 60 days. NGA will sponsor those investigations as needed.
Can foreign nationals work on this contract?
Not without prior written approval from the Contracting Officer or the COR, and that prohibition applies to unclassified work. If NGA identifies unauthorized personnel, the CO may direct the contractor at its own expense to remove and replace them, without prejudice to other remedies including termination for default. This is stricter than the case-by-case disclosure approach used on ITAR-restricted topics elsewhere in this cycle.
Do I need a facility clearance and a SCIF?
Any classified work performed at contractor sites must be performed in an NGA accredited SCIF, an Other Government Agency SCIF, or an approved secure collateral space with a Memorandum of Agreement, Memorandum of Understanding, Joint Use Agreement, or Co-Use Agreement with NGA for this contract, and all classified work at a non-NGA facility must be COR approved. Accreditation must be at least commensurate with the safeguarding level required by the DD254.
What must my Key Personnel section include?
In addition to the standard content, list the clearance level held by each of the personnel, because, as NGA puts it, we will be operating in a classified environment.
What must my Facilities and Equipment section include?
How Controlled Unclassified Information will be protected in accordance with NIST SP 800-171, your Supplier Performance Risk System score, and, since the work requires a classified environment, your Facility Clearance Level, the safeguarding level for each facility, the CAGE code for each facility, and the associated physical addresses.
Is my security posture actually scored?
Yes. In addition to the DoW Program evaluation criteria, NGA will evaluate a vendor's ability to perform controlled work, meaning CUI and classified, as part of their technical evaluation score. Note also that NGA may use SETA support for the programmatics of executing the evaluation but will not use SETA support to evaluate proposals.
What is the Section 508 requirement?
The Commercialization Strategy must address Section 508 compliance. All systems, hardware, software, software engineering, and information technology associated with the effort must be accessible for people with disabilities per NGA Instruction 8400.4 and Section 508 of the Rehabilitation Act of 1973 as amended in 1998. Information and Communications Technology may use WCAG 2.1 to comply, or alternative designs achieving substantially equivalent or greater access. The proposal should include an outline of specifically how compliance will be achieved, and if an exception is justified the rationale must be submitted as part of the proposal.
What are the cybersecurity requirements?
At minimum, implement 10 U.S.C. Sections 391 and 393 along with NIST SP 800-171. If using an external cloud service provider to store, process, or transmit covered defense information, ensure the provider meets security requirements equivalent to the FedRAMP Moderate baseline. On discovering a cyber incident, notify the DoD Cyber Crime Center and the COR in writing within 72 hours and conduct a compromise review. Isolated malicious software goes to DC3 per COR instructions, not to the COR.
Do I need an insider threat program?
Yes. The contractor will establish and maintain an insider threat program consistent with Executive Order 13587 and the Presidential Memorandum on National Insider Threat Policy and Minimum Standards, and report to the COR events affecting clearance eligibility, indicating insider threat, affecting safeguarding, or indicating classified information has been or is suspected lost or compromised.
What does Phase II actually build?
Software, not hardware. Prototype the agentic AI based cognitive radar system in a modeling and simulation environment, develop software systems to select various parameters of the radar systems and operating environment, and analyze performance of the system in a testing environment. Hardware is Phase III: develop and build the hardware for technology transition.
If Phase II is simulation only, what does credibility rest on?
The fidelity of your simulation environment. With no over-the-air evidence, a reviewer will scrutinize your clutter model, interference model, channel model, platform geometry, and any validation of the simulator against measured data. Modest results from a simulator anchored to real measurements are more persuasive than impressive results from an unvalidated one.
What environmental conditions must the system sense?
The topic names three: interference, clutter, and signal-to-noise ratio. Address all three even though the feasibility metric measures interference.
What transmit parameters must the system control?
The topic names frequency, bandwidth, pulse repetition frequency, and other critical parameters. Say which your system controls and which it does not, and why.
What CMMC level applies?
The projected requirement for this topic is CMMC Level 2 with self-assessment. Note separately that the classified nature of the work imposes personnel and facility requirements well beyond CMMC.
Is this topic ITAR restricted?
No topic-level ITAR or EAR restriction paragraph appears on OSW26BZ06-DV033. The classified environment and the foreign national prohibition impose stricter constraints than an ITAR notice would.
Is the Company Commercialization Report evaluated?
No. Completion of the CCR as Volume 4 is required, but information contained in the CCR will not be considered by NGA during proposal evaluations. That differs from several other components in this cycle. Put your effort into the Commercialization Strategy inside the technical volume instead.
Are there Percentage of Work restrictions?
Yes. NGA will not accept any deviation to the Percentage of Work requirements described in the DoW Program. Note also that subcontractors inherit the security requirements and the DFARS 252.204-7012 and FAR 52.204-21 flow-downs.
What contract type should I expect, and when?
NGA typically provides a firm fixed price contract within 180 days of the proposal due date, roughly mid-April 2027 for this cycle, with contract type at the discretion of the Contracting Officer.
Can I publicize an award?
Only carefully. Contractor personnel shall not release any unclassified information, in any medium, pertaining to any part of the contract or any related program unless the COR has given prior written approval or in performance of a project scoped and negotiated by NGA. The NGA clause on unauthorized use of the NGA name, seal, and initials also applies. Plan your commercialization narrative with those constraints in mind.
When will I hear back?
Within 90 days of the closing date of the topic, approximately January 19, 2027, via email. Note that the NGA text says notification "for a Phase I award," which appears to be residual language given that this topic issues no Phase I award.
Who do I contact with questions?
Technical questions about the topic go through DSIP Topic Q&A, which closes October 7, 2026. Administrative questions about the NGA SBIR/STTR Program and these proposal preparation instructions go to sbir@nga.mil. Protests after award go to Patricia Hill at Patricia.D.Hill@nga.mil.
Positioning Advice for Companies Considering This Topic
Answer the security question before the radar question. Your ability to perform controlled work is a scored technical criterion, all personnel must hold Top Secret clearances with SCI eligibility, and classified work needs an accredited SCIF or an approved arrangement with NGA. If you cannot state a Facility Clearance Level, safeguarding level, CAGE codes, addresses, and an SPRS score, that gap is visible on page one of your Facilities section. Fix it, partner for it, or choose a different topic.
Check your SPRS score today. It is a specific number NGA asks for, a reviewer can look at it immediately, and it takes weeks to improve. An absent or low score on a scored controlled-work criterion is an unforced error.
Deal with the polygraph timeline now. Anyone not polygraphed in five years must schedule immediately and complete within 90 days in no more than three sessions. That constrains who you can actually put on a project team starting in spring 2027, and it is worth mapping before you name Key Personnel.
Use all twenty feasibility pages. NGA gives four times the feasibility allowance most components allow. Full experimental conditions, subsystem architecture, the interference scenarios, the decision logic, and complete before-and-after data all fit. A five-page feasibility section in a twenty-page allowance reads as thin evidence rather than efficient writing.
Ask about the SINR percentage, and report decibels regardless. A percentage SINR is not standard, and how NGA intends it changes what you must show. Ask in Topic Q&A, then report in decibels with conditions specified plus a normalized figure covering the stated range, and explain the choice. Silently reinterpreting a stated requirement is riskier than transparently addressing an ambiguity.
Make the simulation environment a first-class part of the proposal. Phase II is modeling, simulation, and software, so your simulator is your evidence base. Describe the clutter, interference, and channel models, the platform geometry, the validation against any measured data you have, and the separation between the environment where you develop and the testing environment where you evaluate. This is where a strong proposal separates from a plausible one.
Engage Haykin and Guerci explicitly. NGA cited the founding concept paper, the knowledge-aided fully adaptive book, and a software defined radar demonstration. Position your agentic AI layer against that lineage: what does an agentic decision layer add over knowledge-aided fully adaptive processing, and why now. Reinventing cognitive radar without naming that work signals you have not read the references.
Do not skip exploitation. The objective names parameters, waveforms, and exploitation. Most proposals will focus on transmit-side adaptation and treat processing as fixed. A system that adapts both what it sends and how it processes what comes back answers the objective as written, and Guerci's knowledge-aided approach is the reference that supports it.
Address the adversary's adaptation. The threat framing is asymmetric operating environments that adversaries impose. An adversary observing your adaptation policy can exploit it. Saying how your system avoids becoming predictable, and what happens when it is countered, is a differentiator on a resilience topic and most proposals will not raise it.
Define the autonomy boundary and the failure behavior. "Once initiated" implies a human starts it and the system runs. Say exactly what a human sets, what the system decides, what it does outside its training distribution, and how it signals degraded confidence. In an intelligence collection context, a system that knows it is failing is worth more than one that confidently continues.
Write the Section 508 outline. It is a stated proposal requirement, it sits inside the Commercialization Strategy, and it is precisely the item a proposal reused from another agency will omit. Name WCAG 2.1, describe the conformance approach for your operator interface, put accessibility review in the schedule, and use the exception path with rationale if something genuinely cannot comply.
Do not propose options. NGA prohibits it flatly, and an option structure carried over from another component's proposal is an immediate signal that you did not read the instructions.
Price for firm fixed price. NGA typically awards FFP within 180 days. Simulation fidelity work is the least predictable part of this scope, so carry margin there rather than discovering the gap after award.
Confirm your compute environment meets FedRAMP Moderate equivalence. A machine learning program on a commercial cloud that does not meet the standard for covered defense information is an expensive mid-program discovery.
Plan your commercialization story around the publication constraints. You cannot release unclassified information about the contract without written COR approval, and the NGA name and seal are protected. A commercialization strategy that depends on public visibility needs to account for that, and the honest version is more persuasive than one that ignores it.
Budget security administration as real work. HSPD-12 sponsorship coordination, insider threat program maintenance, HCMS data updates, badging and in-processing and out-processing, continuous evaluation enrollment, and departure property returns within four business days are all recurring administrative obligations. Companies new to IC work routinely underprice them.
OSW-FutureG SBIR OSW26BZ06-DV032: Smart Manufacturing, Open-Source Private 5G
Deadline: October 21, 2026
Funding Award Size: $2m
Description: Complete guide to OSW-FutureG SBIR Direct to Phase II topic OSW26BZ06-DV032, open-source private 5G for smart manufacturing. Up to $2,153,927 over 18 months. Closes October 21, 2026.
Quick Answer
OSW26BZ06-DV032 is a Direct to Phase II SBIR topic under the OSW FutureG Office, FY26 SBIR Broad Agency Announcement, Release 6. No Phase I award will be issued. The government wants a private 5G network for factories built entirely from open-source software, with no proprietary core, RAN Intelligent Controller, or Service Management and Orchestration component anywhere in the stack, demonstrated in a real metal-heavy industrial environment with at least three cells. The award must not exceed $2,153,927 over 18 months, with a 20-page technical volume. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The stack is named and mandatory: OCUDU for the Centralized Unit and Distributed Unit, SD-Core for the 5G core, SD-RAN for the Near-Real-Time RIC, and the OSC stack for the Non-Real-Time RIC and SMO. A commercial core or a vendor RIC does not meet the topic.
The performance targets are specific and, unusually, so is the price point. The dominant performance classes are ultra-reliable low-latency communication and high-mobility reliability: sustained sub-30 millisecond threshold to sub-15 millisecond objective latency for machine-vision backhaul, at least 99.5 percent handover success at operational automated guided vehicle and autonomous mobile robot speeds, and at least five-nines availability for safety-critical services, all in dense, metal-heavy RF environments. And the target commercial packaging is a CBRS-based as-a-service offer at $100,000 to $250,000, which is the number that decides whether the business case closes.
The topic is also refreshingly honest about what it is not claiming. On Wi-Fi roaming it says modern standards substantially mitigate fixed-access-point handoff behavior when properly deployed, and that the real advantages sought are deterministic network-scheduled access on interference-managed spectrum, standardized network-controlled mobility management, and quality-of-service guarantees enforceable under load. A proposal that argues Wi-Fi cannot roam is arguing against the topic's own text.
One thing to know before drafting. The topic repeatedly directs proposers to key performance metric tables and to "Section 3.0." No such tables and no numbered Section 3.0 appear in the published document. That gap has its own section below and it is the most important question to ask before the topic closes.
Topic At a Glance
Topic number: OSW26BZ06-DV032
Title: Smart Manufacturing
Agency: Office of the Secretary of War, FutureG Office, under OUSW(R&E)
Solicitation: OSW FutureG Office, FY26 SBIR Broad Agency Announcement, Release 6, Proposal Submission Instructions
Program type: Direct to Phase II only. This topic is accepting Direct to Phase II proposals only, and a formal Phase I award will not be issued
Award: must not exceed $2,153,927
Period of performance: 18 months
Technical volume limit: 20 pages maximum, structured as Part 1 Phase I Justification at 5 pages maximum and Part 2 Phase II Technical Proposal at 15 pages maximum, with the Technology Transition and Commercialization Strategy at no more than 2 pages counting toward the 15
OUSW (R&E) Critical Technology Areas: Applied Artificial Intelligence (AAI), Contested Logistics Technologies (LOG)
Component Technology Priority Areas: FutureG, Advanced Infrastructure and Advanced Manufacturing, Sustainment and Logistics
Projected CMMC level requirement: Level 2. Note that this topic states Level 2 without the parenthetical self-assessment qualifier that appears on the other two topics in this release
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic
Mandatory stack: OCUDU plus SD-Core plus SD-RAN plus OSC, with no proprietary core, RIC, or SMO component anywhere
Dominant performance classes: ultra-reliable low-latency communication and high-mobility reliability
Stated performance targets: sub-30 ms threshold and sub-15 ms objective latency for machine-vision backhaul, at least 99.5 percent handover success at operational AGV and AMR speeds, and at least five-nines availability for safety-critical services
Stated commercial price point: a CBRS-based as-a-service offer at $100,000 to $250,000
Market: roughly 50,000 U.S. manufacturing firms in the 20 to 99 employee band alone, with several thousand more in the 100 to 250 range
Use cases: proposers must address at least two of four, and must identify which their reference architecture and pilot deployment are designed to validate
Mandatory common work package: OCUDU baseline benchmarking and upstream enhancement, Tasks A, B, and C, required regardless of use cases selected
Additional mandatory task: representative facility-specific RF network planning and site engineering
Upstream requirement: all modifications to OCUDU shall be contributed upstream through the project's standard contribution and review process
Demonstration site: must be representative of the target deployment class in RF character, meaning metal-heavy and multipath-rich industrial construction, and in scale, meaning a footprint and cell count of at minimum three cells sufficient to exercise inter-cell handover at operational AGV and AMR speeds
Out of scope for Phase II: hard-real-time machine motion control, meaning isochronous traffic with cycle times of approximately 0.5 to 2 milliseconds per 3GPP TS 22.104
Technical and Business Assistance: up to $50,000 per Phase II project, in addition to the cost ceiling and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5
Percentage of Work: the FutureG Office will not accept any deviation to the POW requirements
Company Commercialization Report: information contained in the CCR will be considered during proposal evaluations
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026
Submission portal: DSIP at dodsbirsttr.mil
Keywords: smart manufacturing, AI, robotics, smart factories, Industry 4.0, 5G connected warehouse
The Feasibility Bar, Which Is the First Thing to Check
This topic is accepting Direct to Phase II proposals only, so a formal Phase I award will not be issued.
To qualify for a Phase II award, proposers must submit Feasibility Documentation as part of their proposal package demonstrating that they have already completed Phase I-type research and development. The purpose of this documentation is to prove that the underlying technology is mature, scientifically sound, and ready to transition immediately into the Phase II prototyping and testing environment.
What the government will accept
The Government will accept a wide variety of documentation styles. Proposers do not need to have a perfect, finished product, but they must show that their core ideas have been successfully tested. One or more of the following items should be included to prove technology readiness.
Test data and metrics: real-world measurements, performance charts, or network test data from previous lab environments or early field trials.
Technical reports and white papers: written summaries explaining your previous research, system designs, or software integration efforts.
Prototype designs and simulation models: diagrams, architectural blueprints, or computer simulation results showing how your proposed software and hardware components interact.
Previous project outcomes: success criteria, milestone reports, or commercialization results from prior private, academic, or non-SBIR federally funded work.
This is among the most accommodating feasibility framings in the 2026 cycle. One or more of the four suffices, simulation models are explicitly acceptable, and the government states you do not need a finished product.
The restriction that still applies
The FutureG Direct to Phase II guidelines impose a hard constraint the topic-level text does not repeat.
Feasibility documentation cannot be based upon or logically extend from any prior or ongoing federally funded SBIR or STTR work. Work submitted within the feasibility documentation must have been substantially performed by the proposer or the principal investigator. If technology in the feasibility documentation is subject to intellectual property, the proposer must either own the IP or must have obtained license rights to such technology prior to proposal submission, to enable it and its subcontractors to legally carry out the proposed work.
The Volume 2 instruction phrases it as "must not be solely based on" prior or ongoing federally funded SBIR or STTR work, which is weaker. Plan against the stricter formulation.
Notice that the topic's own fourth evidence category names "prior private, academic, or non-SBIR federally funded work," which is consistent with the restriction and tells you where to look. Private, academic, and non-SBIR federal work are all fine. Prior SBIR and STTR work is the problem, and open-source 5G integration work in the United States has been substantially SBIR funded, so audit the provenance of every result you intend to cite.
If the proposer fails to demonstrate technical merit and feasibility equivalent to the Phase I level as described in the topic, the related Phase II proposal will not be evaluated.
The Missing KPM Tables, and What the Topic Does Tell You
The topic refers to KPM tables repeatedly. It says other suggested KPMs for the various use cases and system requirements are given below for reference. It says each use case is defined in terms of the operational scenario, the connectivity requirement, and "the corresponding KPM table." Task A requires a baseline against the Phase II KPMs "including the General System and Architectural Requirements." The deliverables require Key Performance Metrics "see Section 3," MVP documentation reporting "the KPI results achieved against Section 3.0 thresholds," and a baseline benchmark report with "measured baseline results against Section 3.0 KPMs." A scope note refers to "the latency-class scope note in Section 3.0."
No KPM tables appear in the published document, and there is no numbered Section 3.0 or General System and Architectural Requirements section.
The governing principle is stated, even though the tables are not
This topic, unlike its companion DV031, spells out how the KPMs are meant to work, and that paragraph is what lets you proceed.
The KPM values in the tables are suggested reference values, not pass or fail contract requirements. Proposers shall propose specific, justified KPM targets in the Technical Volume, calibrated to their selected use cases, demonstration environment, spectrum plan, and channel bandwidth. Deviations from the suggested values must be technically justified. Threshold values represent the intended standard of Phase II demonstration success, and objective values are stretch goals, some of which are expected to mature during Phase III or along the FutureG evolution path. Final KPM targets will be agreed with the Government at kickoff and confirmed at the Critical Design Review. Consistent with the key milestones, partial achievement within a defined and documented scope may be considered successful.
So the practical approach is to propose your own KPM set with threshold and objective values, calibrated to your use cases, demonstration environment, spectrum plan, and channel bandwidth, justify each, and note that final targets are agreed at kickoff and confirmed at CDR. State plainly that you are doing so because the referenced tables do not appear in the published instructions.
Ask anyway. DSIP Topic Q&A closes October 7, 2026, and requesting the tables or a pointer to them is the single highest-value question on this topic.
The numbers the topic does state in its text
Four performance figures appear in the body and you should treat them as the anchors.
Sustained sub-30 millisecond threshold to sub-15 millisecond objective latency for machine-vision backhaul.
At least 99.5 percent handover success at operational AGV and AMR speeds.
At least five-nines availability for safety-critical services.
And a worked example: "The platform must sustain a session success rate greater than 99.5% for AGV/AMR connections while handing over between at least 3 small cells at speeds up to 10 mph."
The topic also names the KPM categories to address: session success rate, handover success rate, latency, jitter, and service availability.
The latency measurement definition, which is unusually precise
Latency KPMs in this document denote round-trip application-layer latency measured between the user equipment application interface and the local edge application endpoint served behind the on-site or edge User Plane Function, meaning device to edge through the RAN and core user plane, excluding external wide-area network transport.
Handover time denotes user-plane interruption time at the RAN, per 3GPP definitions.
For each latency KPM adopted, proposers shall specify the measurement reference points, the traffic type and packet size, and the measurement methodology, for example key performance indicator definitions per 3GPP TS 28.554, and shall include in the Critical Design Document a latency budget decomposition across the air interface, DU and CU processing, fronthaul and F1 transport, core user plane, and application processing.
That latency budget decomposition is a specific, named CDR deliverable across five segments. It is also a genuinely useful engineering discipline and one of the more concrete requirements in the topic. Do not treat it as boilerplate.
The latency scope note, which limits what you must demonstrate
Hard-real-time machine motion control, meaning isochronous traffic with cycle times of approximately 0.5 to 2 milliseconds per 3GPP TS 22.104, is outside the Phase II demonstration scope.
The latency KPMs in this topic target AGV and AMR supervision and control-plane traffic, machine-vision backhaul, and safety alerting.
Proposers shall, however, address in their FutureG evolution path the OCUDU enhancements required to approach the 1 millisecond latency class over time, including deterministic and priority scheduling, Time-Sensitive Communication support, and mini-slot and preemption features.
This is the government being realistic, and it matters. You are not being asked to close a motion-control loop over 5G in 18 months. You are being asked to hit sub-30 to sub-15 milliseconds for supervision, vision backhaul, and alerting, and to document the roadmap toward the 1 millisecond class. A proposal that promises isochronous motion control in Phase II is not more ambitious, it is out of scope.
What the Government Is Actually Buying
The objective
The objective of this Phase II effort is to design, validate, and demonstrate a fully open-source, private 5G network platform for smart manufacturing that integrates OCUDU, SD-Core, SD-RAN, and OSC components into a secure, vendor-neutral architecture.
This platform will deliver highly reliable, low-latency connectivity to support autonomous robotics, machine vision, and connected-worker safety, providing a cost-effective, FutureG-ready solution for both commercial manufacturers and dual-use defense logistics facilities.
The market, as the government describes it
The realistic near-to-mid-term serviceable addressable market for private 5G among U.S. small and medium-sized manufacturers is not all manufacturers. It is the subset with real mobility, reliability, coverage, or security pain: plants using automated guided vehicles, autonomous mobile robots, machine vision, connected workers, outdoor yards, large indoor spaces, or metal-heavy RF environments.
Manufacturing is already among the leading sectors for private mobile network deployments worldwide, and published case studies show double-digit productivity gains and lower infrastructure capital expenditure versus Wi-Fi deployments.
Based on data compiled by the National Association of Manufacturers, roughly 50,000 U.S. manufacturing firms fall in the 20 to 99 employee band alone, with several thousand more in the 100 to 250 range.
Note the qualification. The market is not 50,000 plants, it is the subset of them with one of six named pain characteristics. Your commercialization strategy should segment on those characteristics rather than quoting the headline firm count, because the topic already told you the count is not the addressable market.
How this segment buys, and what stops it
This segment buys on simplicity, speed, fit, and economics, not on global platform standardization.
Its stated barriers to adoption are consistent and compounding: return-on-investment ambiguity, legacy equipment and retrofit costs, spectrum and regulatory fragmentation, uneven device ecosystems, information technology and operational technology integration complexity, in-house skills shortages, and tight capital budgets.
A proprietary platform from a tier-one vendor addresses almost none of these barriers directly. It is priced and supported for large campuses and multi-site accounts, not a single plant with one connectivity problem to solve.
Seven named barriers. That list is effectively a scoring rubric for your commercialization strategy, and addressing each one explicitly is cheap differentiation.
The open-source argument and its honest cost
An all-open-source stack removes the structural cost and lock-in that makes proprietary platforms a poor fit for this segment. Because OCUDU's O-CU and O-DU communicate with the rest of the network over open, standardized interfaces, especially the O-RAN 7.2x split to the O-RU, the network can be assembled from whichever radio unit vendor best fits a given plant's bands, form factor, and budget, rather than a single bundled radio line.
Together, OCUDU, SD-Core, SD-RAN, and OSC form one integrated, fully open-source 5G stack that a smaller solution provider or systems integrator can deploy, support, and price as a right-sized, as-a-service offer for a single plant, instead of selling a broad, vendor-locked platform.
That directly answers this market's stated preference for outcome-first, economically flexible offers, and it supports a Wi-Fi-coexistence posture, meaning private 5G for the hard problem zones and Wi-Fi where it already works, rather than a rip-and-replace sale.
The trade-off is that the burden of proving interoperability, stability, and security shifts from a single vendor's warranty to the integrator and the open-source community itself. Small and medium manufacturer buyers, who by their own account lack in-house cellular and associated cybersecurity expertise, have little tolerance for integration failures discovered after deployment.
This is exactly the gap the ongoing OCUDU Testing and Validation Plan, and its RTEC-centered extensions, are designed to close: independently witnessed, multi-vendor testing of device diversity, RU diversity, outdoor RF performance, Core, RIC, and SMO integration, and security. The proposal should leverage the current, ongoing testing and evaluation activities in RTECs and associated results.
Take the Wi-Fi-coexistence framing seriously. The topic explicitly rejects a rip-and-replace posture, and a proposal that positions private 5G as replacing plant Wi-Fi is arguing against the solicitation's own commercial logic. Private 5G for the hard zones, Wi-Fi where it works, is the stated sale.
The generalization requirement
Although this topic is anchored in a specific vertical to ensure a concrete deployment environment, real users, and a defensible commercialization path, proposers should recognize and are required to present the technical work in terms of the generic network performance classes it advances.
Improvements made to the OCUDU O-CU and O-DU under this effort, meaning scheduler behavior, mobility management, uplink capacity, stability under sustained load, and security features, are expected to generalize across verticals and to benefit the broader community of RAN developers building on OCUDU.
The Technical Volume shall include a mapping of each selected use case, and its associated KPMs, to the performance class or classes it exercises, and shall identify which anticipated OCUDU code or feature enhancements correspond to each class.
This is a required Technical Volume element stated with "shall." It is easy to omit while writing about factories. Reserve space for it.
The Three Research Questions
Research and development for this effort should address, at minimum, the following three questions. Note that this topic asks three where its venue companion asks four; the ISAC evolution question is absent here.
All-open-source economics
What is the fully loaded cost, covering integration, support, spectrum, and hardware, of an OCUDU plus SD-Core plus SD-RAN plus OSC deployment relative to a proprietary platform at small and medium manufacturer scale, and how should that be packaged as a CBRS-based, as-a-service offer within the price point buyers require, stated as $100,000 to $250,000?
The stated price band is the most actionable number in the topic. It bounds the whole design: how many radio units, what class of hardware, how much integration labor, and what recurring service margin. Build a cost model against it and show that it closes, because a technically excellent platform that lands at $600,000 per plant does not answer the question that was asked.
Use-case-first automation
Which of the use cases defined below deliver the fastest, most measurable return on investment on the open-source stack, and what RTEC-executed interoperability tests across Device, RU, Core, RIC and SMO, and Infrastructure dimensions are needed to validate each one end-to-end?
Trust in open source
How does RTEC-executed testing of the full open-source stack, not just OCUDU, reduce the integration risk and skills-shortage barrier that small and medium manufacturer buyers most often cite, and what evidence package best converts community-maintained, no-license-fee software into a procurement-ready proof point for a buyer with no in-house cellular expertise?
That third question is the commercial crux and the one most proposals will answer weakly. The implied deliverable is an evidence package: what does a plant manager with no RF staff need to see, in what form, to sign a contract for a network with no vendor warranty behind it? Treat it as a document design problem, not a testing plan.
Solutions leveraging artificial intelligence and machine learning for predictive maintenance, network optimization, or automated fault resolution are encouraged but not required. The government will consider any novel concept that increases the reliability, economics, and trustworthiness of an all-open-source private 5G and FutureG platform for the manufacturing vertical. Dual-use opportunities are expected across both commercial small and medium manufacturers and DoW facility networks.
Phase II Scope
This Phase II effort will design, validate, and demonstrate a fully open-source private 5G network platform purpose-built for manufacturing facilities, integrating OCUDU, SD-Core, SD-RAN, and the OSC stack with no proprietary core, RIC, or SMO components.
The platform addresses four manufacturing-specific needs: high-mobility connectivity for automated guided vehicles and mobile plant equipment; low-latency video and camera backhaul for quality inspection and process monitoring; worker and plant safety communications; and outdoor yard and logistics coverage extending beyond the factory floor.
Phase II work will produce a validated reference architecture, RTEC-executed interoperability testing across RU, Core, RIC and SMO, and infrastructure dimensions, and a working minimum viable product demonstrated at a representative manufacturing site, along with a documented technical bridge path toward FutureG capability.
Anticipated benefits include a lower-cost, vendor-neutral alternative to proprietary industrial wireless systems, improved reliability for mobile robotics and safety-critical communications in dense industrial RF environments, and a reusable open-source deployment model.
The Four Use Cases, Of Which You Must Address Two
Proposers must address at least two of the following four use cases in their Phase II Statement of Work, and must identify which use cases their reference architecture and pilot deployment are designed to validate.
Device-class diversity and RedCap, which applies across all use cases
Each use-case set spans device classes from full-capability user equipment, meaning equipment modems, cameras, and broadcast and production units, to reduced-capability Internet of Things endpoints, meaning sensors, wearables, tags, and trackers.
Proposers shall address how the platform serves reduced-capability device classes, including 3GPP Release 17 Reduced Capability, RedCap, and Release 18 eRedCap user equipment, and shall identify any OCUDU scheduler or feature enhancements required to support RedCap operation. Such enhancements are strongly encouraged as upstream contributions under the common OCUDU benchmarking and enhancement work package.
Where RedCap-certified devices are not commercially available for a given endpoint type at demonstration time, proposers may demonstrate with available device classes or emulated RedCap user equipment profiles, and shall document the RedCap migration path.
That final allowance is worth using, since RedCap device availability in CBRS bands remains limited.
Industrial IoT service characteristics, which are specific to this topic
For the manufacturing vertical specifically, proposers shall additionally address Industrial IoT service characteristics as framed by 3GPP TS 22.104 on service requirements for cyber-physical control applications, including four things.
The mapping of selected use cases to TS 22.104 communication service classes.
Support for private-network, meaning Non-Public Network, operation.
Awareness of Time-Sensitive Communication and IEEE Time-Sensitive Networking integration concepts in the platform architecture.
And secure information technology and operational technology segmentation for IIoT traffic, for example via network slicing or 5G-LAN group management.
This requirement has no counterpart in the venue topic and it is a genuine technical scope addition. The TS 22.104 service class mapping in particular is a concrete artifact a reviewer can check, and IT and OT segmentation is the requirement that plant IT departments will care about most.
Use Case 1: AGV and AMR connectivity and mobility
Automated guided vehicles and autonomous mobile robots move continuously across the plant floor, between production cells, and into staging or storage areas, requiring an uninterrupted control-plane and telemetry connection as they roam. Roaming interruptions have historically been a common source of dropped sessions and stalled vehicles in Wi-Fi-served plants.
The topic then does something unusual and important. It concedes the counterargument. Modern Wi-Fi roaming standards, meaning IEEE 802.11k neighbor reports, 802.11r fast BSS transition, and 802.11v BSS transition management, substantially mitigate fixed-access-point handoff behavior when properly deployed on capable client devices. The more fundamental challenges in this environment are contention-based access on unlicensed, shared spectrum and the severe attenuation, multipath, and reflection conditions of metal-heavy plants, which degrade any RF system absent careful network planning.
The advantage sought under this topic is therefore not that Wi-Fi cannot roam, but that a 3GPP system provides deterministic, network-scheduled access on interference-managed spectrum, standardized mobility management under network control, and quality-of-service guarantees that remain enforceable under load.
This use case requires the platform to sustain low-latency, high-reliability connectivity and seamless handover as AGVs and AMRs move between small cells, indoors and where applicable into adjoining outdoor areas.
It additionally requires on-floor localization of AGVs, AMRs, and tagged mobile assets for fleet management, geofencing, and safety zoning. Network-native positioning using 3GPP NR positioning methods from Release 16 and 17 is preferred. Hybrid approaches that fuse NR positioning with existing plant localization systems may be proposed with justification.
Two things to take from this. First, do not write a Wi-Fi-cannot-roam argument; the topic pre-refuted it and a reviewer will notice. Make the case on deterministic scheduling, interference-managed spectrum, network-controlled mobility, and enforceable quality of service. Second, localization is a requirement inside this use case, not an optional extra, and network-native NR positioning is preferred over fusion with existing plant systems.
Use Case 2: Machine vision backhaul
Quality-inspection and process-monitoring cameras generate continuous, high-bandwidth video or image streams that must reach an on-premises or edge analytics system with minimal delay and jitter to support real-time defect detection and line-stoppage decisions.
This use case requires the platform to sustain high uplink throughput with low jitter for multiple simultaneous camera streams, including in metal-heavy or RF-reflective areas of the plant where Wi-Fi performance typically degrades.
Note that this is the use case the sub-30 and sub-15 millisecond latency targets attach to, and it is an uplink-heavy problem, which is the harder direction for a 5G system. Uplink capacity is also one of the named OCUDU enhancement areas in Task B, so this use case connects directly to the mandatory work package.
Use Case 3: Connected-worker safety
Plant personnel increasingly carry or wear connected devices such as gas and hazard sensors, push-to-talk radios, panic buttons, or health and location monitors, that must reliably reach a monitoring system without gaps in coverage, including in areas such as mezzanines, tank farms, and loading docks that legacy Wi-Fi does not reliably reach.
This use case requires the platform to sustain consistent, low-latency coverage for safety-critical alerting across the full indoor plant footprint.
This is where the five-nines availability target lives, and it is the use case most likely to involve RedCap and eRedCap devices, since wearables and sensors are exactly the reduced-capability endpoint class.
Use Case 4: Secure outdoor yard and logistics coverage
Loading docks, staging yards, rail sidings, and outdoor storage areas typically fall outside indoor Wi-Fi coverage entirely, yet increasingly require connectivity for yard trucks, RFID and asset tracking, outdoor cameras, and inventory handling equipment.
This use case requires the platform to extend secure, private coverage into outdoor areas immediately adjacent to the plant, using the same OCUDU-based infrastructure rather than a separate outdoor Wi-Fi buildout.
Note that outdoor RF performance is named as one of the dimensions the RTEC testing program covers, which makes this use case comparatively well supported by existing validation work.
Choosing your two
Use Cases 1 and 2 are the most tightly coupled to the topic's stated performance classes, since Use Case 1 carries the handover success target and Use Case 2 carries the latency targets, and together they exercise both dominant performance classes of URLLC and high-mobility reliability. They also both live indoors on the same RF design, which is the cheapest pairing.
Use Case 3 is the natural pair with either, since it shares the indoor footprint and adds the RedCap and availability dimensions at modest additional hardware cost. Use Case 4 requires outdoor cell planning and additional radio units.
The Contested Logistics Technologies Critical Technology Area designation points toward Use Case 4 and the depot and logistics defense narrative, so if the defense transition story matters to you, weigh that.
The Mandatory Common Work Package
All performers under this topic shall execute the following common work package, which is a required element of the Phase II Statement of Work regardless of the use cases selected.
Not optional, and not scoped by your use case choice. Budget and staff it separately.
Task A: Baseline Benchmark
Establish a quantified performance baseline of the integrated open-source stack, meaning OCUDU plus SD-Core plus SD-RAN plus OSC, against the Phase II KPMs relevant to the selected use cases, including the General System and Architectural Requirements.
An emulated end-to-end configuration, using emulated radio units and user equipment, RF channel emulation, or synthetic load generation, is acceptable and encouraged for the baseline, provided the benchmark methodology, tooling, configurations, and results are fully documented and reproducible.
Where an RTEC-validated reference configuration already exists for the proposed RU, Core, RIC, and SMO combination, the baseline shall incorporate available RTEC results rather than duplicate them.
Baseline methodology and results shall be presented to the OCUDU Test and Evaluation Working Group.
The baseline benchmark report is expected by Month 3, which means your emulation environment must stand up in the first weeks of the award.
Task B: Code and Feature Enhancement
Identify the gaps between baseline performance and threshold and objective values, and develop the OCUDU code improvements and features required to close them, for example scheduler and quality-of-service enhancements, mobility and handover optimization, uplink capacity improvements, stability hardening, and security features.
All modifications to OCUDU shall be contributed upstream through the project's standard contribution and review process. Enhancements that cannot be upstreamed shall be documented with rationale.
Progress against each targeted KPM, and the status of each upstream contribution, shall be reported in the Monthly Status Reports presented to the OCUDU Test and Evaluation Working Group.
Two consequences worth confronting in your proposal. Your code improvements go into a public project on that project's review timeline, which is schedule risk you do not own, and the deliverable log explicitly includes review status because acceptance is not guaranteed. And your commercial differentiation cannot be the OCUDU code itself; it has to be the integration, the RF engineering, the evidence package, the service model, and the support relationship. Say so in the commercialization strategy.
Note that mobility and handover optimization and uplink capacity improvements are both named enhancement examples, and both map directly onto Use Cases 1 and 2. That alignment is worth making explicit in your performance-class mapping.
Task C: Benchmark and Regression Harness
Deliver the emulation-based end-to-end benchmark suite developed under Task A as a repeatable, documented, open-source harness suitable for adoption by the OCUDU community and RTECs for regression testing of future OCUDU releases.
This work package complements, and does not replace, the interoperability testing and the physical MVP demonstration required elsewhere in this topic. Emulated results establish the baseline and guide enhancement work. Over-the-air performance with physical radio units and user equipment at the RTEC or representative demonstration site remains the standard of evidence for final KPM achievement.
RF Network Planning and Site Engineering, a Second Mandatory Task
This requirement is unique to the manufacturing topic and it is one of the more substantive engineering asks in the release.
Industrial facilities are among the most difficult RF environments for any wireless system. Dense metal structures, racking, and machinery produce severe attenuation, multipath, and reflection conditions that directly affect handover performance, throughput, jitter, and coverage completeness.
Performers shall execute a representative facility-specific RF engineering task comprising three elements.
Predictive propagation modeling incorporating facility-specific features such as metal racking, machinery, mezzanines, tank farms, and loading areas.
An RF design that mitigates identified dead spots and multipath-driven impairments through cell placement, antenna selection and orientation, and mobility-parameter tuning.
And installation engineering practices that protect radio hardware, including antenna placement clearances from nearby reflective metal and verification of antenna-system return loss and voltage standing wave ratio at commissioning, with monitoring thereafter, to prevent reflected-power damage to radio unit front ends.
That third element is notably practical and it is the kind of detail that signals the requirement was written by someone who has damaged a radio front end. Return loss and VSWR verification at commissioning, with ongoing monitoring, is an installation and operations procedure, and an RF Design Report including representative coverage maps and return-loss and VSWR commissioning approaches is a named deliverable.
Security Requirements
Security shall be a first-class design requirement of the platform, not a demonstration afterthought.
Performers shall implement and document a security architecture covering the following.
3GPP security per TS 33.501, including mutual authentication and air-interface encryption and integrity protection.
Protection of the O-RAN open interfaces, meaning open fronthaul, E2, A1, and O1, per O-RAN WG11 specifications.
Zero-trust principles per NIST SP 800-207, including least-privilege access and separation of management and user traffic.
Monitoring of Common Vulnerabilities and Exposures affecting OCUDU and its dependencies, and timely upstream patching.
Security features and hardening developed for OCUDU shall be contributed upstream under the common benchmarking and enhancement work package.
The security architecture shall be documented at the Critical Design Review and validated in RTEC testing, and the MVP demonstration shall include at least one security capability shown live, for example rejection of an unauthorized device, encrypted fronthaul, or detection of a simulated intrusion.
Pick your live security demonstration early and design for it. Note also that the IIoT requirement above asks for secure IT and OT segmentation via network slicing or 5G-LAN group management, which is a second security-adjacent requirement specific to this topic and worth addressing alongside the architecture.
Milestones, Demonstration Site, and Deliverables
Milestones as stated
Month 1: Kickoff and Technical Interchange Meeting.
Monthly Status Reports throughout.
Month 12: Critical Design Review.
Month 16: Prototype demonstration.
Month 14: Final design review, demonstration, and assessment.
Month 18: Final Phase II Report.
The published list places the Month 16 prototype demonstration before the Month 14 final design review, which cannot be the intended sequence. The same inversion appears in the companion topic DV031, which points to a shared drafting error. Confirm through DSIP Topic Q&A and state your assumed sequence in your work plan.
Monthly Status Reports must be presented to the OCUDU Test and Evaluation Working Group as well, bringing the community up to speed on progress. That is a recurring external commitment and it should be staffed.
The demonstration site, which is specified more tightly here than in the venue topic
Prototype demonstrations will be performed at the proposer's site, ideally an operating or representative manufacturing facility such as a partner small or medium manufacturer plant, a manufacturing institute or applied-research factory floor, or a comparable industrial or laboratory environment.
The demonstration environment must be representative of the target deployment class in RF character, meaning metal-heavy, multipath-rich industrial construction, and in scale, meaning a footprint and cell count at minimum three cells, sufficient to exercise inter-cell handover at operational AGV and AMR speeds. Its fidelity to plant conditions must be documented in the MVP demonstration package.
Partial solutions may be considered successful if effective within a defined scope. A final technical report detailing the capabilities demonstrated will be required. Extended user evaluations or additional prototypes may be pursued based on utility.
The three-cell minimum is a hard, checkable number and it should drive your site selection and your hardware budget. The topic's own worked KPM example describes handing over between at least three small cells at speeds up to 10 miles per hour, so three cells and roughly 10 miles per hour is the demonstration you should plan.
Note that the site sentence in the published text reads "ideally an operating or representative manufacturing facility (such as a partner SMM plant, a manufacturing institute or applied-research factory floor, or a comparable industrial or laboratory environment) where live plant access is not feasible during Phase II," which parses oddly. The companion venue topic contains the parallel construction "or a full-scale representative test bed where live-venue access is not feasible during Phase II," which suggests the alternative clause was dropped here. The sensible reading is that an operating plant is preferred and a comparable industrial or laboratory environment is the fallback where live plant access is not feasible. Either way, the RF character and three-cell scale requirements govern.
OCUDU integration and scalability
The MVP demonstration, including physical radio units and user equipment, will need to occur at the factory-representative site, as close to a real environment as possible.
Performers should address scalability, including testing across multiple radio unit vendors and hardware-accelerator options, and should leverage RTEC-executed interoperability testing to address integration risks ahead of the demonstration wherever a validated reference configuration already exists.
Phase II deliverables
Kickoff and Technical Interchange Meeting slides.
Monthly Status Reports.
A Critical Design Document containing the full reference architecture across OCUDU, SD-Core, SD-RAN, and OSC.
Key Performance Metrics.
MVP Demonstration slides and documentation, including a description of the demonstration site's fidelity to factory conditions and the KPI results achieved against the referenced thresholds.
A Reference Configuration Package comprising executables, integration documentation, and RTEC test results for the validated RU, Core, RIC, and SMO combinations used.
Integration of the platform into an RTEC-affiliated test and evaluation network or a factory-representative demonstration site, demonstrating at least one of the two selected use cases.
An OCUDU Baseline Benchmark Report covering methodology, emulation environment description, configurations, and measured baseline results, expected by Month 3.
An Upstream Contribution Log, itemizing OCUDU code contributions such as patches and pull requests, their review status, and the KPM gap each addresses, updated in each Monthly Status Report with the final version in the Final Technical Report.
An open-source benchmark and regression harness, with documentation sufficient for independent execution by RTECs and the OCUDU community.
An RF Design Report, including representative coverage maps and return-loss and VSWR commissioning approaches.
A Final Design Document.
A Final Technical Report.
Note the relationship between addressing at least two use cases and "demonstrating at least one of the two selected use cases." You scope two in the Statement of Work and physically demonstrate at least one. That is a meaningful reduction in demonstration burden and it should shape your pairing: pick two where one is demonstrable at your site and the other is architecturally addressed.
Note also that the Critical Design Document here does not carry the venue topic's requirement to include a 6G and ISAC evolution path, consistent with this topic having no ISAC use case. It does, however, require the latency budget decomposition described earlier, and the Phase II scope calls for a documented technical bridge path toward FutureG capability.
Phase III Dual Use
The development of an open-source private 5G network platform for smart manufacturing offers significant dual-use potential, benefiting both commercial industry and Department operations.
For the commercial sector, this technology provides small and medium-sized manufacturers with a low-cost, secure, and vendor-neutral wireless solution. It directly addresses key manufacturing needs such as enhancing automated guided vehicle mobility, enabling real-time machine vision for quality control, improving connected-worker safety, and extending secure connectivity to outdoor logistics yards.
For the Department, this same technology can be applied to its own industrial and logistical environments. It offers a pathway to modernize DoW-affiliated depots, maintenance facilities, and logistics operations with resilient, high-mobility wireless connectivity. This is particularly relevant for contested logistics, where reliable, secure, and private communication networks are critical for maintaining operational tempo and supply chain integrity. The platform's open-source nature reduces dependency on proprietary systems and enhances security, aligning with key modernization goals.
The depot and maintenance facility case is the strongest defense hook and it is structurally identical to the commercial one. Depots are large, metal-heavy, multipath-rich industrial environments with mobile equipment, asset tracking needs, and outdoor yards, run by organizations that also lack in-house cellular engineering staff. If you can name a specific depot, maintenance center, or logistics activity, that is worth more than the general claim.
Funding, Cost Structure, and FutureG Mechanics
The award
Direct to Phase II proposals must not exceed a cost of $2,153,927 and a duration of 18 months.
Be realistic about scope. A full open-source stack integration, RTEC interoperability testing, OCUDU code enhancement with upstream contribution, a facility-specific RF engineering task with predictive propagation modeling, a security architecture with a live demonstration, an emulation benchmark harness, and a physical three-cell MVP at a factory-representative site, in 18 months for $2.15 million, is a full program. Existing OCUDU experience, an existing plant or applied-research factory floor relationship, and existing RTEC engagement are worth more than headcount.
Cost volume
A detailed Phase II Cost Volume must be submitted online in the proper format shown in the Cost Breakdown Guidance in the DoW 2026 SBIR BAA. Some items may not apply, and there is no need to provide information for every item. Provide enough information to allow evaluators to assess your plans to use the requested funds.
Justify items of equipment to be purchased, including Government Furnished Equipment. All requirements for government furnished equipment or other assets, and associated costs, must be determined and agreed to during Phase II contract negotiations. At least three radio units, user equipment across full-capability and reduced-capability classes, hardware accelerators, CBRS Spectrum Access System service, channel emulation and load generation for the Task A baseline, propagation modeling tools, and VSWR and return-loss test equipment all belong in the cost discussion.
Percentage of Work, with no exceptions
Review the updated Percentage of Work calculation details included in the DoW SBIR Program BAA. The FutureG Office will not accept any deviation to the POW requirements.
The natural team here includes a radio unit vendor, a systems integrator, a plant partner, possibly an RTEC, and possibly a manufacturing institute or university. Model your POW before you assemble it.
Technical and Business Assistance
Phase II awardees may request up to $50,000 per Phase II project. TABA funding is in addition to the Phase II cost ceiling and is not subject to profit or fee.
All requests for TABA must be completed using the SBIR/STTR TABA Request Form, and the completed form must be included in Volume 5 of the proposal submission in DSIP. OSW will not accept requests for TABA that do not utilize the form or that are not included as a submission document in Volume 5.
For this topic the strongest uses are commercial go-to-market development, since the topic asks explicitly for an as-a-service offer inside a stated price band, and spectrum and regulatory support for the CBRS deployment.
The 20-page structure
Volume 2 is 20 pages maximum: Part 1, Phase I Justification, 5 pages maximum, and Part 2, Phase II Technical Proposal, 15 pages maximum, with the Technology Transition and Commercialization Strategy at no more than 2 pages counting toward the 15.
So 5 pages of feasibility, 13 pages of technical proposal, 2 pages of commercialization. Against that you must fit: three research questions, two use cases, RedCap handling, the IIoT and TS 22.104 requirements, the required performance-class mapping, three common work package tasks, the RF network planning and site engineering task, the security architecture, the latency measurement definitions and budget approach, the milestone plan, key personnel, facilities, and consultants. This topic has more mandatory content than its venue companion and the same page allowance. Plan the allocation before drafting.
The FutureG instructions do not state that figures, tables, charts, and references count inside the page limit, and do not prohibit appendices. They defer to the DoW SBIR Program BAA formatting requirements, so read that rather than assuming another component's stricter rule applies.
What the technical proposal must contain
The Phase II Technical Objectives and Approach section must list specific technical objectives and provide a detailed technical approach, and must include these named subsections.
Phase II Work Plan, with an explicit, detailed description of the approach, indicating what is planned, how and where the work will be carried out, a schedule of major events, and the final product to be developed.
Related Work, describing significant activities directly related to the effort including those of the Principal Investigator, the firm, consultants, or others, and demonstrating awareness of the state of the art.
Relationship with Future Research or Research and Development, stating anticipated results and the significance of the Phase II effort as a foundation for Phase III.
Technology Transition and Commercialization Strategy, at no more than 2 pages counting toward the 15-page limit, addressing five specific questions: what is the first product this technology will go into; who will be your customers and what is your estimate of the market size; how much funding will you need to bring the technology to market and how will you raise those funds; does your company contain marketing expertise and if not how do you intend to bring it in; and who are your competitors and what is your price or quality advantage.
Key Personnel, including the Principal Investigator, with directly related education, experience, and relevant publications, and a concise resume of the PI.
Facilities and Equipment, describing available instrumentation and physical facilities, justifying equipment purchases including Government Furnished Equipment, and stating whether facilities meet federal, state, and local environmental laws across the named groupings.
Consultants, describing in detail any involvement of universities, academic institutions, or other consultants and identifying them in the Cost Volume.
Answer the five commercialization questions as five distinct answers, and note that the topic hands you the answer to the pricing part of question three: the $100,000 to $250,000 band.
The Company Commercialization Report is evaluated
Completion of the CCR as Volume 4 is required. The information contained in the CCR will be considered during proposal evaluations.
FutureG states this consistently in both its Phase I and Direct to Phase II sections. It is separate from the commercialization strategy in Volume 2: the CCR covers what you have done with past Phase II awards, the strategy covers how you propose to commercialize this research.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW solicitation.
Proposing firms will be notified of selection or non-selection status within 90 days of the closing date of the topic via DSIP. The FutureG text says "for a Phase I award," which appears to be residual language given that this topic issues no Phase I award. The notification will be sent to the individual listed as the Corporate Official on the proposal cover sheet, so make sure that is someone who will act on it.
Ninety days from October 21, 2026 is approximately January 19, 2027.
Refer to the DoW solicitation for procedures to protest the announcement. Protests after award should be submitted, as prescribed in FAR 33.106(b) and FAR 52.233-3, to osd.ncr.ousd-r-e.mbx.SBIR-STTR-Protest@mail.mil.
Questions
Specific questions pertaining to the administration of the FutureG SBIR Program and these proposal preparation instructions should be directed to the OUSW(R&E) FutureG Office at OSDRE-FutureG@groups.mail.mil.
The FutureG instructions do not state that DSIP Topic Q&A is unavailable, so the standard DoW process applies and Topic Q&A closes two weeks before the topic closes, on October 7, 2026.
The Reference
One, and it is a link to a vendor explainer: smart manufacturing, at ibm.com.
As with the companion venue topic, the formal reference list is not where the substance is. The real citations are embedded in the topic text and they are the reading list that matters: 3GPP TS 22.104 on service requirements for cyber-physical control applications, which you must map your use cases to; 3GPP TS 28.554 on key performance indicator definitions, named as an acceptable latency measurement methodology; 3GPP Release 16 and 17 NR positioning methods; 3GPP Release 17 RedCap and Release 18 eRedCap; 3GPP TS 33.501 on security; O-RAN WG11 specifications; NIST SP 800-207 on zero trust; the O-RAN 7.2x split; IEEE 802.11k, 802.11r, and 802.11v for the Wi-Fi comparison the topic makes; and IEEE Time-Sensitive Networking concepts.
That is ten substantive references embedded in prose against one marketing link in the reference section. The Related Work section is required to demonstrate awareness of the state of the art, and you carry that burden entirely. Bring the OCUDU project documentation, the OCUDU Testing and Validation Plan the topic refers to, the SD-Core and SD-RAN and OSC project documentation, and published RTEC test results.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
DSIP Topic Q&A closes: October 7, 2026, two weeks before the topic closes, per the DoW SBIR Program BAA
Proposal deadline: October 21, 2026
Selection notification: within 90 days of the closing date, approximately January 19, 2027
Period of performance: 18 months from award
A working backward plan
Before September 23. Audit the funding provenance of every feasibility result you intend to cite, since work based upon or logically extending from prior or ongoing federally funded SBIR or STTR work is excluded and failing the feasibility bar means the proposal is not evaluated. Resolve intellectual property ownership or license rights. Secure your demonstration site, confirming it is metal-heavy and multipath-rich and can host at minimum three cells with room to exercise handover at AGV speeds. Build the cost model against the $100,000 to $250,000 as-a-service price band and confirm it closes. Engage RTECs and identify which validated reference configurations exist for your intended RU, Core, RIC, and SMO combination. Get familiar with the OCUDU contribution process and the OCUDU Test and Evaluation Working Group cadence. Choose your two use cases. Plan your CBRS spectrum approach. Read TS 22.104 and prepare your service class mapping, and read TS 28.554 for the latency methodology. Model your Percentage of Work before assembling a team including an RU vendor, an integrator, a plant partner, and possibly a manufacturing institute. Confirm SAM registration and your CMMC posture.
September 23 through October 5. Draft the 5-page Phase I justification using the four accepted evidence categories, leaning on private, academic, and non-SBIR federally funded work. Draft the 13-page technical proposal covering the three research questions, your two use cases, RedCap handling, the IIoT and TS 22.104 requirements including NPN support and TSC and TSN awareness and IT and OT segmentation, the required performance-class mapping, the three common work package tasks, the RF network planning and site engineering task, the security architecture with your chosen live demonstration, your latency measurement reference points and budget decomposition approach, and the milestone plan with your assumed Month 14 and Month 16 sequence. Propose your own KPM set with threshold and objective values calibrated to your use cases, demonstration environment, spectrum plan, and channel bandwidth, and justify each. Draft the 2-page commercialization strategy answering all five enumerated questions, using the $100,000 to $250,000 band and the seven named adoption barriers. Draft the 3,000 character cover sheet abstract and the 3,000 character anticipated benefits and commercial applications discussion.
October 6 through October 7. Submit questions through DSIP Topic Q&A before it closes. The essential ones are the missing KPM tables, Section 3.0, and General System and Architectural Requirements section, and the Month 14 versus Month 16 milestone ordering. Send administrative questions to OSDRE-FutureG@groups.mail.mil.
October 8 through October 14. Build the cost volume online following the Cost Breakdown Guidance in the DoW 2026 SBIR BAA, against the $2,153,927 and 18-month ceiling. Price at least three radio units across multiple vendors, user equipment including reduced-capability classes, hardware accelerators, CBRS Spectrum Access System service, channel emulation and load generation, propagation modeling tools, VSWR and return-loss test equipment, RTEC testing engagement, site access and installation, OCUDU development labor including upstream contribution effort, security architecture work, and the recurring OCUDU T&E Working Group reporting. Identify Government Furnished Equipment needs. Complete the SBIR/STTR TABA Request Form and place it in Volume 5.
October 15 through October 18. Complete Volume 4, the Company Commercialization Report, carefully, since FutureG states it is considered during evaluations. Assemble Volume 5 with the TABA form and any letters from plant partners, manufacturing institutes, or RTECs that substantiate specific claims. Complete Volume 6 training and the Volume 7 foreign affiliations webform, remembering it must be the webform and will not be accepted as a PDF in Volume 5, and that no previous versions should be uploaded there. Run compliance: 5 plus 15 pages with the 2-page commercialization strategy inside the 15, no proprietary or classified information on the cover sheet, 3,000 character limits per cover sheet section.
October 19 through October 20. Submit and certify in DSIP.
Frequently Asked Questions
What is OSW-FutureG SBIR topic OSW26BZ06-DV032?
OSW26BZ06-DV032 is a Direct to Phase II SBIR topic titled "Smart Manufacturing," released under the OSW FutureG Office FY26 SBIR Broad Agency Announcement, Release 6. The objective is to design, validate, and demonstrate a fully open-source private 5G network platform for smart manufacturing integrating OCUDU, SD-Core, SD-RAN, and OSC components into a secure, vendor-neutral architecture supporting autonomous robotics, machine vision, and connected-worker safety.
How much funding is available?
Direct to Phase II proposals must not exceed a cost of $2,153,927 and a duration of 18 months. Phase II awardees may also request up to $50,000 in Technical and Business Assistance, in addition to the cost ceiling and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5.
When is the proposal deadline?
The topic opens September 23, 2026 and proposals are due October 21, 2026 through the Defense SBIR/STTR Innovation Portal at dodsbirsttr.mil.
Can I submit a Phase I proposal?
No. This topic is accepting Direct to Phase II proposals only, and a formal Phase I award will not be issued.
What software stack is required?
OCUDU for the Centralized Unit and Distributed Unit, SD-Core for the 5G core, SD-RAN for the Near-Real-Time RIC, and the OSC stack for the Non-Real-Time RIC and SMO, integrated into a single fully open-source reference architecture with no proprietary core, RIC, or SMO component anywhere in the stack.
What are the stated performance targets?
Sustained sub-30 millisecond threshold to sub-15 millisecond objective latency for machine-vision backhaul, at least 99.5 percent handover success at operational AGV and AMR speeds, and at least five-nines availability for safety-critical services, all in dense, metal-heavy RF environments. The topic also gives a worked example: a session success rate greater than 99.5 percent for AGV and AMR connections while handing over between at least 3 small cells at speeds up to 10 miles per hour.
Where are the KPM tables the topic refers to?
They do not appear in the published document, and there is no numbered Section 3.0 or General System and Architectural Requirements section. Raise it through DSIP Topic Q&A before it closes on October 7. The topic does state the governing principle: the KPM values in the tables are suggested reference values rather than pass or fail contract requirements, proposers shall propose specific justified KPM targets calibrated to their use cases, demonstration environment, spectrum plan, and channel bandwidth, deviations must be technically justified, threshold values represent the intended standard of demonstration success while objective values are stretch goals, and final targets are agreed with the Government at kickoff and confirmed at the Critical Design Review.
How is latency defined?
Round-trip application-layer latency measured between the user equipment application interface and the local edge application endpoint served behind the on-site or edge User Plane Function, meaning device to edge through the RAN and core user plane, excluding external WAN transport. Handover time means user-plane interruption time at the RAN per 3GPP definitions. For each latency KPM you must specify measurement reference points, traffic type and packet size, and methodology such as KPI definitions per 3GPP TS 28.554, and include a latency budget decomposition in the Critical Design Document across the air interface, DU and CU processing, fronthaul and F1 transport, core user plane, and application processing.
Do I have to demonstrate hard-real-time motion control?
No. Hard-real-time machine motion control, meaning isochronous traffic with cycle times of approximately 0.5 to 2 milliseconds per 3GPP TS 22.104, is explicitly outside the Phase II demonstration scope. The latency KPMs target AGV and AMR supervision and control-plane traffic, machine-vision backhaul, and safety alerting. You must, however, address in your FutureG evolution path the OCUDU enhancements required to approach the 1 millisecond latency class over time, including deterministic and priority scheduling, Time-Sensitive Communication support, and mini-slot and preemption features.
What is the target commercial price point?
A CBRS-based as-a-service offer at $100,000 to $250,000. This is the most actionable number in the topic and it bounds the entire design: radio unit count and class, integration labor, and recurring service margin.
What does the topic say about the market?
Roughly 50,000 U.S. manufacturing firms in the 20 to 99 employee band alone, with several thousand more in the 100 to 250 range, based on National Association of Manufacturers data. But the serviceable addressable market is not all manufacturers, it is the subset with real mobility, reliability, coverage, or security pain: plants using AGVs, AMRs, machine vision, connected workers, outdoor yards, large indoor spaces, or metal-heavy RF environments.
What are the stated barriers to adoption?
Seven, described as consistent and compounding: return-on-investment ambiguity, legacy equipment and retrofit costs, spectrum and regulatory fragmentation, uneven device ecosystems, IT and OT integration complexity, in-house skills shortages, and tight capital budgets. Addressing each explicitly in your commercialization strategy is cheap differentiation.
Should I argue that Wi-Fi cannot roam?
No, and the topic pre-empts it. It states that modern Wi-Fi roaming standards, IEEE 802.11k, 802.11r, and 802.11v, substantially mitigate fixed-access-point handoff behavior when properly deployed on capable client devices. The advantage sought is that a 3GPP system provides deterministic network-scheduled access on interference-managed spectrum, standardized mobility management under network control, and quality-of-service guarantees enforceable under load. It also endorses Wi-Fi coexistence rather than rip-and-replace.
How many use cases must I address?
At least two of the four, in your Phase II Statement of Work, and you must identify which your reference architecture and pilot deployment are designed to validate. The deliverables require demonstrating at least one of the two selected use cases, so you scope two and physically demonstrate at least one.
What are the four use cases?
AGV and AMR connectivity and mobility, which also requires on-floor localization. Machine vision backhaul. Connected-worker safety. And secure outdoor yard and logistics coverage.
Is localization required?
Yes, within Use Case 1. That use case additionally requires on-floor localization of AGVs, AMRs, and tagged mobile assets for fleet management, geofencing, and safety zoning. Network-native positioning using 3GPP NR positioning methods from Release 16 and 17 is preferred, and hybrid approaches fusing NR positioning with existing plant localization systems may be proposed with justification.
What Industrial IoT requirements apply?
For the manufacturing vertical specifically, proposers shall address IIoT service characteristics as framed by 3GPP TS 22.104, including mapping selected use cases to TS 22.104 communication service classes, support for private-network or Non-Public Network operation, awareness of Time-Sensitive Communication and IEEE Time-Sensitive Networking integration concepts in the platform architecture, and secure IT and OT segmentation for IIoT traffic such as via network slicing or 5G-LAN group management. This requirement has no counterpart in the companion venue topic.
What is the mandatory common work package?
Three tasks, required regardless of use cases selected. Task A, Baseline Benchmark, establishing a quantified performance baseline of the integrated stack, with emulated end-to-end configuration acceptable and encouraged, incorporating existing RTEC results where available, presented to the OCUDU Test and Evaluation Working Group, with the report expected by Month 3. Task B, Code and Feature Enhancement, closing gaps between baseline and threshold and objective values, with all OCUDU modifications contributed upstream. Task C, Benchmark and Regression Harness, delivering the emulation-based benchmark suite as a repeatable, documented, open-source harness for adoption by the OCUDU community and RTECs.
Is there an additional mandatory engineering task?
Yes, and it is unique to this topic. A representative facility-specific RF engineering task comprising predictive propagation modeling incorporating facility-specific features such as metal racking, machinery, mezzanines, tank farms, and loading areas; an RF design mitigating dead spots and multipath impairments through cell placement, antenna selection and orientation, and mobility-parameter tuning; and installation engineering practices protecting radio hardware, including antenna placement clearances from reflective metal and verification of antenna-system return loss and VSWR at commissioning with monitoring thereafter to prevent reflected-power damage to RU front ends. An RF Design Report with representative coverage maps and commissioning approaches is a deliverable.
Do I have to contribute my code upstream?
Yes. All modifications to OCUDU shall be contributed upstream through the project's standard contribution and review process, and enhancements that cannot be upstreamed shall be documented with rationale. Progress and upstream status are reported in Monthly Status Reports to the OCUDU T&E Working Group, and an itemized Upstream Contribution Log is a deliverable. This means your commercial differentiation cannot be the OCUDU code itself.
What are the security requirements?
A documented security architecture covering 3GPP security per TS 33.501 including mutual authentication and air-interface encryption and integrity protection; protection of the O-RAN open interfaces, meaning open fronthaul, E2, A1, and O1, per O-RAN WG11 specifications; zero-trust principles per NIST SP 800-207 including least-privilege access and separation of management and user traffic; and CVE monitoring for OCUDU and its dependencies with timely upstream patching. The architecture is documented at CDR, validated in RTEC testing, and the MVP demonstration must include at least one security capability shown live.
What are the Phase II milestones?
Month 1 kickoff and Technical Interchange Meeting, Monthly Status Reports throughout, Month 12 Critical Design Review, Month 16 prototype demonstration, Month 14 final design review and demonstration and assessment, and Month 18 Final Phase II Report. The published list places Month 16 before Month 14, which cannot be the intended order and appears as the same inversion in the companion topic DV031. Confirm through DSIP Topic Q&A and state your assumption in your work plan.
What does the demonstration site have to be?
Ideally an operating or representative manufacturing facility such as a partner small or medium manufacturer plant, a manufacturing institute or applied-research factory floor, or a comparable industrial or laboratory environment. It must be representative of the target deployment class in RF character, meaning metal-heavy and multipath-rich industrial construction, and in scale, meaning at minimum three cells sufficient to exercise inter-cell handover at operational AGV and AMR speeds. Its fidelity to plant conditions must be documented in the MVP demonstration package.
Is emulation acceptable?
For the Task A baseline, yes, and it is encouraged, provided the methodology, tooling, configurations, and results are fully documented and reproducible. But over-the-air performance with physical radio units and user equipment at the RTEC or representative demonstration site remains the standard of evidence for final KPM achievement.
How do I handle RedCap devices?
Address how the platform serves reduced-capability device classes including 3GPP Release 17 RedCap and Release 18 eRedCap, and identify any OCUDU scheduler or feature enhancements required, which are strongly encouraged as upstream contributions. Where RedCap-certified devices are not commercially available for a given endpoint type at demonstration time, you may demonstrate with available device classes or emulated RedCap UE profiles, and shall document the RedCap migration path.
What CMMC level applies?
The projected requirement for this topic is CMMC Level 2. Note that this topic states Level 2 without the parenthetical self-assessment qualifier that appears on the other two topics in this release, so if the distinction affects your compliance planning it is worth confirming through DSIP Topic Q&A.
Is this topic ITAR restricted?
No topic-level ITAR or EAR restriction paragraph appears on OSW26BZ06-DV032, and none appears on any of the three topics in this FutureG release.
How long can my technical volume be?
Twenty pages maximum, divided into Part 1 Phase I Justification at 5 pages maximum and Part 2 Phase II Technical Proposal at 15 pages maximum, with the Technology Transition and Commercialization Strategy at no more than 2 pages counting toward the 15. The FutureG instructions do not state that figures, tables, charts, and references count inside the limit or prohibit appendices, deferring instead to the DoW SBIR Program BAA formatting requirements.
Is the Company Commercialization Report evaluated?
Yes. FutureG states in both its Phase I and Direct to Phase II sections that information contained in the CCR will be considered during proposal evaluations. It is separate from the commercialization strategy in Volume 2.
Are there Percentage of Work restrictions?
Yes. The FutureG Office will not accept any deviation to the Percentage of Work requirements described in the DoW SBIR Program BAA. Model your POW before assembling a team that includes a radio unit vendor, an integrator, a plant partner, and possibly a manufacturing institute or university.
When will I hear back, and who gets notified?
Within 90 days of the closing date of the topic, approximately January 19, 2027, via DSIP. The notification goes to the individual listed as the Corporate Official on the proposal cover sheet. Note that the FutureG text says notification "for a Phase I award," which appears to be residual language given that this topic issues no Phase I award.
What is the defense application?
Modernizing DoW-affiliated depots, maintenance facilities, and logistics operations with resilient, high-mobility wireless connectivity, described as particularly relevant for contested logistics where reliable, secure, private communication networks are critical for maintaining operational tempo and supply chain integrity. Depots are structurally the same problem as commercial plants: large, metal-heavy, multipath-rich, with mobile equipment and outdoor yards.
Who do I contact with questions?
Technical questions about the topic go through DSIP Topic Q&A, which closes October 7, 2026. Administrative questions about the FutureG SBIR Program and these proposal preparation instructions go to the OUSW(R&E) FutureG Office at OSDRE-FutureG@groups.mail.mil.
Positioning Advice for Companies Considering This Topic
Audit your feasibility provenance before anything else. Feasibility documentation cannot be based upon or logically extend from any prior or ongoing federally funded SBIR or STTR work, and a proposal that fails the feasibility bar is not evaluated. Open-source 5G integration work in the United States has been heavily SBIR funded, so this is a real risk for the most qualified bidders. The topic's own fourth evidence category tells you where to look: private, academic, or non-SBIR federally funded work.
Build the cost model to $100,000 to $250,000 and show it closes. This is the only topic in the release that states a target price point, and it is the question the government asked. A platform that performs beautifully at $600,000 per plant does not answer it. Show the radio unit count, hardware class, integration labor, and recurring service margin that land inside the band, and be explicit about what you trade to get there.
Do not argue that Wi-Fi cannot roam. The topic concedes that 802.11k, 802.11r, and 802.11v substantially mitigate handoff behavior when properly deployed, and it names the real advantages: deterministic network-scheduled access on interference-managed spectrum, standardized network-controlled mobility, and enforceable quality of service under load. Make that case instead. And adopt the Wi-Fi coexistence posture the topic endorses rather than a rip-and-replace pitch.
Secure a metal-heavy three-cell site early. The demonstration environment must be representative in RF character and in scale, with at minimum three cells sufficient to exercise inter-cell handover at operational AGV speeds, and its fidelity must be documented. A partner plant, a manufacturing institute, or an applied-research factory floor is a dependency you cannot buy quickly. Name it on page one.
Ask about the missing KPM tables, then propose your own. The topic points at tables and a Section 3.0 that are not published, but it also tells you how the KPMs are meant to work: suggested reference values, not pass or fail, with proposers proposing justified targets calibrated to their environment and final targets agreed at kickoff and confirmed at CDR. Raise the gap in Topic Q&A and handle it professionally in the proposal.
Do the latency budget decomposition properly. Five named segments, air interface, DU and CU processing, fronthaul and F1 transport, core user plane, and application processing, in the Critical Design Document, with measurement reference points, traffic type, packet size, and methodology specified per latency KPM. It is a named requirement, it is genuinely useful engineering, and most proposals will state a latency number without decomposing it.
Respect the isochronous scope boundary. Hard-real-time motion control at 0.5 to 2 millisecond cycle times is out of scope for Phase II. Promising it reads as not having read the topic. Address the 1 millisecond class in your FutureG evolution path with deterministic and priority scheduling, TSC support, and mini-slot and preemption features, which is exactly what was asked.
Take the TS 22.104 mapping seriously. Mapping your selected use cases to TS 22.104 communication service classes is a concrete, checkable artifact, and NPN support, TSC and TSN awareness, and secure IT and OT segmentation are named requirements unique to this topic. IT and OT segmentation in particular is what a plant IT department will scrutinize, so network slicing or 5G-LAN group management deserves real treatment.
Pick Use Cases 1 and 2 unless you have a reason not to. They carry the topic's two stated performance targets, handover success and machine-vision latency, they exercise both dominant performance classes, they share the indoor RF design, and they align with the two named OCUDU enhancement areas of mobility and handover optimization and uplink capacity. Use Case 3 pairs cheaply with either and adds the RedCap and five-nines dimensions.
Answer the trust question as a document. What does a plant manager with no RF staff need to see, in what form, to sign for a network with no vendor warranty? Sketching that evidence package, even as a table of contents, differentiates you from proposals that answer with a test matrix.
Treat the RF engineering task as a differentiator, not a chore. Predictive propagation modeling in metal-heavy environments, dead-spot mitigation through cell placement and antenna orientation, and VSWR verification at commissioning to prevent reflected-power damage to RU front ends. That last item is a practitioner's requirement, and answering it with actual installation practice signals that you have deployed in a factory rather than modeled one.
Budget the common work package separately. Tasks A, B, and C are required regardless of use case, the baseline report is due by Month 3, and the harness is an open-source deliverable documented for independent execution. Folding it into general engineering underprices it.
Plan for upstream review you do not control. All OCUDU changes go upstream on the project's timeline, the log includes review status, and non-upstreamable enhancements need documented rationale. Say how you sequence contributions and what happens to a KPM claim if a patch is still in review at Month 18.
State how you make money when the code is public. Integration, RF engineering, the evidence package, the as-a-service model, and support are your differentiation. A commercialization strategy that avoids this looks naive to a reviewer who wrote the upstream requirement.
Address the seven adoption barriers one by one. ROI ambiguity, legacy equipment and retrofit costs, spectrum and regulatory fragmentation, uneven device ecosystems, IT and OT integration complexity, skills shortages, and tight capital budgets. The topic says a tier-one proprietary platform addresses almost none of them. Showing that you address each is the clearest possible articulation of why you should be funded.
Plan the page budget before drafting. This topic has more mandatory content than its venue companion, including the IIoT requirements and the RF engineering task, and the same 5 plus 13 plus 2 page allowance. Decide the allocation first.
OSW-FutureG SBIR OSW26BZ06-DV031: Smart Venues and Stadiums, Open-Source Private 5G
Deadline: October 21, 2026
Funding Award Size: $2m
Description: Complete guide to OSW-FutureG SBIR Direct to Phase II topic OSW26BZ06-DV031, open-source private 5G for stadiums and large venues with 6G ISAC path. Up to $2,153,927 over 18 months. Closes October 21, 2026.
Quick Answer
OSW26BZ06-DV031 is a Direct to Phase II SBIR topic under the OSW FutureG Office, FY26 SBIR Broad Agency Announcement, Release 6. No Phase I award will be issued. The government wants a private 5G network for stadiums and large venues built entirely from open-source software, with no proprietary core, RAN Intelligent Controller, or Service Management and Orchestration component anywhere in the stack, validated at a real venue under real event-day congestion, and architected to evolve toward 6G Integrated Sensing and Communication. The award must not exceed $2,153,927 over 18 months, with a 20-page technical volume. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The specific stack is named and it is not optional: OCUDU for the Centralized Unit and Distributed Unit, SD-Core for the 5G core, SD-RAN for the Near-Real-Time RIC, and the OSC stack for the Non-Real-Time RIC and SMO. If your architecture includes a commercial core or a vendor RIC, it does not meet the topic.
The economic argument is unusually detailed for a solicitation. Distributed Antenna System deployments in large public venues have grown from a few million dollars in early builds to routinely tens of millions. Active multi-carrier DAS hardware and installation is commonly quoted at $5 to $10 per square foot, meaning even a mid-sized 300,000 square foot concourse and bowl footprint implies a multi-million-dollar build before integration, spectrum, and ongoing carrier-management costs. And mobile network operator appetite to fund new DAS builds outside Tier 1 venues, meaning 70,000-plus capacity or 200-plus events per year, is shrinking, leaving small and mid-sized venues with a widening connectivity gap.
There is one thing you should know before reading further. The topic repeatedly directs proposers to key performance metrics in "the table below" and to "Section 3.0" thresholds. No such tables and no numbered Section 3.0 appear in the published document. That gap is discussed in its own section below, and it is the most important question to ask the government before the topic closes.
Topic At a Glance
Topic number: OSW26BZ06-DV031
Title: Smart Venues and Stadiums
Agency: Office of the Secretary of War, FutureG Office, under OUSW(R&E)
Solicitation: OSW FutureG Office, FY26 SBIR Broad Agency Announcement, Release 6, Proposal Submission Instructions
Program type: Direct to Phase II only. This topic is accepting Direct to Phase II proposals only, and a formal Phase I award will not be issued
Award: must not exceed $2,153,927
Period of performance: 18 months
Technical volume limit: 20 pages maximum, structured as Part 1 Phase I Justification at 5 pages maximum and Part 2 Phase II Technical Proposal at 15 pages maximum, with the Technology Transition and Commercialization Strategy at no more than 2 pages counting toward the 15
OUSW (R&E) Critical Technology Areas: Applied Artificial Intelligence (AAI), Contested Logistics Technologies (LOG)
Component Technology Priority Areas: FutureG, Sustainment and Logistics
Projected CMMC level requirement: Level 2 (Self)
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic
Mandatory stack: OCUDU plus SD-Core plus SD-RAN plus OSC, with no proprietary core, RIC, or SMO component anywhere
Dominant performance classes: extreme user density, ultra-secure operation, and integrated sensing
Use cases: proposers must address at least two of four, and must identify which their reference architecture and pilot deployment are designed to validate
Mandatory common work package: OCUDU baseline benchmarking and upstream enhancement, Tasks A, B, and C, required regardless of use cases selected
Upstream requirement: all modifications to OCUDU shall be contributed upstream through the project's standard contribution and review process
Demonstration site: the proposer's site, ideally an operating or representative venue or stadium environment, or a full-scale representative test bed where live-venue access is not feasible during Phase II
Technical and Business Assistance: up to $50,000 per Phase II project, in addition to the cost ceiling and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5
Percentage of Work: the FutureG Office will not accept any deviation to the POW requirements
Company Commercialization Report: information contained in the CCR will be considered during proposal evaluations
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026
Submission portal: DSIP at dodsbirsttr.mil
Keywords: smart venues and stadiums, connected stadiums, 5G stadiums
The Feasibility Bar, Which Is the First Thing to Check
This topic is accepting Direct to Phase II proposals only, so a formal Phase I award will not be issued.
To qualify for a Phase II award, proposers must submit Feasibility Documentation as part of their proposal package demonstrating that they have already completed Phase I-type research and development. The purpose of this documentation is to prove that the underlying technology is mature, scientifically sound, and ready to transition immediately into the Phase II prototyping and testing environment.
What the government will accept
The Government will accept a wide variety of documentation styles. Proposers do not need to have a perfect, finished product, but they must show that their core ideas have been successfully tested. One or more of the following items should be included to prove technology readiness.
Test data and metrics: real-world measurements, performance charts, or network test data from previous lab environments or early field trials.
Technical reports and white papers: written summaries explaining your previous research, system designs, or software integration efforts.
Prototype designs and simulation models: diagrams, architectural blueprints, or computer simulation results showing how your proposed software and hardware components interact.
Previous project outcomes: success criteria, milestone reports, or commercialization results from prior private, academic, or non-SBIR federally funded work.
This is among the most accommodating feasibility framings in the 2026 cycle. "One or more of the following" and "do not need to have a perfect, finished product" are real allowances, and simulation models are explicitly acceptable.
The restriction that still applies, and it matters here
The FutureG Direct to Phase II guidelines impose a hard constraint that the topic-level text does not repeat.
Feasibility documentation cannot be based upon or logically extend from any prior or ongoing federally funded SBIR or STTR work. Work submitted within the feasibility documentation must have been substantially performed by the proposer or the principal investigator. If technology in the feasibility documentation is subject to intellectual property, the proposer must either own the IP or must have obtained license rights to such technology prior to proposal submission, to enable it and its subcontractors to legally carry out the proposed work.
The Volume 2 instruction phrases it as "must not be solely based on" prior or ongoing federally funded SBIR or STTR work, which is weaker. Plan against the stricter formulation.
Notice that the topic's own fourth acceptable evidence category says "prior private, academic, or non-SBIR federally funded work," which is consistent with the restriction and tells you exactly where to look. Non-SBIR federally funded work is explicitly fine. Private and academic work is fine. Prior SBIR and STTR work is the problem.
This is a live risk for exactly the companies most likely to bid, because open-source 5G integration work in the United States has been substantially SBIR funded. Audit the provenance of every result you intend to cite.
And the consequence is severe: if the proposer fails to demonstrate technical merit and feasibility equivalent to the Phase I level as described in the topic, the related Phase II proposal will not be evaluated.
The Missing KPM Tables
This section exists because the gap is material and you should not discover it while drafting.
The topic refers to key performance metric tables and to a numbered section repeatedly. It says other suggested KPMs for the various use cases and system requirements are given below for reference. In Use Case 2 it says the positioning and velocity accuracy, anomaly alert, and missed-detection and false-alarm KPMs "in the table below" apply to the Tier 1 demonstrated functions. Task A requires establishing a baseline against the KPMs relevant to the selected use cases "including the General System and Architectural Requirements." The deliverables list requires MVP demonstration documentation reporting "the KPI results achieved against Section 3.0 thresholds" and a baseline benchmark report with "measured baseline results against Section 3.0 KPMs."
No KPM tables appear in the published document. There is no numbered Section 3.0, and no section titled General System and Architectural Requirements.
What to do about it
Ask. DSIP Topic Q&A closes October 7, 2026, two weeks before the topic closes, and this is the single highest-value question available on this topic. Ask whether the KPM tables and the General System and Architectural Requirements section were omitted from the published instructions and, if so, request them or a pointer to them.
In the meantime, the topic gives you enough to proceed and in fact tells you to. It says proposers should address specific, relevant, measurable, and quantifiable KPMs including, for example, concurrent-device density, sensing detection accuracy, latency, jitter, and service availability. And it gives one worked example: "The platform must sustain a session success rate greater than 99.5% for fan-facing connections at a concurrent-device density of 3,000 devices per acre in the lower bowl."
The companion topic in this same release, DV032 Smart Manufacturing, states the governing principle explicitly and it is reasonable to read it as the program office's general intent: the KPM values in the tables are suggested reference values, not pass/fail contract requirements; proposers shall propose specific, justified KPM targets in the Technical Volume, calibrated to their selected use cases, demonstration environment, spectrum plan, and channel bandwidth; deviations from suggested values must be technically justified; threshold values represent the intended standard of Phase II demonstration success while objective values are stretch goals; and final KPM targets will be agreed with the Government at kickoff and confirmed at the Critical Design Review.
So the practical approach is to propose your own KPM set with threshold and objective values, justify each against your architecture and spectrum plan, state plainly that you are doing so because the referenced tables do not appear in the published instructions, and note that final targets will be agreed at kickoff and confirmed at CDR. That is a defensible position and it demonstrates that you read the document carefully, which is itself worth something.
What the Government Is Actually Buying
The objective
The objective of this Phase II effort is to design, validate, and demonstrate a secure, high-density private 5G network platform for stadiums and large venues, integrating open-source OCUDU, SD-Core, SD-RAN, and OSC components, to deliver resilient, multi-service communication alongside a 6G-ready Integrated Sensing and Communication architecture.
This platform will be validated at a representative venue site to prove stable performance under extreme device congestion, enable native device localization for crowd analytics, and establish a cost-effective, vendor-neutral dual-use framework for both commercial entertainment venues and dense military installation environments.
The market, as the government describes it
The realistic near-to-mid-term serviceable addressable market for private 5G and 6G-ready platforms among U.S. venues and stadiums is a subset of venues with acute connectivity, safety, and operational pain: major-league and collegiate stadiums, multi-purpose arenas, outdoor amphitheaters, and convention and exhibition centers that host recurring high-density events, require public-safety-grade coverage throughout bowls, concourses, tunnels, and back-of-house areas, and are increasingly asked to support fan-facing digital experiences, cashierless concessions, camera and sensor networks, and first-responder communications simultaneously on event day.
The United States features an extensive footprint of sports and entertainment infrastructure, spanning an estimated 2,000 to 5,000 major stadiums and arenas when accounting for all professional, collegiate, and municipal facilities, which further scales to over 15,000 total venues if relatively large localized high school stadiums are included. Beyond stadiums and arenas, industry directories catalog well over 400 U.S. convention and exhibition centers, ranging from single-hall regional facilities to multi-million-square-foot complexes. Together, these figures put the large-venue addressable base at several hundred U.S. facilities, before counting mid-sized arenas, amphitheaters, and secondary convention space. Localized large gatherings such as fairs add up to several thousand more.
Note the careful narrowing. Fifteen thousand venues exist, but the serviceable addressable base is "several hundred U.S. facilities." Your commercialization strategy should use the government's own narrower figure rather than the headline number, because inflating it against the topic's own text is an easy weakness to spot.
The pain points
Venues already invest heavily in connectivity infrastructure, yet several published industry reports indicate persistent cost and coverage pain. Distributed Antenna System deployments in large public venues have grown from a few million dollars in early builds to routinely tens of millions of dollars. Independent of venue scale, active multi-carrier DAS hardware and installation is commonly quoted at $5 to $10 per square foot, meaning even a mid-sized 300,000 square foot concourse and bowl footprint implies a multi-million-dollar build before integration, spectrum, and ongoing carrier-management costs.
Despite this infrastructure spend, published reviews consistently cite coverage dead zones in lower bowls and concourses, congestion during peak moments such as kickoff, halftime, and entry and egress, and multi-year vendor lock-in as recurring pain points. Industry reporting suggests that mobile network operator appetite to fund new DAS builds outside Tier 1 venues, defined as 70,000-plus capacity or 200-plus events per year, is shrinking, leaving small and mid-sized venues with a widening connectivity gap.
How this segment buys
This segment buys on measurable fan and operational outcomes, specifically dropped-call rate at kickoff, point-of-sale uptime, camera uptime, and incident-response time, not on global platform standardization.
Its stated barriers to adopting anything other than the incumbent DAS and neutral-host model are consistent: high capital expenditure and multi-year contract terms, uncertainty about interoperability across multiple wireless carriers that must all be hosted on one system, integration complexity with legacy venue information technology and operational technology and public-safety radio systems, and a lack of in-house RF and cellular engineering staff.
Those four named metrics are the ones to build your value proposition around. Dropped-call rate at kickoff, POS uptime, camera uptime, incident-response time. They are also natural KPMs, and using the buyer's own language is more persuasive than throughput figures.
The open-source argument, and its honest cost
An all-open-source stack removes the structural cost and lock-in that makes proprietary DAS platforms a poor fit for this segment's economics. Because OCUDU communicates with the rest of the network over open, standardized interfaces, especially the O-RAN 7.2x split to the O-RU, the network can be assembled from whichever radio unit vendor and antenna form factor best fits a given venue's bowl geometry, concourse layout, and budget.
Together, OCUDU, SD-Core, SD-RAN, and OSC form one integrated, fully open-source 5G stack that a systems integrator can deploy, support, and price as a right-sized, as-a-service offer for a single venue, rather than a broad, vendor-locked, multi-carrier neutral-host platform.
The trade-off of an all-open-source stack is the same one seen in other verticals: the burden of proving interoperability, stability, and security shifts from a single vendor's warranty to the integrator and the open-source community.
Venue operators, who by their own account do not carry in-house cellular or RF engineering expertise, have essentially zero tolerance for connectivity failures discovered live, on a broadcast event, in front of tens of thousands of people.
That last sentence is the commercial crux of the topic and it deserves a direct answer in your proposal. The topic gives you the intended answer: this is exactly the gap the ongoing OCUDU Testing and Validation Plan, and its RTEC-centered extensions, are designed to close, through independently witnessed, multi-vendor testing of device diversity, RU diversity, high-density and high-mobility RF performance, Core, RIC, and SMO integration, and security. The proposal should leverage the current, ongoing testing and evaluation activities in RTECs and associated results.
The generalization requirement
Although this topic is anchored in a specific vertical to ensure a concrete deployment environment, real users, and a defensible commercialization path, proposers should recognize and are required to present the technical work in terms of the generic network performance classes it advances.
Improvements made to the OCUDU O-CU and O-DU under this effort, meaning scheduler behavior, mobility management, uplink capacity, stability under sustained load, and security features, are expected to generalize across verticals and to benefit the broader community of RAN developers building on OCUDU.
The Technical Volume shall include a mapping of each selected use case, and its associated KPMs, to the performance class or classes it exercises, and shall identify which anticipated OCUDU code or feature enhancements correspond to each class.
This is a required Technical Volume element, stated with "shall." It is a mapping table in substance, and it is easy to omit while writing about the venue application. Do not omit it.
For this topic, the dominant performance classes are extreme user density, ultra-secure operation, and integrated sensing: tens of thousands of concurrent devices per venue with sharp event-day load peaks, public-safety-grade availability and access control across bowls, concourses, and back-of-house areas, and an architecture evolvable to 6G Integrated Sensing and Communication.
The Four Research Questions
Research and development for this effort should address, at minimum, the following four questions. Treat them as required sections.
All-open-source economics
What is the fully loaded cost, covering integration, support, spectrum, hardware, and RF design, of an OCUDU plus SD-Core plus SD-RAN plus OSC deployment at venue scale relative to a proprietary DAS and neutral-host platform, and how should that be packaged as a CBRS-based, as-a-service offer within the price point venue operators require?
Note "CBRS-based." Citizens Broadband Radio Service spectrum is the assumed band, which shapes your RF design, your device ecosystem, and your Spectrum Access System dependency. Note also "as-a-service," meaning the business model is recurring rather than capital sale. Unlike the companion manufacturing topic, this one does not state a target price point, so you must establish it from the DAS comparison the topic gave you.
Use-case-first deployment
Which of the use cases defined below deliver the fastest, most measurable return on investment on the open-source stack, and what RTEC-executed interoperability tests across Device, RU, Core, RIC and SMO, and Infrastructure dimensions are needed to validate each one end-to-end under stadium-scale, high-density loading conditions?
6G and ISAC evolution path
What architectural changes across waveform, RU capability, RIC application, and SMO orchestration are required to evolve the platform from a 5G communications-only deployment to a 6G-ready platform capable of Integrated Sensing and Communication, and which ISAC-enabled use cases can be prototyped now using 5G-Advanced sensing features as a bridge to native 6G ISAC?
Trust in open source
How does RTEC-executed testing of the full open-source stack reduce the integration risk and skills-shortage barrier venue operators most often cite, and what evidence package best converts community-maintained, no-license-fee software into a procurement-ready proof point for a buyer with no in-house cellular expertise?
That fourth question is the most commercially important and the one most proposals will answer weakly. The deliverable it implies is an evidence package: what does a venue general manager with no RF staff need to see, in what form, to sign a contract for a network with no vendor warranty behind it? Answer it as a document design problem, not as a testing plan.
Solutions leveraging artificial intelligence and machine learning for crowd analytics, predictive maintenance, network optimization, or automated fault resolution are encouraged but not required. The government will consider any novel concept that increases the reliability, economics, and trustworthiness of an all-open-source private 5G and FutureG platform for the venues and stadiums vertical. Dual-use opportunities are expected across commercial venue operators, DoW-affiliated event and installation venues, and public-safety agencies.
The Four Use Cases, Of Which You Must Address Two
Proposers must address at least two of the following four use cases in their Phase II Statement of Work and must identify which use cases their reference architecture and pilot deployment are designed to validate.
Device-class diversity and RedCap, which applies across all use cases
Each use-case set spans device classes from full-capability user equipment, meaning equipment modems, cameras, and broadcast and production units, to reduced-capability Internet of Things endpoints, meaning sensors, wearables, tags, and trackers.
Proposers shall address how the platform serves reduced-capability device classes, including 3GPP Release 17 Reduced Capability, RedCap, and Release 18 eRedCap user equipment, and shall identify any OCUDU scheduler or feature enhancements required to support RedCap operation. Such enhancements are strongly encouraged as upstream contributions under the common OCUDU benchmarking and enhancement work package.
Where RedCap-certified devices are not commercially available for a given endpoint type at demonstration time, proposers may demonstrate with available device classes or emulated RedCap user equipment profiles, and shall document the RedCap migration path.
That final allowance is a practical accommodation worth using. RedCap device availability in CBRS bands is limited, and the topic explicitly permits emulated profiles with a documented migration path.
Use Case 1: High-density fan connectivity and broadcast and production backhaul
Tens of thousands of fans, concession point-of-sale terminals, ticketing scanners, and broadcast and production camera and audio feeds must all operate simultaneously within a single bowl and concourse footprint, with demand spiking sharply at kickoff, halftime, and egress. Legacy DAS and Wi-Fi are commonly saturated at these peak moments, producing dropped sessions and stalled transactions.
This use case requires the platform to sustain high concurrent-device density with consistent per-user throughput, plus dedicated low-latency, low-jitter uplink capacity for broadcast and production camera backhaul, across the full bowl and concourse footprint.
Note that this bundles two different problems: massive downlink-and-uplink density for consumer devices, and guaranteed low-jitter uplink for a small number of professional video feeds, on the same network at the same moment. The quality-of-service and slicing story is the answer, and it should be explicit.
Use Case 2: ISAC-enabled crowd analytics and situational awareness, 6G-ready
Venue operators and public-safety personnel need continuous, accurate awareness of crowd density, flow, and anomalies such as bottlenecks at egress, unattended objects, and unauthorized drone incursion into restricted airspace above the bowl, to prevent crushing incidents and stampedes and respond quickly to emerging threats. Deploying a separate dedicated sensor network of radar, lidar, or additional camera arrays to provide this awareness is costly and adds another system to integrate.
This use case requires the platform to support Integrated Sensing and Communication, which uses the same radio infrastructure and spectrum that carries communication traffic to also sense the physical environment, through bistatic or monostatic sensing from gNB and O-RU hardware correlated with RIC-hosted analytics applications, without requiring a separate sensing-only network build.
The topic cites 3GPP Release 19 Technical Report 22.837, which identifies more than 30 ISAC use cases spanning object and intruder detection, environmental monitoring, motion sensing, and public-safety scenarios directly applicable to stadiums, arenas, and convention centers. It also cites 3GPP TS 22.137, which establishes a standardized set of ISAC performance metrics including positioning accuracy, velocity accuracy, sensing resolution, sensing range, refresh rate, latency, and missed-detection and false-alarm probabilities.
Proposers should treat 5G-Advanced sensing features as the near-term bridge path toward native 6G ISAC.
The hybrid localization requirement
Proposers shall employ a hybrid architecture in which standardized network-native user equipment positioning, meaning 3GPP Release 16 and 17 NR positioning using uplink time difference of arrival, multi-round-trip-time, and angle of arrival and departure, over the connected device population provides the primary, near-term source of crowd density, flow, and bottleneck analytics, with appropriate aggregation and anonymization for fan-facing devices, and RF sensing, meaning ISAC, is applied to non-cooperative targets that carry no connected device.
Crowd-analytics KPMs may be satisfied via UE positioning in the Phase II demonstration.
That last sentence is a significant de-risking allowance. You can meet the crowd analytics metrics using standardized positioning of connected phones rather than true RF sensing. Read it together with the tiering below.
The tiered scope, which is the most important paragraph in this use case
Sensing functions under this use case are tiered by 12-month feasibility.
Tier 1, for the Phase II demonstration, comprises two things. First, crowd density and flow analytics derived from network-native UE positioning of connected devices, per the hybrid-localization requirement, aggregated and anonymized. Second, at least one non-cooperative RF-sensing bridge function achievable at 5G-Advanced maturity and deployed sensing bandwidths, for example detection of unauthorized drone incursion exploiting Doppler and motion signatures, demonstrated in a laboratory or limited field configuration.
Tier 2 is roadmap only: unattended-object detection and fine-grained tracking of individuals within dense, multi-directional crowds, which are limited by achievable sensing resolution of approximately c over 2B at deployable bandwidths. These shall be addressed in the documented 6G and ISAC evolution path with quantified bandwidth, waveform, aperture, and sensor-fusion requirements rather than demonstrated in Phase II.
Custom or modified waveforms are not required for Phase II.
The positioning and velocity accuracy, anomaly alert, and missed-detection and false-alarm KPMs referenced in the tables apply to the Tier 1 demonstrated functions.
This tiering is the government being realistic, and it is written by someone who understands the physics. The c over 2B range resolution limit means that at CBRS-scale bandwidths you cannot resolve individual people in a crowd, and the topic says so rather than asking you to pretend otherwise. Respect the tiering. A proposal that promises Tier 2 capability in Phase II is not more ambitious, it is less credible.
Note also that a laboratory or limited field configuration is acceptable for the Tier 1 non-cooperative sensing function. Drone detection by Doppler signature in a controlled setting satisfies it.
Use Case 3: Public safety and first responder priority communications
First responders, venue security, and event staff require reliable, prioritized push-to-talk and data communications throughout the bowl, concourses, tunnels, loading docks, and below-grade back-of-house areas that legacy DAS and Wi-Fi frequently fail to reach or fail to prioritize during network congestion.
This use case requires the platform to sustain low-latency, high-reliability, priority and preemption-capable connectivity for public-safety traffic across the full indoor venue footprint, including areas outside typical fan-facing coverage design.
Priority and preemption is a specific 3GPP capability set and it is the part of this use case that has to actually work under the Use Case 1 load. If you address both use cases, show the interaction: public-safety traffic preempting fan traffic at kickoff is the test.
Use Case 4: Venue operations, logistics, and outdoor perimeter coverage
Loading docks, outdoor plazas, parking structures, and perimeter areas surrounding a venue increasingly require connectivity for autonomous cleaning and security robots, connected cameras, RFID and asset tracking for concessions and merchandise logistics, and credentialed-access control, yet typically fall outside indoor DAS and Wi-Fi coverage entirely.
This use case requires the platform to extend secure, private coverage into outdoor plaza, parking, and perimeter areas immediately adjacent to the venue, using the same OCUDU-based infrastructure rather than a separate outdoor Wi-Fi or cellular repeater buildout.
Choosing your two
Some combinations are cheaper than others. Use Cases 1 and 3 share the indoor bowl and concourse RF design and differ mainly in quality-of-service treatment, which makes them the most economical pair and the most directly aligned with the buying metrics the topic named. Use Case 2 requires ISAC-capable radio units or software-defined sensing functions plus RIC-hosted analytics, which is additional hardware and additional integration, though the Tier 1 tiering and the UE-positioning allowance make it more tractable than it first appears. Use Case 4 requires outdoor cell planning and additional radio units.
State your choice explicitly and say what your reference architecture and pilot deployment are designed to validate, because the topic requires you to identify that.
The Mandatory Common Work Package
All performers under this topic shall execute the following common work package, which is a required element of the Phase II Statement of Work regardless of the use cases selected.
This is not optional and it is not scoped by your use case choice. Budget it and staff it separately.
Task A: Baseline Benchmark
Establish a quantified performance baseline of the integrated open-source stack, meaning OCUDU plus SD-Core plus SD-RAN plus OSC, against the KPMs relevant to the selected use cases, including the General System and Architectural Requirements.
An emulated end-to-end configuration, using emulated radio units and user equipment, RF channel emulation, or synthetic load generation, is acceptable and encouraged for the baseline, provided the benchmark methodology, tooling, configurations, and results are fully documented and reproducible.
Where an RTEC-validated reference configuration already exists for the proposed RU, Core, RIC, and SMO combination, the baseline shall incorporate available RTEC results rather than duplicate them.
Baseline methodology and results shall be presented to the OCUDU Test and Evaluation Working Group.
Note that the baseline benchmark report is expected by Month 3. That is early, and it means your emulation environment has to be standing up in the first weeks of the award.
Task B: Code and Feature Enhancement
Identify the gaps between baseline performance and the threshold and objective values, and develop the OCUDU code improvements and features required to close them, for example scheduler and quality-of-service enhancements, mobility and handover optimization, uplink capacity improvements, stability hardening, and security features.
All modifications to OCUDU shall be contributed upstream through the project's standard contribution and review process. Enhancements that cannot be upstreamed shall be documented with rationale.
Progress against each targeted KPM, and the status of each upstream contribution, shall be reported in the Monthly Status Reports presented to the OCUDU Test and Evaluation Working Group.
The upstream contribution requirement has real business consequences. Your code improvements go into a public open-source project, reviewed by that project's maintainers on their timeline. That means your differentiation cannot be the OCUDU code itself. It has to be the integration, the RF engineering, the evidence package, the service model, and the support relationship. Say so in your commercialization strategy rather than leaving a reviewer to wonder how you make money.
It also means schedule risk you do not control. Upstream review timelines belong to the project, and the deliverable is an itemized log including review status, which acknowledges that acceptance is not guaranteed. Plan for contributions in review at the end of the period of performance.
Task C: Benchmark and Regression Harness
Deliver the emulation-based end-to-end benchmark suite developed under Task A as a repeatable, documented, open-source harness suitable for adoption by the OCUDU community and RTECs for regression testing of future OCUDU releases.
This work package complements, and does not replace, the interoperability testing and the physical MVP demonstration required elsewhere in this topic. Emulated results establish the baseline and guide enhancement work. Over-the-air performance with physical radio units and user equipment at the RTEC or representative demonstration site remains the standard of evidence for final KPM achievement.
That final sentence is the one to underline. Emulation gets you the baseline. Over-the-air with real hardware is what counts for final KPM achievement.
Security Requirements
Security shall be a first-class design requirement of the platform, not a demonstration afterthought.
Performers shall implement and document a security architecture covering the following.
3GPP security per TS 33.501, including mutual authentication and air-interface encryption and integrity protection.
Protection of the O-RAN open interfaces, meaning open fronthaul, E2, A1, and O1, per O-RAN WG11 specifications.
Zero-trust principles per NIST SP 800-207, including least-privilege access and separation of management and user traffic.
Monitoring of Common Vulnerabilities and Exposures affecting OCUDU and its dependencies, and timely upstream patching.
Security features and hardening developed for OCUDU shall be contributed upstream under the common benchmarking and enhancement work package.
The security architecture shall be documented at the Critical Design Review and validated in RTEC testing, and the MVP demonstration shall include at least one security capability shown live, for example rejection of an unauthorized device, encrypted fronthaul, or detection of a simulated intrusion.
Four named standards and one live demonstration requirement. The live security demonstration is easy to underplan, so pick which capability you will show and design the demonstration around it early. Note also that ultra-secure operation is one of the three dominant performance classes for this topic, alongside extreme user density and integrated sensing, so security is a scored dimension rather than a compliance section.
Milestones and Deliverables
Milestones as stated
Month 1: Kickoff and Technical Interchange Meeting.
Monthly Status Reports throughout.
Month 12: Critical Design Review.
Month 16: Prototype demonstration.
Month 14: Final design review, demonstration, and assessment.
Month 18: Final Phase II Report.
Note the ordering. The published list places the Month 16 prototype demonstration before the Month 14 final design review, which cannot be the intended sequence. The same inversion appears in the companion topic DV032 in this release, which suggests a shared drafting error rather than a deliberate structure. The sensible reading is a Month 14 prototype demonstration followed by a Month 16 final design review and assessment, or the reverse; either way, confirm through DSIP Topic Q&A before you build a schedule around it, and state your assumed sequence in your work plan.
Monthly Status Reports are to be presented to the OCUDU Test and Evaluation Working Group as well, bringing the community up to speed on progress. That is a recurring external commitment, not just an internal report, and it should be staffed.
The demonstration site
Prototype demonstrations will be performed at the proposer's site, ideally an operating or representative venue or stadium environment, or a full-scale representative test bed where live-venue access is not feasible during Phase II.
Partial solutions may be considered successful if effective within a defined scope. A final technical report detailing the capabilities demonstrated will be required. Extended user evaluations or additional prototypes may be pursued based on utility.
The MVP demonstration, including physical radio units and user equipment, and for Use Case 2 ISAC-capable RU and sensing hardware or software-defined sensing functions, will need to occur at the venue-representative site.
Securing a venue relationship is the practical crux of this proposal. An operating stadium, arena, amphitheater, or convention center that will let you install radio units and run tests, or a full-scale representative test bed, is a dependency you do not control and cannot buy quickly. If you have a venue partner, name it on page one. If you have a test bed, describe its fidelity to venue conditions, because the deliverables require a description of the demonstration site's fidelity to venue and stadium conditions.
Scalability and RTEC leverage
Performers should address scalability, including testing across multiple radio unit vendors and hardware-accelerator options, and should leverage RTEC-executed interoperability testing to address integration risks ahead of the demonstration wherever a validated reference configuration already exists.
Phase II deliverables
Kickoff and Technical Interchange Meeting slides.
Monthly Status Reports.
A Critical Design Document containing the full reference architecture across OCUDU, SD-Core, SD-RAN, and OSC, including the 6G and ISAC evolution path.
Key Performance Metrics.
MVP Demonstration slides and documentation, including a description of the demonstration site's fidelity to venue and stadium conditions and the KPI results achieved against the referenced thresholds.
A Reference Configuration Package comprising executables, integration documentation, and RTEC test results for the validated RU, Core, RIC, and SMO combinations used.
Integration of the platform into an RTEC-affiliated test and evaluation network or a venue-representative demonstration site, demonstrating at least one of the two selected use cases.
An OCUDU Baseline Benchmark Report covering methodology, emulation environment description, configurations, and measured baseline results, expected by Month 3.
An Upstream Contribution Log, itemizing OCUDU code contributions such as patches and pull requests, their review status, and the KPM gap each addresses, updated in each Monthly Status Report with the final version in the Final Technical Report.
An open-source benchmark and regression harness, with documentation sufficient for independent execution by RTECs and the OCUDU community.
A Final Design Document.
A Final Technical Report.
Two notes on this list. The published deliverables include a duplicated integration bullet, one referring to a venue-representative demonstration site and one referring to a factory-representative demonstration site. The factory reference is plainly carried over from the companion Smart Manufacturing topic and does not apply here.
Also note the relationship between "address at least two use cases" and "demonstrating at least one of the two selected use cases." You scope two in the Statement of Work and demonstrate at least one physically. That is a meaningful reduction in demonstration burden and it should shape which two you select: pick a pair where one is demonstrable at your site and the other is architecturally addressed.
Phase III Dual Use
The advanced private 5G and FutureG networking platform presents significant dual-use opportunities by leveraging a single, open-source technology stack to serve both commercial venues and critical DoW operational needs.
For commercial interests, this technology offers a cost-effective and vendor-neutral alternative to proprietary systems in stadiums, arenas, and convention centers. It aims to enhance fan experiences with high-density connectivity, support broadcast backhauling, and improve venue operations through integrated sensing for crowd analytics and situational awareness.
For the Department, the same platform provides resilient, secure, and high-density wireless coverage essential for DoW-affiliated event venues, training facilities, and large-scale installation assembly spaces. The platform's capability for Integrated Sensing and Communication can be adapted for enhanced situational awareness, asset tracking, and security monitoring in dynamic military environments, ensuring that advancements in commercial wireless technology directly bolster defense capabilities.
The defense case is worth developing beyond the stated text. Large-scale installation assembly spaces, base-wide event venues, training facilities, and deployed camp environments all involve dense transient populations in fixed footprints, which is structurally the same problem as a stadium. The Contested Logistics Technologies Critical Technology Area designation also points at the logistics and asset-tracking angle in Use Case 4 more than at fan connectivity, so if you want the defense narrative to land, Use Case 4 and the public-safety priority communications of Use Case 3 are the stronger hooks.
Funding, Cost Structure, and FutureG Mechanics
The award
Direct to Phase II proposals must not exceed a cost of $2,153,927 and a duration of 18 months.
Be realistic about scope. A full open-source stack integration, RTEC interoperability testing, OCUDU code enhancement with upstream contribution, a venue-scale RF design, an ISAC or positioning capability if you select Use Case 2, an emulation benchmark harness, a security architecture with a live demonstration, and a physical MVP at a venue-representative site, in 18 months for $2.15 million, is a full program. Existing OCUDU experience, an existing venue or test bed relationship, and existing RTEC engagement are worth more here than headcount.
Cost volume
A detailed Phase II Cost Volume must be submitted online in the proper format shown in the Cost Breakdown Guidance in the DoW 2026 SBIR BAA. Some items in the cost volume template may not apply, and there is no need to provide information for every item. Provide enough information to allow evaluators to assess your plans to use the requested funds.
Justify items of equipment to be purchased, including Government Furnished Equipment. All requirements for government furnished equipment or other assets, and associated costs, must be determined and agreed to during Phase II contract negotiations. Radio units, user equipment, CBRS Spectrum Access System service, channel emulators, and load generators all belong in the cost discussion.
Percentage of Work, with no exceptions
Review the updated Percentage of Work calculation details included in the DoW SBIR Program BAA. The FutureG Office will not accept any deviation to the POW requirements.
This is a serious constraint on this topic. The natural team includes a radio unit vendor, a systems integrator, a venue partner, possibly an RTEC, and possibly a university. Model your POW before you assemble it, because a plan that pushes too much work outside your firm cannot be negotiated back into compliance.
Technical and Business Assistance
Phase II awardees may request up to $50,000 per Phase II project. TABA funding is in addition to the Phase II cost ceiling and is not subject to profit or fee.
All requests for TABA must be completed using the SBIR/STTR TABA Request Form, and the completed form must be included in Volume 5 of the proposal submission in DSIP. OSW will not accept requests for TABA that do not utilize the form or that are not included as a submission document in Volume 5.
For this topic the strongest uses are commercial go-to-market development, since the topic is explicitly asking for an as-a-service offer at a price point, and spectrum and regulatory support for the CBRS deployment.
The 20-page structure
Volume 2 is 20 pages maximum: Part 1, Phase I Justification, 5 pages maximum, and Part 2, Phase II Technical Proposal, 15 pages maximum. Within the 15 pages, the Technology Transition and Commercialization Strategy is not to exceed 2 pages and counts toward the limit.
So 5 pages of feasibility, 13 pages of technical proposal, 2 pages of commercialization. Against that you must fit: the four research questions, two use cases, the RedCap discussion, the hybrid localization architecture if you select Use Case 2, the performance-class mapping, three common work package tasks, the security architecture, the RF design approach, the milestone plan, key personnel, facilities, and consultants. Plan the page allocation before drafting, because this is one of the tightest content-to-page ratios in the cycle.
Note that the FutureG instructions do not state that figures, tables, charts, and references count inside the page limit, and do not prohibit appendices. They defer to the DoW SBIR Program BAA formatting requirements. Read that BAA rather than assuming another component's stricter rule applies.
What the technical proposal must contain
The Phase II Technical Objectives and Approach section must list specific technical objectives and provide a detailed technical approach, and must include these named subsections.
Phase II Work Plan, with an explicit, detailed description of the approach, indicating what is planned, how and where the work will be carried out, a schedule of major events, and the final product to be developed.
Related Work, describing significant activities directly related to the effort including those of the Principal Investigator, the firm, consultants, or others, and demonstrating awareness of the state of the art.
Relationship with Future Research or Research and Development, stating anticipated results and the significance of the Phase II effort as a foundation for Phase III.
Technology Transition and Commercialization Strategy, at no more than 2 pages counting toward the 15-page limit, addressing five specific questions: what is the first product this technology will go into; who will be your customers and what is your estimate of the market size; how much funding will you need to bring the technology to market and how will you raise those funds; does your company contain marketing expertise and if not how do you intend to bring it in; and who are your competitors and what is your price or quality advantage.
Key Personnel, including the Principal Investigator, with directly related education, experience, and relevant publications, and a concise resume of the PI.
Facilities and Equipment, describing available instrumentation and physical facilities, justifying equipment purchases including Government Furnished Equipment, and stating whether facilities meet federal, state, and local environmental laws across the named groupings.
Consultants, describing in detail any involvement of universities, academic institutions, or other consultants and identifying them in the Cost Volume.
Answer the five commercialization questions as five distinct answers. They are enumerated and a reviewer will look for each.
The Company Commercialization Report is evaluated
Completion of the CCR as Volume 4 is required. The information contained in the CCR will be considered during proposal evaluations.
FutureG states this in both its Phase I and Direct to Phase II sections, consistently. Complete it fully. Note that it is separate from the commercialization strategy in Volume 2: the CCR covers what you have done with past Phase II awards, the strategy covers how you propose to commercialize this research.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW solicitation.
Proposing firms will be notified of selection or non-selection status within 90 days of the closing date of the topic via DSIP. Note that the FutureG text says "for a Phase I award," which appears to be residual language given that this topic issues no Phase I award. The notification will be sent to the individual listed as the Corporate Official on the proposal cover sheet, so make sure that is someone who will act on it.
Ninety days from October 21, 2026 is approximately January 19, 2027.
Refer to the DoW solicitation for procedures to protest the announcement. Protests after award should be submitted, as prescribed in FAR 33.106(b) and FAR 52.233-3, to osd.ncr.ousd-r-e.mbx.SBIR-STTR-Protest@mail.mil.
Questions
Specific questions pertaining to the administration of the FutureG SBIR Program and these proposal preparation instructions should be directed to the OUSW(R&E) FutureG Office at OSDRE-FutureG@groups.mail.mil.
The FutureG instructions do not state that DSIP Topic Q&A is unavailable, so the standard DoW process applies and Topic Q&A closes two weeks before the topic closes, on October 7, 2026. Use it, and use it on the missing KPM tables above all.
The Reference
One, and it is a link to an industry article: the rise of smart stadiums, at cellnex.com.
That is the entire cited reference list, which is remarkable for a topic of this technical density. The substantive citations are embedded in the topic text rather than in the reference section, and they are the ones that matter: 3GPP Release 19 TR 22.837 for ISAC use cases, 3GPP TS 22.137 for ISAC performance metrics, 3GPP Release 16 and 17 NR positioning methods, 3GPP Release 17 RedCap and Release 18 eRedCap, 3GPP TS 33.501 for security, O-RAN WG11 specifications for open interface protection, NIST SP 800-207 for zero trust, and the O-RAN 7.2x split.
Those eight are your real reading list. The Related Work section is required to demonstrate awareness of the state of the art, and with only a marketing article in the formal reference list you carry that burden entirely. Bring the OCUDU project documentation, the OCUDU Testing and Validation Plan the topic refers to, the SD-Core and SD-RAN and OSC project documentation, and the published RTEC test results the topic tells you to leverage.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
DSIP Topic Q&A closes: October 7, 2026, two weeks before the topic closes, per the DoW SBIR Program BAA
Proposal deadline: October 21, 2026
Selection notification: within 90 days of the closing date, approximately January 19, 2027
Period of performance: 18 months from award
A working backward plan
Before September 23. Audit the funding provenance of every feasibility result you intend to cite, because work based upon or logically extending from prior or ongoing federally funded SBIR or STTR work is excluded and failing the feasibility bar means the proposal is not evaluated. Resolve intellectual property ownership or license rights. Secure your venue or full-scale test bed relationship in writing, since the MVP demonstration requires physical radio units and user equipment at a venue-representative site. Engage the RTEC community and identify which validated reference configurations already exist for your intended RU, Core, RIC, and SMO combination, since the baseline shall incorporate available RTEC results rather than duplicate them. Get familiar with the OCUDU contribution process and the OCUDU Test and Evaluation Working Group cadence, since Monthly Status Reports go to that group. Choose your two use cases and be able to say which your reference architecture and pilot deployment validate. Plan your CBRS spectrum approach. Read the eight embedded standards references. Model your Percentage of Work before assembling a team that includes an RU vendor, an integrator, a venue, and possibly a university. Confirm SAM registration and CMMC Level 2 self-assessment in SPRS.
September 23 through October 5. Draft the 5-page Phase I justification using the four accepted evidence categories, leaning on private, academic, and non-SBIR federally funded work. Draft the 13-page technical proposal covering the four research questions, your two use cases, RedCap handling, the hybrid localization architecture and Tier 1 and Tier 2 scoping if you select Use Case 2, the required performance-class mapping, the three common work package tasks, the security architecture with your chosen live demonstration, the RF design approach, and the milestone plan with your assumed Month 14 and Month 16 sequence. Draft your own KPM set with threshold and objective values and justify each. Draft the 2-page commercialization strategy answering all five enumerated questions, using the government's own "several hundred U.S. facilities" figure and the four buying metrics. Draft the 3,000 character cover sheet abstract and the 3,000 character anticipated benefits and commercial applications discussion.
October 6 through October 7. Submit questions through DSIP Topic Q&A before it closes. The essential ones are the missing KPM tables and General System and Architectural Requirements section, and the Month 14 versus Month 16 milestone ordering. Send administrative questions to OSDRE-FutureG@groups.mail.mil.
October 8 through October 14. Build the cost volume online following the Cost Breakdown Guidance in the DoW 2026 SBIR BAA, against the $2,153,927 and 18-month ceiling. Price radio units across multiple vendors, user equipment, hardware accelerators, CBRS Spectrum Access System service, channel emulation and load generation for the Task A baseline, RTEC testing engagement, venue site access and installation, ISAC or sensing hardware if applicable, OCUDU development labor including upstream contribution effort, security architecture work, and the recurring OCUDU T&E Working Group reporting. Identify Government Furnished Equipment needs. Complete the SBIR/STTR TABA Request Form and place it in Volume 5.
October 15 through October 18. Complete Volume 4, the Company Commercialization Report, carefully, since FutureG states it is considered during evaluations. Assemble Volume 5 with the TABA form and any letters from venue partners or RTECs that substantiate specific claims. Complete Volume 6 training and the Volume 7 foreign affiliations webform, remembering it must be the webform and will not be accepted as a PDF in Volume 5, and that no previous versions should be uploaded there. Run compliance: 5 plus 15 pages with the 2-page commercialization strategy inside the 15, no proprietary or classified information on the cover sheet, 3,000 character limits per cover sheet section.
October 19 through October 20. Submit and certify in DSIP.
Frequently Asked Questions
What is OSW-FutureG SBIR topic OSW26BZ06-DV031?
OSW26BZ06-DV031 is a Direct to Phase II SBIR topic titled "Smart Venues and Stadiums," released under the OSW FutureG Office FY26 SBIR Broad Agency Announcement, Release 6. The objective is to design, validate, and demonstrate a secure, high-density private 5G network platform for stadiums and large venues, integrating open-source OCUDU, SD-Core, SD-RAN, and OSC components, delivering resilient multi-service communication alongside a 6G-ready Integrated Sensing and Communication architecture, validated at a representative venue site.
How much funding is available?
Direct to Phase II proposals must not exceed a cost of $2,153,927 and a duration of 18 months. Phase II awardees may also request up to $50,000 in Technical and Business Assistance, in addition to the cost ceiling and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5.
When is the proposal deadline?
The topic opens September 23, 2026 and proposals are due October 21, 2026 through the Defense SBIR/STTR Innovation Portal at dodsbirsttr.mil.
Can I submit a Phase I proposal?
No. This topic is accepting Direct to Phase II proposals only, and a formal Phase I award will not be issued.
What software stack is required?
OCUDU for the Centralized Unit and Distributed Unit, SD-Core for the 5G core, SD-RAN for the Near-Real-Time RIC, and the OSC stack for the Non-Real-Time RIC and SMO, integrated into a single fully open-source reference architecture with no proprietary core, RIC, or SMO component anywhere in the stack.
What feasibility evidence does the topic accept?
One or more of four categories: test data and metrics from previous lab environments or early field trials; technical reports and white papers explaining previous research, system designs, or software integration efforts; prototype designs and simulation models including diagrams, architectural blueprints, or simulation results; and previous project outcomes including success criteria, milestone reports, or commercialization results from prior private, academic, or non-SBIR federally funded work. The government states that proposers do not need to have a perfect, finished product.
Can my feasibility evidence come from a prior SBIR award?
No. The FutureG Direct to Phase II guidelines state that feasibility documentation cannot be based upon or logically extend from any prior or ongoing federally funded SBIR or STTR work, and that the work must have been substantially performed by the proposer or the Principal Investigator. The Volume 2 instruction uses the weaker phrasing "must not be solely based on"; plan against the stricter reading. Note that the topic's own fourth evidence category names private, academic, or non-SBIR federally funded work, which is consistent with the restriction.
Where are the KPM tables the topic refers to?
They do not appear in the published document. The topic refers to KPMs in "the table below," to "Section 3.0" thresholds, and to a "General System and Architectural Requirements" section, none of which are present. This is the most important thing to raise through DSIP Topic Q&A before it closes on October 7. In the meantime, propose your own justified KPM set with threshold and objective values, state why you are doing so, and note that final targets are agreed at kickoff and confirmed at the Critical Design Review.
What KPMs does the topic name in its text?
Concurrent-device density, sensing detection accuracy, latency, jitter, and service availability, given as examples. It also provides one worked example: the platform must sustain a session success rate greater than 99.5 percent for fan-facing connections at a concurrent-device density of 3,000 devices per acre in the lower bowl. The four buying metrics the segment uses are also natural KPMs: dropped-call rate at kickoff, point-of-sale uptime, camera uptime, and incident-response time.
How many use cases must I address?
At least two of the four, in your Phase II Statement of Work, and you must identify which your reference architecture and pilot deployment are designed to validate. Note that the deliverables require demonstrating at least one of the two selected use cases, so you scope two and physically demonstrate at least one.
What are the four use cases?
High-density fan connectivity and broadcast and production backhaul. ISAC-enabled crowd analytics and situational awareness, 6G-ready. Public safety and first responder priority communications. And venue operations, logistics, and outdoor perimeter coverage.
Do I have to build true RF sensing for the crowd analytics use case?
Not entirely. Proposers shall employ a hybrid architecture where standardized network-native UE positioning using 3GPP Release 16 and 17 NR positioning provides the primary near-term source of crowd density, flow, and bottleneck analytics, with RF sensing applied to non-cooperative targets carrying no connected device. The topic states that crowd-analytics KPMs may be satisfied via UE positioning in the Phase II demonstration.
What is the tiered sensing scope?
Tier 1, for the Phase II demonstration, is crowd density and flow analytics from network-native UE positioning, aggregated and anonymized, plus at least one non-cooperative RF-sensing bridge function achievable at 5G-Advanced maturity and deployed bandwidths, such as unauthorized drone incursion detection using Doppler and motion signatures, demonstrated in a laboratory or limited field configuration. Tier 2 is roadmap only: unattended-object detection and fine-grained tracking of individuals in dense crowds, limited by achievable sensing resolution of approximately c over 2B at deployable bandwidths, to be addressed in the documented 6G and ISAC evolution path rather than demonstrated. Custom or modified waveforms are not required for Phase II.
What is the mandatory common work package?
Three tasks, required in the Phase II Statement of Work regardless of use cases selected. Task A, Baseline Benchmark, establishing a quantified performance baseline of the integrated stack, with emulated end-to-end configuration acceptable and encouraged, incorporating existing RTEC results where available, presented to the OCUDU Test and Evaluation Working Group. Task B, Code and Feature Enhancement, closing the gaps between baseline and threshold and objective values, with all OCUDU modifications contributed upstream. Task C, Benchmark and Regression Harness, delivering the emulation-based benchmark suite as a repeatable, documented, open-source harness for adoption by the OCUDU community and RTECs.
Do I have to contribute my code upstream?
Yes. All modifications to OCUDU shall be contributed upstream through the project's standard contribution and review process, and enhancements that cannot be upstreamed shall be documented with rationale. Progress and upstream contribution status are reported in Monthly Status Reports to the OCUDU T&E Working Group, and an itemized Upstream Contribution Log is a deliverable. This means your commercial differentiation cannot be the OCUDU code itself.
What are the security requirements?
A documented security architecture covering 3GPP security per TS 33.501 including mutual authentication and air-interface encryption and integrity protection; protection of the O-RAN open interfaces, meaning open fronthaul, E2, A1, and O1, per O-RAN WG11 specifications; zero-trust principles per NIST SP 800-207 including least-privilege access and separation of management and user traffic; and CVE monitoring for OCUDU and its dependencies with timely upstream patching. The architecture is documented at CDR, validated in RTEC testing, and the MVP demonstration must include at least one security capability shown live.
What are the Phase II milestones?
Month 1 kickoff and Technical Interchange Meeting, Monthly Status Reports throughout, Month 12 Critical Design Review, Month 16 prototype demonstration, Month 14 final design review and demonstration and assessment, and Month 18 Final Phase II Report. Note that the published list places Month 16 before Month 14, which cannot be the intended order and appears as the same inversion in the companion topic DV032. Confirm the sequence through DSIP Topic Q&A and state your assumption in your work plan.
Where must the demonstration happen?
At the proposer's site, ideally an operating or representative venue or stadium environment, or a full-scale representative test bed where live-venue access is not feasible during Phase II. The MVP demonstration, including physical radio units and user equipment, must occur at the venue-representative site, and the deliverables require documenting the site's fidelity to venue and stadium conditions.
Is emulation acceptable?
For the Task A baseline, yes, and it is encouraged, provided the methodology, tooling, configurations, and results are fully documented and reproducible. But over-the-air performance with physical radio units and user equipment at the RTEC or representative demonstration site remains the standard of evidence for final KPM achievement.
What spectrum does the topic assume?
CBRS. The economics question asks how the deployment should be packaged as a CBRS-based, as-a-service offer within the price point venue operators require. Unlike the companion manufacturing topic, this one does not state a target price point, so you must establish it from the DAS cost comparison the topic provides.
What does the topic say about the market size?
An estimated 2,000 to 5,000 major stadiums and arenas across professional, collegiate, and municipal facilities, scaling to over 15,000 total venues including larger high school stadiums, plus well over 400 U.S. convention and exhibition centers. But the topic narrows the serviceable addressable base to several hundred U.S. facilities before counting mid-sized arenas, amphitheaters, and secondary convention space. Use the narrower figure in your commercialization strategy.
What is the required performance-class mapping?
The Technical Volume shall include a mapping of each selected use case, and its associated KPMs, to the generic network performance class or classes it exercises, and shall identify which anticipated OCUDU code or feature enhancements correspond to each class. For this topic the dominant performance classes are extreme user density, ultra-secure operation, and integrated sensing. This is a required element stated with "shall" and it is easy to omit.
How do I handle RedCap devices?
Address how the platform serves reduced-capability device classes including 3GPP Release 17 RedCap and Release 18 eRedCap, and identify any OCUDU scheduler or feature enhancements required, which are strongly encouraged as upstream contributions. Where RedCap-certified devices are not commercially available for a given endpoint type at demonstration time, you may demonstrate with available device classes or emulated RedCap UE profiles, and shall document the RedCap migration path.
What CMMC level applies?
The projected requirement for this topic is CMMC Level 2 with self-assessment.
Is this topic ITAR restricted?
No topic-level ITAR or EAR restriction paragraph appears on OSW26BZ06-DV031, and none appears on any of the three topics in this FutureG release.
How long can my technical volume be?
Twenty pages maximum, divided into Part 1 Phase I Justification at 5 pages maximum and Part 2 Phase II Technical Proposal at 15 pages maximum, with the Technology Transition and Commercialization Strategy at no more than 2 pages counting toward the 15. The FutureG instructions do not state that figures, tables, charts, and references count inside the limit or prohibit appendices, deferring instead to the DoW SBIR Program BAA formatting requirements.
What must the commercialization strategy address?
Five specific questions in no more than 2 pages. What is the first product this technology will go into. Who will be your customers and what is your estimate of the market size. How much funding will you need to bring the technology to market and how will you raise those funds. Does your company contain marketing expertise and if not how do you intend to bring it in. Who are your competitors and what is your price or quality advantage.
Is the Company Commercialization Report evaluated?
Yes. FutureG states in both its Phase I and Direct to Phase II sections that information contained in the CCR will be considered during proposal evaluations. It is separate from the commercialization strategy in Volume 2: the CCR covers what you have done with past Phase II awards, the strategy covers how you propose to commercialize this research.
Are there Percentage of Work restrictions?
Yes. The FutureG Office will not accept any deviation to the Percentage of Work requirements described in the DoW SBIR Program BAA. This is a real constraint given that the natural team includes a radio unit vendor, an integrator, a venue partner, possibly an RTEC, and possibly a university. Model your POW before assembling it.
When will I hear back, and who gets notified?
Within 90 days of the closing date of the topic, approximately January 19, 2027, via DSIP. The notification goes to the individual listed as the Corporate Official on the proposal cover sheet. Note that the FutureG text says notification "for a Phase I award," which appears to be residual language given that this topic issues no Phase I award.
Who do I contact with questions?
Technical questions about the topic go through DSIP Topic Q&A, which closes October 7, 2026. Administrative questions about the FutureG SBIR Program and these proposal preparation instructions go to the OUSW(R&E) FutureG Office at OSDRE-FutureG@groups.mail.mil.
Positioning Advice for Companies Considering This Topic
Audit your feasibility provenance before anything else. Feasibility documentation cannot be based upon or logically extend from any prior or ongoing federally funded SBIR or STTR work, and a proposal that fails the feasibility bar is not evaluated at all. Open-source 5G integration work in the United States has been heavily SBIR funded, so this is a real risk for the most qualified bidders. The topic's own fourth evidence category tells you where to look: private, academic, or non-SBIR federally funded work.
Ask about the missing KPM tables, then proceed anyway. The topic points at tables and a Section 3.0 that are not in the published document. Raise it in Topic Q&A before October 7, and in the proposal propose your own justified KPM set with threshold and objective values, saying plainly why. Noting the gap and handling it professionally is a strength, not a complaint.
Secure the venue relationship first. The MVP demonstration requires physical radio units and user equipment at a venue-representative site, and the deliverables require documenting that site's fidelity to venue conditions. A signed venue partner or a described full-scale test bed is the single most valuable thing on page one. Nothing technical compensates for not having somewhere to demonstrate.
Answer the trust question as a document, not a test plan. The fourth research question asks what evidence package best converts community-maintained, no-license-fee software into a procurement-ready proof point for a buyer with no in-house cellular expertise. That is a deliverable a venue general manager reads and acts on. Designing it, and showing a mockup or a table of contents, differentiates you from proposals that answer with a testing matrix.
Use the buyer's four metrics. Dropped-call rate at kickoff, point-of-sale uptime, camera uptime, incident-response time. The topic says this segment buys on those, not on platform standardization. Building your KPMs and your value proposition on those four speaks the customer's language and the government's at the same time.
Pick your two use cases on cost, then justify on mission. Use Cases 1 and 3 share the indoor RF design and differ mainly in quality-of-service treatment, which makes them the most economical pair and the one closest to the named buying metrics. If you address both, show priority and preemption for public safety working under peak fan load, because that interaction is the real test.
Respect the Tier 1 and Tier 2 sensing boundary. The topic tells you that fine-grained tracking of individuals in dense crowds is limited by c over 2B at deployable bandwidths and puts it in the roadmap, not the demonstration. A proposal promising Tier 2 in Phase II reads as not having done the physics. Use the UE-positioning allowance for crowd analytics and pick one tractable non-cooperative sensing function, drone detection being the topic's own example, in a laboratory or limited field setting.
Do the performance-class mapping. It is a "shall" requirement on the Technical Volume: map each use case and its KPMs to the performance classes it exercises and identify the OCUDU code or feature enhancements corresponding to each class. It is also the element most likely to get squeezed out by venue-specific content. Reserve space for it.
Budget the common work package as a separate line. Tasks A, B, and C are required regardless of use case, the baseline benchmark report is due by Month 3, and the harness is an open-source deliverable with documentation sufficient for independent execution. Proposals that fold this into general engineering will underprice it.
Plan for upstream review you do not control. All OCUDU modifications go upstream through the project's process, the deliverable is a log including review status, and enhancements that cannot be upstreamed need documented rationale. That is schedule risk owned by someone else. Say how you sequence contributions and what happens to a KPM claim if a patch is still in review at Month 18.
State how you make money when the code is public. Your differentiation is integration, RF engineering, the evidence package, the as-a-service model, and support, not the OCUDU code. A commercialization strategy that does not confront this looks naive to a reviewer who wrote the upstream requirement.
Use the government's own market number. Several hundred U.S. facilities is the serviceable addressable base the topic states, after listing 15,000 venues. Quoting the big number when the topic already narrowed it is an easy weakness to spot. The credible move is to use the narrow number and show that the unit economics work at that scale.
Engage RTECs before you propose. The baseline shall incorporate available RTEC results rather than duplicate them, RTEC-executed interoperability testing is referenced throughout, and the security architecture is validated in RTEC testing. Knowing which validated reference configurations already exist for your RU, Core, RIC, and SMO combination saves you money and shows the reviewer you are inside the community.
Plan the page budget before drafting. Five, thirteen, and two pages against four research questions, two use cases, RedCap, hybrid localization, performance-class mapping, three work package tasks, a security architecture, an RF design, and a milestone plan. Decide the allocation first or the technical proposal will lose whichever section you write last.
Make the Corporate Official someone who watches email. FutureG sends the selection notification to the Corporate Official on the cover sheet and does not state that the Principal Investigator is copied.
OSW-FutureG SBIR OSW26BZ06-NV028: 5G Signature Tracking Mitigation via Cyber Deception
Deadline: October 21, 2026
Funding Award Size: $2m
Description: Complete guide to OSW-FutureG SBIR Phase I topic OSW26BZ06-NV028, 5G signature tracking mitigation via cyber deception. Up to $323,090 over 6 months. Closes October 21, 2026.
Quick Answer
OSW26BZ06-NV028 is a Phase I SBIR topic under the OSW FutureG Office, FY26 SBIR Broad Agency Announcement, Release 6. When military and government personnel use commercial cellular networks abroad, the metadata their phones generate is enough for an adversary to map their routines, spot when something unusual is happening, and locate sensitive facilities. This topic funds a system that generates convincing fake cellular signatures to bury the real ones. The award is up to $323,090 over 6 months, with a 20-page technical volume. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The threat model is stated concisely and it is worth reading twice. Accessing foreign mobile phone networks exposes individuals and organizations to surveillance and espionage threats due to their digital exhaust, meaning the metadata required to operate on a cellular network such as credentials, location data, and device identifiers. Adversaries can leverage this data to track daily routines, identify deviations indicating special events such as VIP visits or mission preparation, and map sensitive facilities based on cell phone concentration or periods of inactivity.
That last item is the subtle one. A facility can be identified not only by phones being present but by their absence, and a sudden change in the pattern is itself the signal. Any deception system has to be believable in both directions.
Phase I is a design study, not a build. Six months produces initial application concepts, algorithms, and a comprehensive design document outlining the software architecture, plus a detailed plan for a Phase II prototype effort aiming at Technology Readiness Level 7.
Topic At a Glance
Topic number: OSW26BZ06-NV028
Title: 5G Signature Tracking Mitigation via Cyber Deception
Agency: Office of the Secretary of War, FutureG Office, under OUSW(R&E)
Solicitation: OSW FutureG Office, FY26 SBIR Broad Agency Announcement, Release 6, Proposal Submission Instructions
Program type: Phase I
Base award: must not exceed $323,090
Base period of performance: 6 months
Technical volume limit: not to exceed 20 pages. Pages over 20 will not be considered in proposal evaluations
OUSW (R&E) Critical Technology Area: Contested Logistics Technologies (LOG)
Component Technology Priority Areas: FutureG, Emerging Threat Reduction, Sustainment and Logistics
Projected CMMC level requirement: Level 2 (Self)
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic, and none appears on any of the three topics in this release
Phase II target: Technology Readiness Level 7, with follow-on Phase II funding of $2,153,927 over 18 months
Phase II form factors: hardened decoy boxes, for example Wi-Fi router-sized team devices and puck-sized personal devices
Technical and Business Assistance: Phase I up to $6,500, Phase II up to $50,000 per project, in addition to the cost ceilings and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5
Percentage of Work: the FutureG Office will not accept any deviation to the POW requirements
Company Commercialization Report: information contained in the CCR will be considered by the FutureG Office during proposal evaluations
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026
Submission portal: DSIP at dodsbirsttr.mil
Keywords: digital signature management, digital exhaust, operating on untrusted networks
The Objective and the Threat
The objective
The objective of this effort is to develop an advanced digital signature management and deception system that obfuscates cellular metadata, described as digital exhaust, to protect personnel, small units, and critical facilities from adversarial surveillance, tracking, and pattern-of-life analysis on third-party commercial networks.
Three protected entities are named, and they are different problems. An individual needs a personal device small enough to carry. A small unit needs coordinated signatures that do not contradict each other. A critical facility needs a persistent baseline of plausible activity that does not go quiet when the real population leaves.
The problem
As military and government personnel increasingly rely on commercial mobile networks, the management of digital signatures has become a critical operational concern.
Accessing foreign mobile phone networks exposes individuals and organizations to surveillance and espionage threats due to their digital exhaust, the metadata required to operate on a cellular network, for example credentials, location data, and device IDs.
Adversaries can leverage this data to track daily routines, identify deviations indicating special events, for example VIP visits or mission preparation, and map sensitive facilities based on cell phone concentration or periods of inactivity.
What that implies technically
The phrase "third-party commercial networks" is doing a lot of work. You do not control the network. The metadata that leaks is the metadata the network requires to function, which means you cannot simply suppress it and still have a working phone. The only options are to change what the metadata says, to add plausible metadata that dilutes the real signal, or both. The topic's answer is deception rather than suppression, and it names the two mechanics in the Phase I deliverable list: AI-persona generation and identity-swapping.
Note also that the adversary in this model is doing automated analysis at population scale, not manual surveillance of one target. That is why realism against machine learning matters more than realism to a human analyst, and the performance metrics reflect it.
Key Performance Metrics, Which You Must Propose Yourself
Proposers must define specific, relevant, measurable, and quantifiable key performance metrics for their proposed solutions. The topic offers suggested metrics with threshold and objective values, explicitly stated as including but not limited to the following four.
Persona capacity
The metric is AI-powered personas broadcast simultaneously per decoy device. The suggested threshold, meaning minimum success, is 2 personas. The suggested objective, meaning desired success, is 4 or more personas.
Two personas per device is the floor, which tells you the intended ratio: a decoy is not a one-to-one substitute for a phone, it is a multiplier. Four or more per device is where a small team's footprint becomes genuinely ambiguous.
Operational endurance
The metric is battery life for highly mobile and pluggable decoy form factors. The suggested threshold is 12 hours. The suggested objective is 24 to 36 hours.
Twelve hours is a working day. Twenty-four to thirty-six hours is an operational cycle without recharge. Note the phrase "highly mobile and pluggable," which suggests two power models, a carried battery-powered unit and a unit that plugs into available power, and the endurance metric applies to the mobile case.
System weight
The metric is the weight of the personal decoy device. The suggested threshold is under 2 pounds. The suggested objective is approximately 1 pound.
Two pounds for a carried radio with multiple simultaneous cellular personas and 12 hours of battery is a demanding target, and one pound is aggressive. The Phase II description calls the personal form factor puck-sized, which is the physical envelope to design against.
Obfuscation realism
The metric is the ability of AI-generated traffic to mimic realistic patterns of life to defeat automated network analysis. The suggested threshold is defeating basic heuristic analysis. The suggested objective is defeating advanced machine-learning-based behavioral tracking.
This is the metric that decides whether the system works, and it is the hardest to quantify. The threshold and objective are stated qualitatively, which means the burden is on you to make them measurable. That is exactly what the topic asks for when it says proposers must define specific, relevant, measurable, and quantifiable metrics.
A credible answer proposes an adversary model and an evaluation method. What classifier are you testing against, what features does it use, what does its detection rate look like against your generated traffic versus real traffic, and how do you avoid overfitting your generator to one detector. Proposing the red-team methodology is as important as proposing the generator, and most proposals will neglect it.
Why the metrics section deserves real effort
The topic says proposers must define the metrics and that the four above are suggestions. That is an invitation with a catch: a proposal that simply restates the four suggested metrics has not done what was asked. Add the metrics your architecture actually turns on. Plausible candidates include persona persistence over time, the number of distinct network attach events per hour and whether that rate is itself anomalous, geographic consistency between a persona's claimed movement and physically possible movement, cross-persona correlation, meaning whether your personas accidentally look like each other, and time to detection under sustained observation.
What Phase I Requires
During Phase I, proposers are required to present a comprehensive concept design and feasibility study for a digital exhaust deception system. Solutions should outline the proprietary software required to conduct AI-enabled mission planning and operations.
Phase I will consist of a six-month period of performance to develop the initial application concepts, algorithms, and a comprehensive design document that outlines the software architecture for future development and deployment.
Key activities
Phase kickoff, comprising a Technical Interchange Meeting and a Preliminary Design Review.
Before the end of Phase I, the performer must develop and present a detailed plan for addressing Phase II prototype development, testing, and evaluation, aiming to advance the technology from its current state to Technology Readiness Level 7.
Phase I deliverables
Kickoff and Technical Interchange Meeting slides.
Monthly Status Reports.
A Preliminary Design Document detailing AI-persona generation and identity-swapping mechanics.
A Phase II Plan.
A Final Phase I Report.
Reading the Phase I scope
This is a software architecture and algorithm effort with a formal design review, not a hardware build. Five deliverables in six months, three of which are reporting artifacts. The substantive output is the Preliminary Design Document, and the topic names its two required subjects precisely: AI-persona generation and identity-swapping mechanics.
Note "AI-enabled mission planning and operations." The system is not only a signature generator, it is a planning tool. Someone has to decide what personas to run, where, for how long, and with what pattern, and the topic says the software supports that decision. Address the operator's workflow, not just the signal generation.
Note also the phrase "advance the technology from its current state to TRL 7." TRL 7 means a prototype demonstrated in an operational environment. A six-month Phase I design study followed by an 18-month Phase II reaching TRL 7 implies you are starting well above TRL 1. Be explicit in your proposal about your actual current maturity, because the Phase II plan has to be credible against it.
Phase II, For Planning Purposes
Phase II proposals may only be submitted by Phase I awardees. For successful Phase I efforts, there may be a follow-on Phase II topic released within six months of Phase I completion, and Phase II efforts may be funded at $2,153,927 for an 18-month period of performance.
A Phase II follow-on effort will focus on prototype production, testing, and evaluation to reach a TRL 7 capability. The effort should result in the delivery of fully integrated mission management software and hardened decoy box form factors, for example Wi-Fi router-sized team devices and puck-sized personal devices.
Phase II milestones as stated
Month 1: Monthly Status Reports, and Kickoff and Technical Interchange Meeting.
Month 6: Critical Design Review.
Month 9: Prototype demonstration of single and multi-node networked operations.
Month 11: Final design review, field demonstration, for example 5G test network execution, and assessment.
Months 12 through 18: Final Phase II Report.
What the Phase II milestones tell you
Prototype demonstrations will be performed at the proposer's site or an approved DoW test range. Performers should address scalability, including the ability to run multiple linked decoy boxes from a single operations laptop.
Three things worth planning for now, in Phase I.
Two form factors, not one. A router-sized team device and a puck-sized personal device are different thermal, power, and antenna problems from the same software base. Your Phase I architecture should anticipate both.
Networked multi-node operation is a Month 9 milestone. Multiple decoy boxes linked and coordinated, run from a single operations laptop, means your persona generation has to be consistent across devices. Two decoys independently generating personas will produce contradictions an analyst or classifier can exploit. Design the coordination layer in Phase I.
A 5G test network is named as the Phase II field demonstration environment. Access to a 5G test network, whether your own, a partner's, or a DoW range, is a Phase II dependency worth identifying in your Phase I Phase II Plan.
Note that the Phase II milestone list places the final design review at Month 11, before the Months 12 through 18 reporting window, which leaves a long tail after the last technical milestone. If you are writing the Phase II Plan deliverable, propose what fills Months 12 through 17 rather than treating it as report-writing time.
Phase III Dual Use
The proposed digital signature management technology presents significant dual-use potential. While fundamentally designed to protect military assets from advanced signals intelligence, the core capability of obfuscating cellular metadata directly translates to protecting high-value corporate assets, personnel, and intellectual property from commercial espionage and tracking.
The commercial case is real but it needs specificity to be persuasive. Executive protection, corporate travel to high-risk jurisdictions, journalist and non-governmental organization safety, merger and acquisition activity where travel patterns leak deal information, and research facility protection are all identifiable buyers with the same exposure. Naming a segment and a buyer is stronger than the general claim.
There is also a legal and policy dimension worth acknowledging rather than avoiding. A device that broadcasts fabricated cellular identities operates in a regulated radio environment and interacts with third-party carrier networks. A proposal that addresses spectrum authorization, the difference between operating on a test network and on a live commercial network, and the compliance posture for commercial sale will read as more mature than one that treats the technology as purely technical. This is a legitimate place to say what you would use TABA for.
Funding, Cost Structure, and FutureG Mechanics
The award
The Phase I base amount must not exceed $323,090 and a duration of 6 months.
Note the phrasing: this is a not-to-exceed ceiling on both cost and duration, stated in the Volume 3 instruction rather than in a per-topic table. Roughly $323,000 over six months is a high burn rate, which suits a focused software architecture effort with an existing team and does not suit one that has to hire and ramp.
Cost volume
A detailed cost volume must be submitted online in the proper format shown in the Cost Breakdown Guidance in the DoW 2026 SBIR BAA. Some items in the cost volume template may not apply to the proposed project, and there is no need to provide information for every item. Provide enough information to allow evaluators to assess your plans to use the requested funds.
Justify items of equipment to be purchased, including Government Furnished Equipment. All requirements for government furnished equipment or other assets, and associated costs, must be determined and agreed to during Phase II contract negotiations. If your Phase I work needs access to cellular test equipment, network emulation, or a test network, say so.
Percentage of Work, with no exceptions
Review the updated Percentage of Work calculation details included in the DoW SBIR Program BAA. The FutureG Office will not accept any deviation to the POW requirements.
For this topic, the temptation is to subcontract the cellular protocol stack work, the machine learning generator, or the hardware form factor design. Model your POW before you assemble the team.
Technical and Business Assistance
The OSW SBIR/STTR Program will consider TABA requests in accordance with 15 U.S.C. 638(q). Phase I awardees may request up to $6,500 in TABA funding. Phase II awardees may request up to $50,000 per Phase II project. TABA funding is in addition to the Phase I and Phase II cost ceilings and is not subject to profit or fee.
All requests for TABA must be completed using the SBIR/STTR TABA Request Form. The completed form must be included in Volume 5 of the proposal submission in DSIP. OSW will not accept requests for TABA that do not utilize the SBIR/STTR TABA Request Form or that are not included as a submission document in Volume 5.
The form requirement is absolute. A TABA request made any other way is not accepted. For this topic, regulatory and spectrum counsel is the standout use, given that the commercial product transmits fabricated cellular identities.
The Company Commercialization Report is evaluated
Completion of the CCR as Volume 4 is required. Information contained in the CCR will be considered by the FutureG Office during proposal evaluations.
FutureG states this consistently in both its Phase I and Direct to Phase II instructions, which is worth noting because it is not universal across components. Your prior Phase II commercialization history is part of your score. Fill it out completely rather than treating it as a formality.
Note that the commercialization strategy in the technical volume is separate from the CCR. The strategy addresses how you propose to commercialize this research; the CCR covers what you have done to commercialize the results of past Phase II awards.
Proposal format
The technical volume is not to exceed 20 pages and must follow the formatting requirements provided in the DoW SBIR Program BAA. Any pages in the technical volume over 20 pages will not be considered in proposal evaluations. Use the Phase I Proposal Template provided in the DoW SBIR Program BAA, Appendix A.
Note what this document does not say. It does not state that figures, tables, charts, and references count inside the page limit, and it does not prohibit appendices. Some other OSW program offices impose those rules explicitly; FutureG defers to the DoW SBIR Program BAA formatting requirements instead. Read that BAA for the governing rules rather than assuming another component's restrictions apply.
The proposal cover sheet must include a brief technical abstract describing the proposed R&D project and an anticipated benefits and potential commercial applications discussion, each no more than 3,000 characters. Do not include proprietary or classified information in the cover sheet. If selected for negotiation and possible award, the technical abstract and anticipated benefits discussion may be publicly released online.
DSIP assigns a proposal number once the cover sheet is saved, and you may modify the cover sheet as needed until the topic closes.
Volumes 5 through 7
Volume 5 is for additional documentation supporting the cover sheet, technical volume, and cost volume, and it is where the TABA request form goes if you are requesting TABA.
Volume 6 is Fraud, Waste and Abuse training, which must be thoroughly reviewed once per year to proceed with proposal submission.
Volume 7 is the Disclosures of Foreign Affiliations or Relationships to Foreign Countries webform. It will not be accepted as a PDF supporting document in Volume 5, and you should not upload any previous versions of the form to Volume 5.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW solicitation.
Proposing firms will be notified of selection or non-selection status for a Phase I award within 90 days of the closing date of the topic via DSIP. This notification will be sent to the individual listed as the Corporate Official on the proposal cover sheet.
Ninety days from October 21, 2026 is approximately January 19, 2027. Note that the notification goes to the Corporate Official only, not to the Principal Investigator, so make sure the Corporate Official on the cover sheet is someone who will act on it.
Refer to the DoW solicitation for procedures to protest the announcement. Protests after award should be submitted, as prescribed in FAR 33.106(b) and FAR 52.233-3, to osd.ncr.ousd-r-e.mbx.SBIR-STTR-Protest@mail.mil.
Questions
Specific questions pertaining to the administration of the FutureG SBIR Program and these proposal preparation instructions should be directed to the OUSW(R&E) FutureG Office at OSDRE-FutureG@groups.mail.mil.
The FutureG instructions do not state that DSIP Topic Q&A is unavailable, so the standard DoW SBIR Program BAA Topic Q&A process applies. Topic Q&A closes to new questions two weeks before the topic closes, which is October 7, 2026. Use it for technical questions about the topic and the program office address for administrative ones.
The References
One, and it is a link rather than a citation: DoD Instruction 8520.02, at esd.whs.mil.
DoDI 8520.02 governs Public Key Infrastructure and Public Key enabling across the Department. Its relevance here is credentials. The topic's own definition of digital exhaust names credentials first among the metadata that leaks, and a system that swaps identities on a commercial network has to manage credentials somehow.
The single reference is a signal about the topic's character. Unlike topics that cite a research literature to point at an approach, this one gives you a policy instruction and leaves the technical approach entirely open. That means two things. Your Related Work discussion carries the full burden of demonstrating awareness of the state of the art, so bring the relevant literature yourself: cellular metadata privacy research, IMSI and subscription identifier handling in 5G including SUPI and SUCI concealment, pattern-of-life analysis methods, and generative modeling of network traffic. And you should read DoDI 8520.02 and say something concrete about how your credential handling relates to it, because a reviewer who put that reference in the topic will look for it.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
DSIP Topic Q&A closes: October 7, 2026, two weeks before the topic closes, per the DoW SBIR Program BAA
Proposal deadline: October 21, 2026
Selection notification: within 90 days of the closing date, approximately January 19, 2027
Period of performance: 6 months from award
A working backward plan
Before September 23. Define your adversary model, because obfuscation realism is the metric that decides whether the system works and it is stated only qualitatively. Decide what classifier or analysis method you are defeating and how you will measure that. Draft your own key performance metrics beyond the four suggested, since the topic requires you to define specific, measurable metrics rather than restate its suggestions. Assess your current technology readiness honestly against a Phase II that must reach TRL 7 in 18 months. Identify your cellular test environment, whether an in-house network emulator, a partner network, or a DoW range. Read DoDI 8520.02. Model your Percentage of Work before subcontracting protocol stack, machine learning, or hardware work. Confirm SAM registration and CMMC Level 2 self-assessment in SPRS. Download the DoW SBIR Program BAA Appendix A Phase I Proposal Template.
September 23 through October 5. Draft the 20-page technical volume against the DoW Appendix A template. Structure it around the threat model, the persona generation approach, the identity-swapping mechanics, the AI-enabled mission planning workflow, your proposed key performance metrics with justification, the software architecture, and the path to two Phase II form factors and networked multi-node operation. Address the legal and spectrum dimension rather than avoiding it. Draft the 3,000 character cover sheet abstract and the 3,000 character anticipated benefits and commercial applications discussion.
October 6 through October 7. Submit any technical questions through DSIP Topic Q&A before it closes, and send administrative questions to OSDRE-FutureG@groups.mail.mil.
October 8 through October 14. Build the cost volume online following the Cost Breakdown Guidance in the DoW 2026 SBIR BAA, against the $323,090 and 6-month ceiling. Price the software development, the machine learning work, the red-team evaluation effort, any network emulation or test access, and the design review and reporting deliverables. Identify any Government Furnished Equipment needs. Complete the SBIR/STTR TABA Request Form if you want the $6,500 and place it in Volume 5.
October 15 through October 18. Complete Volume 4, the Company Commercialization Report, carefully, since FutureG states it will be considered during evaluations. Assemble Volume 5. Complete Volume 6 training and the Volume 7 foreign affiliations webform, remembering it must be the webform and will not be accepted as a PDF in Volume 5, and that no previous versions should be uploaded there. Run compliance: 20 pages maximum, no proprietary or classified information on the cover sheet, 3,000 character limits on each cover sheet section.
October 19 through October 20. Submit and certify in DSIP.
Frequently Asked Questions
What is OSW-FutureG SBIR topic OSW26BZ06-NV028?
OSW26BZ06-NV028 is a Phase I SBIR topic titled "5G Signature Tracking Mitigation via Cyber Deception," released under the OSW FutureG Office FY26 SBIR Broad Agency Announcement, Release 6. The objective is to develop an advanced digital signature management and deception system that obfuscates cellular metadata, described as digital exhaust, to protect personnel, small units, and critical facilities from adversarial surveillance, tracking, and pattern-of-life analysis on third-party commercial networks.
How much funding is available?
The Phase I base amount must not exceed $323,090 with a duration of 6 months. Phase I awardees may also request up to $6,500 in Technical and Business Assistance, in addition to the cost ceiling and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5.
When is the proposal deadline?
The topic opens September 23, 2026 and proposals are due October 21, 2026 through the Defense SBIR/STTR Innovation Portal at dodsbirsttr.mil.
How long can my technical volume be?
Not to exceed 20 pages, following the formatting requirements in the DoW SBIR Program BAA. Any pages over 20 will not be considered in proposal evaluations. Use the Phase I Proposal Template in Appendix A of the DoW SBIR Program BAA.
Do figures and references count against the 20 pages?
The FutureG instructions do not say so. They defer to the DoW SBIR Program BAA formatting requirements. Some other OSW program offices explicitly require figures, tables, charts, and references to count inside the limit and evaluate no appendices; FutureG does not state that, so follow the DoW BAA rules rather than assuming another component's restriction applies.
What is digital exhaust?
The topic defines it as the metadata required to operate on a cellular network, giving credentials, location data, and device identifiers as examples. Adversaries use it to track daily routines, identify deviations indicating special events such as VIP visits or mission preparation, and map sensitive facilities based on cell phone concentration or periods of inactivity.
What are the suggested key performance metrics?
Four, stated as suggestions and not limits. Persona capacity, measured as AI-powered personas broadcast simultaneously per decoy device, with a threshold of 2 and an objective of 4 or more. Operational endurance, measured as battery life for highly mobile and pluggable decoy form factors, with a threshold of 12 hours and an objective of 24 to 36 hours. System weight for the personal decoy device, with a threshold under 2 pounds and an objective of approximately 1 pound. And obfuscation realism, the ability of AI-generated traffic to mimic realistic patterns of life to defeat automated network analysis, with a threshold of defeating basic heuristic analysis and an objective of defeating advanced machine-learning-based behavioral tracking.
Do I have to use those metrics?
You have to define your own. The topic states that proposers must define specific, relevant, measurable, and quantifiable key performance metrics and that the four listed are suggestions including but not limited to. A proposal that only restates the four suggestions has not done what was asked. Add the metrics your architecture actually turns on and justify them.
How do I make obfuscation realism measurable?
That is the central proposal problem, since the threshold and objective are stated qualitatively. A credible answer proposes an adversary model and an evaluation method: what classifier you test against, what features it uses, its detection rate against your generated traffic versus real traffic, and how you avoid overfitting the generator to one detector. Proposing the red-team methodology matters as much as proposing the generator.
What does Phase I actually deliver?
Five deliverables. Kickoff and Technical Interchange Meeting slides, Monthly Status Reports, a Preliminary Design Document detailing AI-persona generation and identity-swapping mechanics, a Phase II Plan, and a Final Phase I Report. Key activities include a phase kickoff comprising a Technical Interchange Meeting and a Preliminary Design Review.
Is Phase I a hardware build?
No. Phase I develops initial application concepts, algorithms, and a comprehensive design document outlining the software architecture for future development and deployment. The hardware form factors appear in Phase II.
What is the Phase II target?
Technology Readiness Level 7, meaning a prototype demonstrated in an operational environment. Phase II focuses on prototype production, testing, and evaluation, delivering fully integrated mission management software and hardened decoy box form factors such as Wi-Fi router-sized team devices and puck-sized personal devices.
What are the Phase II milestones?
Month 1 for Monthly Status Reports plus Kickoff and Technical Interchange Meeting. Month 6 for Critical Design Review. Month 9 for prototype demonstration of single and multi-node networked operations. Month 11 for final design review, field demonstration such as 5G test network execution, and assessment. Months 12 through 18 for the Final Phase II Report.
How much is Phase II funded at?
Phase II efforts may be funded at $2,153,927 for an 18-month period of performance. Phase II proposals may only be submitted by Phase I awardees, and for successful Phase I efforts there may be a follow-on Phase II topic released within six months of Phase I completion.
Where will the Phase II prototype be demonstrated?
At the proposer's site or an approved DoW test range. The topic names 5G test network execution as an example field demonstration, so identifying your test network access is worth doing in the Phase I Phase II Plan.
What does networked multi-node operation require?
Running multiple linked decoy boxes from a single operations laptop, which the topic lists under scalability. That means persona generation has to be coordinated across devices, since independently generated personas can produce contradictions an analyst or classifier could exploit. Design the coordination layer in Phase I.
What CMMC level applies?
The projected requirement for this topic is CMMC Level 2 with self-assessment.
Is this topic ITAR restricted?
No topic-level ITAR or EAR restriction paragraph appears on OSW26BZ06-NV028, and none appears on any of the three topics in this FutureG release.
Is the Company Commercialization Report evaluated?
Yes. FutureG states in both its Phase I and Direct to Phase II instructions that information contained in the CCR will be considered during proposal evaluations. Complete it fully rather than treating it as a formality. Note that it is separate from the commercialization strategy in the technical volume: the CCR covers what you have done with past Phase II awards, the strategy covers how you propose to commercialize this research.
Are there Percentage of Work restrictions?
Yes. The FutureG Office will not accept any deviation to the Percentage of Work requirements described in the DoW SBIR Program BAA. Model your POW before subcontracting cellular protocol stack, machine learning, or hardware form factor work.
How do I request TABA?
Using the SBIR/STTR TABA Request Form, included in Volume 5 of the DSIP submission. OSW will not accept requests that do not use the form or that are not submitted in Volume 5. Phase I is up to $6,500 and Phase II up to $50,000 per project, both in addition to the cost ceilings and not subject to profit or fee.
Can I ask questions about the topic?
Yes. The FutureG instructions do not state that DSIP Topic Q&A is unavailable, so the standard DoW process applies and Topic Q&A closes two weeks before the topic closes, on October 7, 2026. Administrative questions about the FutureG SBIR Program and these proposal instructions go to OSDRE-FutureG@groups.mail.mil.
When will I hear back, and who gets notified?
Within 90 days of the closing date of the topic, approximately January 19, 2027, via DSIP. The notification is sent to the individual listed as the Corporate Official on the proposal cover sheet. Note that unlike some components, FutureG does not state that the Principal Investigator is also notified, so make sure the Corporate Official is someone who will act on it.
What is the commercial market?
The topic identifies protection of high-value corporate assets, personnel, and intellectual property from commercial espionage and tracking. Concrete segments worth naming include executive protection, corporate travel to high-risk jurisdictions, journalist and non-governmental organization safety, merger and acquisition activity where travel patterns leak deal information, and research facility protection.
What is the single reference about?
DoD Instruction 8520.02, which governs Public Key Infrastructure and Public Key enabling. Its relevance is credentials, which the topic names first among the metadata that leaks. Read it and say something concrete about how your credential handling relates to it, because a reviewer who included that reference will look for it.
Positioning Advice for Companies Considering This Topic
Define your adversary before you define your system. Obfuscation realism is the metric that determines whether any of this works, and the topic states it qualitatively: defeat basic heuristic analysis at threshold, defeat advanced machine-learning-based behavioral tracking at objective. Name the analysis you are defeating, the features it uses, and how you measure success against it. A proposal that describes a persona generator without describing the detector it beats has skipped the hard half.
Propose your own metrics, and say why. The topic requires specific, relevant, measurable, and quantifiable metrics and offers four as suggestions. Restating those four is the minimum and it will not distinguish you. Add persona persistence, attach-event rate plausibility, geographic consistency, cross-persona correlation, and time to detection under sustained observation, with threshold and objective values you can defend.
Solve the absence problem, not just the presence problem. The threat description says adversaries map sensitive facilities based on cell phone concentration or periods of inactivity. Deception that adds plausible activity is half the answer. Maintaining a plausible baseline when the real population leaves, and avoiding a suspicious change in pattern when it returns, is the other half, and most proposals will only address the first.
Design for three protected entities, not one. An individual, a small unit, and a facility have different requirements. A puck on one person, a coordinated set of personas across a team that do not contradict each other, and a persistent facility signature are three modes of the same system. Say which you optimize for and how the others follow.
Take the mission planning software seriously. The topic asks for AI-enabled mission planning and operations, and the Phase II deliverable is fully integrated mission management software. Someone has to decide which personas to run, where, for how long, and with what pattern of life, under time pressure and without deep technical knowledge. Show that workflow.
Design the coordination layer in Phase I. Multi-node networked operation is a Month 9 Phase II milestone and running multiple linked decoy boxes from one operations laptop is a stated scalability requirement. Independently generated personas will collide statistically. Solving that in Phase I architecture rather than discovering it in Phase II integration is the difference between a plan and a hope.
Be honest about your current TRL. Phase II must reach TRL 7 in 18 months following a 6-month design study. That arithmetic only closes if you are starting from working components. State what you have, and let the Phase II Plan be credible against it rather than aspirational.
Address the legal and spectrum question head on. A device broadcasting fabricated cellular identities on or near third-party commercial networks raises authorization questions, and the commercial market raises them again. A proposal that names the compliance posture, distinguishes test network operation from live network operation, and identifies where it needs counsel reads as more mature than one that treats the problem as purely technical. It is also a good use for TABA.
Read DoDI 8520.02 and use it. It is the topic's only reference and it is about credentials, which the threat description names first. A paragraph connecting your identity and credential handling to that instruction shows you read what the Government pointed you at.
Bring your own literature. With one policy reference and no research citations, your Related Work section carries the whole burden of demonstrating awareness of the state of the art, which the DoW template requires. Cellular metadata privacy research, subscription identifier concealment in 5G, pattern-of-life analysis, and generative traffic modeling all belong there.
Fill out the Company Commercialization Report properly. FutureG says twice that it is considered during evaluations. If you have prior Phase II awards, their outcomes are part of your score here.
Model Percentage of Work first. No deviations accepted, and this topic invites subcontracting the protocol stack, the machine learning, and the hardware. Run the calculation before you commit to a team.
Make the Corporate Official someone who watches email. FutureG sends the selection notification to the Corporate Official on the cover sheet and, unlike some components, does not say the Principal Investigator is copied.
Plan for the weight and endurance targets from the start. Under 2 pounds with 12 hours of battery, objective approximately 1 pound with 24 to 36 hours, for a device broadcasting multiple simultaneous cellular personas, is a genuine radio and power engineering problem. Even in a Phase I software effort, a preliminary power and thermal budget against the puck form factor makes the Phase II plan credible.
OSW-Reliance 21 SBIR OSW26BZ06-DV027: Monolithic Graphene-CMOS Broadband UV to LWIR Focal Plane Arrays
Deadline: October 21, 2026
Funding Award Size: $2m
Description: Complete guide to OSW-Reliance 21 SBIR Direct to Phase II topic OSW26BZ06-DV027, monolithic graphene-CMOS broadband UV to LWIR focal plane arrays. $2M over 24 months. Closes October 21, 2026.
Quick Answer
OSW26BZ06-DV027 is a Direct to Phase II SBIR topic under the Office of the Secretary of War, Reliance 21, 2026 SBIR Broad Agency Announcement, Release 6. Phase I proposals will not be accepted. The Navy wants one detector that sees from 300 nanometers to 10,000 nanometers, built by putting graphene directly on a CMOS readout wafer instead of bonding a separate detector array to it. The award is $2,000,000 over 24 months, with a 20-page technical volume. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The argument for it is one sentence in the topic and it is a good one: you cannot detect something you cannot see. Today every wavelength band needs its own material system. Aluminum gallium nitride for ultraviolet, silicon-based devices for near infrared, indium gallium arsenide for short-wave infrared, mercury cadmium telluride and III-V strain-layer superlattices for mid-wave and long-wave infrared. Each is a separate supply chain, a separate cost structure, and a separate camera.
The incumbent approach also has specific, quantified limits. Indium gallium arsenide photodiode arrays for short-wave infrared are hybridized to a silicon readout integrated circuit by indium-bump flip-chip bonding. That is a mature approach but constrained by high unit cost, a limited domestic foundry base, pixel pitches that are difficult to scale below roughly 15 micrometers, array formats limited by hybridization yield, and a spectral response that typically cuts off near 1,700 nanometers.
Monolithic graphene on CMOS removes the bonding step, which is where most of that cost and yield penalty comes from. The topic is not asking you to prove the concept. Graphene-on-CMOS broadband imagers have been demonstrated in the literature, and the topic cites the 2017 Nature Photonics paper that did it. It is asking you to retire four specific risks that stand between those demonstrations and a fieldable naval imager.
Topic At a Glance
Topic number: OSW26BZ06-DV027
Title: Monolithic Graphene-CMOS Broadband (UV-LWIR) Focal Plane Arrays for target detection in low-light conditions
Agency: Office of the Secretary of War, Reliance 21, administered by the OUSW(R&E) SBIR Program
Solicitation: OSW-Reliance 21, 2026 SBIR Broad Agency Announcement, Release 6, Proposal Submission Instructions
Program type: Direct to Phase II only. Phase I proposals will not be accepted
Base award: $2,000,000
Base period of performance: 24 months
Technical volume limit: 20 pages, structured as 5 pages of Phase I justification plus 15 pages of Phase II technical proposal, with the transition and commercialization strategy at no more than 2 pages counting toward the 15
OUSW (R&E) Critical Technology Area: Quantum and Battlefield Information Dominance
Component Technology Priority Areas: Integrated Sensing and Cyber, Microelectronics, Quantum Science, Space Technology, Advanced Materials
Projected CMMC level requirement: Level 1
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic
Spectral requirement: broadband imaging from 300 nanometers to 10,000 nanometers
Architecture: wafer-scale, monolithically integrated graphene-on-CMOS image sensor
Operating condition: compact, low-power, uncooled focal plane array
Mobility goal: a substantial increase in integrated-device carrier mobility, with a goal approaching 10,000 square centimeters per volt-second
Phase II end state: a functioning broadband camera, characterized against Navy-relevant targets including passive low-light atmospheric nightglow imaging and imaging through obscurants such as fog and haze
Phase III path: a pilot qualification lot on the order of tens of wafers with defined yield and performance metrics
Technical and Business Assistance: up to $50,000 per Phase II project, in addition to the cost ceiling and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5
Cost volume: the DSIP online Cost Volume webform is required. No separate Excel template
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026
Submission portal: DSIP at dodsbirsttr.mil
Keywords: graphene, focal plane array, monolithic CMOS integration, two-dimensional materials, broadband imaging, night vision, wafer-scale manufacturing, uncooled detector, quantum-dot and metal oxide nanocrystal sensitization
The Feasibility Bar, Which Is the First Thing to Check
This topic solicits Direct to Phase II proposals only. Phase I proposals will not be accepted. Offerors must document that Phase I feasibility has already been established through prior work.
What counts as acceptable evidence
Acceptable feasibility evidence includes the fabrication and measured characterization of graphene phototransistor or graphene-on-CMOS coupon test structures exhibiting broadband ultraviolet-through-long-wave-infrared photoresponse, and demonstration of die- or wafer-scale graphene integration with quantified device yield and uniformity.
Documentation may comprise technical reports, peer-reviewed publications, measured data, and design artifacts sufficient to substantiate the scientific, technical, and commercial merit required to enter Phase II.
Two distinct capability claims are named, and the word "includes" suggests they are examples rather than an exhaustive list. The first is device physics: coupon structures with measured broadband photoresponse across the band. The second is integration engineering: die- or wafer-scale graphene integration with quantified yield and uniformity. Notice that yield and uniformity must be quantified, not described. Numbers.
What you explicitly do not need yet
Requirements definition for the naval use case, detailed focal-plane-array performance modeling, and pixel-architecture and back-end-of-line integration-flow design need not be complete at proposal. These activities are carried into the Phase II Base effort.
This is a useful and unusually explicit allowance. Three substantial engineering activities are carved out of the feasibility requirement and assigned to the funded Base task. It tells you where to spend your five pages of justification: on measured graphene photoresponse and measured integration yield, not on system modeling or pixel design you have not been paid to do yet.
The restriction that will disqualify some proposers
The OSW-Reliance 21 Direct to Phase II guidelines impose a constraint stricter than most components apply.
Feasibility documentation cannot be based upon or logically extend from any prior or ongoing federally funded SBIR or STTR work. Work submitted within the feasibility documentation must have been substantially performed by the proposer or the principal investigator. If technology in the feasibility documentation is subject to intellectual property, the proposer must either own the IP or must have obtained license rights to such technology prior to proposal submission, to enable it and its subcontractors to legally carry out the proposed work.
The Volume 2 instruction phrases it slightly differently, saying feasibility documentation must not be solely based on work performed under prior or ongoing federally funded SBIR or STTR work. The two formulations do not match, and the stricter one, "cannot be based upon or logically extend from," is the one to plan against.
This matters more on this topic than on most. Graphene photodetector and two-dimensional material integration work in the United States has been heavily federally funded, and a meaningful share of it through SBIR and STTR. Audit the provenance of every result you intend to cite. Internally funded work, privately funded work, non-SBIR government contract work, foundry-partnered development, and published academic work you performed yourself under other funding are all cleaner ground.
Note also the intellectual property requirement. Graphene transfer, encapsulation, and sensitization processes are frequently licensed from universities or partners. You must own the IP or hold license rights before you submit, and be able to show it.
And the consequence of falling short is not a low score: if the proposer fails to demonstrate technical merit and feasibility equivalent to the Phase I level as described in the topic, the related Phase II proposal will not be evaluated.
What the Navy Is Actually Buying
The objective
Develop, fabricate, and demonstrate a wafer-scale, monolithically integrated graphene-on-CMOS image sensor that provides broadband imaging from 300 nanometers to 10,000 nanometers in a compact, low-power, uncooled focal plane array.
Every word in that sentence is a requirement. Wafer-scale, not die-scale. Monolithically integrated, not hybridized. 300 to 10,000 nanometers, which spans ultraviolet through long-wave infrared. Compact, low-power, and uncooled, which rules out the cryogenic cooling that mercury cadmium telluride and superlattice detectors typically need and is a large part of the size, weight, power, and cost argument.
The operational problem
The Navy continues to field multiple seeker technologies to detect, identify, and engage targets in multi-domain operations through obscurants. However, you cannot detect something you cannot see. Each wavelength range provides unique opportunity to glean information.
Currently, for each wavelength range a distinct material system technology is required: aluminum gallium nitride for ultraviolet, analog devices for near infrared, indium gallium arsenide for short-wave infrared, mercury cadmium telluride and III-V strain-layer superlattice detectors for mid-wave and long-wave infrared.
The incumbent's specific limits
Indium gallium arsenide photodiode arrays for short-wave infrared are hybridized to a silicon readout integrated circuit by indium-bump flip-chip bonding. This is a mature approach but constrained by five things: high unit cost, a limited domestic foundry base, pixel pitches that are difficult to scale below roughly 15 micrometers, array formats limited by hybridization yield, and a spectral response that typically cuts off near 1,700 nanometers.
These constraints limit the resolution, affordability, and proliferation of imagers across size, weight, power, and cost constrained platforms.
Those five constraints are your value proposition checklist. A strong proposal states, one by one, what monolithic graphene-on-CMOS does to each: what the unit cost becomes at volume, why a CMOS foundry base is broader than a compound semiconductor one, what pixel pitch you can reach without hybridization, what array format becomes possible when yield is not bounded by bump bonding, and how far past 1,700 nanometers you actually get.
A note on the source text
The paragraph describing incumbent technology is broken in the source document. The sentence beginning "In addition, the incumbent detector technologies, for example indium gallium arsenide (InGaAs)" ends abruptly, the four risk items are inserted, and then the sentence resumes several lines later with "photodiode arrays for SWIR are hybridized to a silicon read-out integrated circuit (ROIC) by indium-bump flip-chip bonding."
Reassembled, it reads as a single statement about InGaAs photodiode arrays being flip-chip bonded to a silicon ROIC. This page presents it in that reconstructed order because that is plainly the intended meaning. Nothing about the requirement changes, but be aware that the topic text as published is out of sequence, and if you quote it, quote it carefully.
The Four Risks to Retire
This is the heart of the topic. It seeks to retire the principal risks that stand between existing demonstrations and a fieldable naval imager. Four are named, and they make a natural outline for your technical proposal.
Risk one: graphene material quality and uniformity at wafer scale
Carrier mobility, Dirac-point control, low hysteresis, and low defect density, for example Raman D-to-G intensity ratio, held tight across full wafers and lot to lot.
Four material metrics and a named measurement for the fourth. "Held tight across full wafers and lot to lot" is the operative phrase: this is a statistical process control requirement, not a best-coupon requirement. Report distributions, not champions.
Dirac-point control and low hysteresis deserve particular attention because they are what make a graphene device stable enough to calibrate. A focal plane whose pixels drift in threshold cannot hold a flat field.
Risk two: back-end-of-line integration on CMOS
Repeatable, high-yield BEOL transfer, encapsulation, patterning, and contacting of graphene on CMOS read-out wafers.
Four process steps, all constrained by the fact that they happen after the CMOS is built. BEOL means low thermal budget, no damage to underlying metal and dielectric layers, and compatibility with a foundry's process flow. Graphene grown at high temperature elsewhere and transferred is the usual answer, and transfer at wafer scale with high yield is the hardest single manufacturing problem in this topic. The topic cites Neumaier, Pindl, and Lemme on integrating graphene into semiconductor fabrication lines, which is exactly this problem.
Risk three: absorber sensitization and passivation
Absorber sensitization and passivation that deliver uniform, stable, low-noise broadband response.
Graphene alone absorbs only a few percent of incident light, so a broadband imager needs a sensitizer that absorbs and transfers charge to the graphene channel. The keywords name the candidates: quantum-dot and metal oxide nanocrystal sensitization. Uniformity, stability, and low noise are the three qualities demanded, and stability is the one that usually fails, since colloidal quantum dot films are prone to drift and degradation.
Note that getting to 10,000 nanometers is the hardest part of the spectral requirement. Quantum dot sensitization comfortably covers visible through short-wave infrared. Long-wave infrared response at 10 micrometers from an uncooled graphene device is a much more demanding claim, and it likely involves a different physical mechanism than the short-wave case. Address the long-wave end explicitly rather than presenting a single sensitization story for the whole band.
Risk four: the focal plane array itself
An imaging focal plane array meeting the noise-equivalent irradiance, dynamic range, frame rate, operability, and stability required for naval intelligence, surveillance, and reconnaissance.
Five system metrics. Note "operability," which in focal plane array practice means the fraction of pixels meeting specification, and it is where monolithic integration should beat hybridization. If your yield story is good, operability is where you show it numerically.
The stated performance target
Proposers should target a substantial increase in integrated-device carrier mobility, with a goal approaching 10,000 square centimeters per volt-second, tight pixel-to-pixel uniformity, and a manufacturable process flow with a clear path to pilot-scale production.
Note "integrated-device carrier mobility." Not mobility measured on a pristine transferred graphene film, but mobility in the finished device after transfer, encapsulation, patterning, and contacting. That is a considerably harder number, and stating your current integrated-device mobility honestly, with the measurement conditions, is more persuasive than citing a film-level figure.
Phase II Structure and Requirements
Design, fabricate, and deliver a prototype monolithic graphene-CMOS broadband and short-wave infrared focal plane array, maturing the technology from documented feasibility through a demonstrated camera. The effort is organized into a Base task that completes design and risk reduction and a Prototype task that builds and characterizes the imager.
Base task: design and risk reduction
Define imager requirements against a representative naval ISR use case.
Develop and validate focal plane array performance models covering spectral response, responsivity, noise-equivalent irradiance, dynamic range, and operability.
Complete the pixel-architecture and back-end-of-line integration-flow design for a foundry CMOS readout integrated circuit.
Fabricate and characterize coupon-level and small-array test structures to confirm material quality, integration yield, and broadband photoresponse.
Finalize the wafer-scale process definition.
Note "for a foundry CMOS ROIC." Your integration flow has to target a real foundry process, which means a foundry relationship and a process design kit. If you already have one, name it early. If you do not, getting one is a Base task dependency you do not fully control.
Prototype task: fabrication and demonstration
Mature the wafer-scale process, meaning graphene growth and transfer, encapsulation, patterning, contacting, and absorber sensitization, to demonstrate high device yield and pixel-to-pixel uniformity across full wafers.
Build and demonstrate a functioning broadband camera, meaning sensor plus read-out and minimal supporting electronics and optics.
Characterize spectral response, responsivity, noise-equivalent irradiance, dynamic range, frame rate, operability, fixed-pattern noise, and stability against Navy-relevant targets, including passive low-light atmospheric nightglow imaging and imaging through obscurants such as fog and haze.
Conduct an initial environmental and reliability assessment.
The two named demonstration scenarios
Passive low-light imaging under atmospheric nightglow, and imaging through obscurants such as fog and haze. These are not generic performance claims and they should drive your test planning and your budget.
Atmospheric nightglow is a specific and demanding illumination condition. The night sky radiates in the near infrared and short-wave infrared, notably from hydroxyl emission bands, at levels well below what a visible sensor can use. Passive imaging by nightglow is the operational argument for short-wave infrared night vision, and demonstrating it requires either a genuinely dark field site on a moonless night or a calibrated low-light chamber that reproduces the spectral distribution. Both are real costs and real schedule.
Obscurant penetration through fog and haze needs either a fog chamber or opportunistic field testing, and quantifying it requires a measured obscurant condition rather than a qualitative before-and-after image. Say how you will measure the obscurant.
Deliverables
Validated performance models and final focal plane array design.
Prototype focal plane arrays and cameras.
The documented wafer-scale process flow with measured yield and uniformity.
A full test and evaluation data package.
A Phase III transition and manufacturing plan, including a path to a pilot qualification lot on the order of tens of wafers with defined yield and performance metrics.
That last item is the most commercially meaningful. "Tens of wafers with defined yield and performance metrics" is a manufacturing qualification plan, not a research roadmap. It tells you the Navy is thinking about a production path and expects you to price and schedule one.
A terminology note
The objective and the title specify 300 to 10,000 nanometers, which is ultraviolet through long-wave infrared. The Phase II section repeatedly says "broadband and short-wave infrared," and the Phase III section refers to "broadband and short-wave infrared imaging." The Phase II characterization scenarios, nightglow and obscurant penetration, are both short-wave infrared applications.
The most reasonable reading is that the full 300 to 10,000 nanometer band is the objective and the short-wave infrared portion is the near-term naval application driving the demonstration. Plan to demonstrate broadband response, as the feasibility evidence requires, while designing the camera demonstration around the short-wave infrared use cases the topic actually names. If the discrepancy affects your architecture choices, raise it through DSIP Topic Q&A before it closes on October 7.
Phase III Dual Use
Transition the technology to naval and joint platforms that require affordable, compact, broadband and short-wave infrared imaging: unmanned aerial, surface, and undersea vehicles; handheld, weapon-mounted, and helmet-mounted night-vision and threat-warning systems; shipboard situational-awareness and navigation suites; and distributed or expendable sensor nodes. Also to hyperspectral and multispectral payloads for camouflage and decoy discrimination and littoral mine countermeasures.
Establish and qualify a domestic, CMOS-foundry-based manufacturing capability for monolithic two-dimensional material focal plane arrays.
The same platform carries broad commercial dual-use value in automotive and autonomous-vehicle vision, seeing through fog, smoke, and darkness; machine vision and semiconductor and solar wafer inspection; agricultural and food-quality sorting; medical and biometric imaging; and environmental and industrial process monitoring, supporting an economically sustainable supply base.
Two things worth pulling out. First, "establish and qualify a domestic, CMOS-foundry-based manufacturing capability" is a supply chain objective, not just a product objective. The topic named a limited domestic foundry base as one of the incumbent's five constraints, and domestic manufacturability is part of what is being bought. Say something concrete about which domestic foundry path you would use.
Second, automotive short-wave infrared vision is the largest commercial market on that list by an order of magnitude, and it is driven by exactly the cost and pixel pitch constraints the topic identifies. If your cost model closes for automotive volumes, that is the strongest possible commercialization argument, and it also happens to be the market that would fund the pilot line the Navy wants.
Funding, Cost Structure, and OSW-Reliance 21 Mechanics
The award
$2,000,000 over 24 months.
Be realistic about scope against that budget. Wafer-scale process development, a foundry CMOS run, camera integration, and a characterization campaign including low-light and obscurant testing is a lot for two million dollars over two years. Existing foundry relationships, existing test infrastructure, and existing graphene transfer capability are worth more here than headcount.
Cost volume mechanics
OSW-Reliance 21 requires the use of the DSIP online Cost Volume webform. No separate Excel template is required. If supplementary cost detail is desired, it may be uploaded as a PDF attachment within Volume 3.
A detailed Phase II cost volume must be submitted online in the proper format shown in the Cost Breakdown Guidance in the DoW 2026 SBIR BAA. Provide enough information to allow evaluators to assess your plans to use the requested funds.
Justify items of equipment to be purchased, including Government Furnished Equipment. All requirements for government furnished equipment or other assets, and associated costs, must be determined and agreed to during Phase II contract negotiations. If you need government-furnished readout wafers, test articles, or facility access, say so.
Percentage of Work, with no exceptions
Review the updated Percentage of Work calculation details included in the DoW solicitation. OSW-Reliance 21 will not accept any deviation to the POW requirements.
This is a live risk on this topic. A monolithic graphene-CMOS effort naturally involves a CMOS foundry, possibly a graphene supplier, possibly a university for materials characterization, and possibly a camera integrator. Model your Percentage of Work before you assemble that team, because a plan that pushes too much work outside your firm cannot be negotiated back into compliance.
Technical and Business Assistance
Phase II awardees may request up to $50,000 per Phase II project. TABA funding is in addition to the Phase II cost ceiling and is not subject to profit or fee.
All requests for TABA must be completed using the SBIR/STTR TABA Request Form, and the completed form must be included in Volume 5 of the proposal submission in DSIP. OSW will not accept requests for TABA that do not utilize the form or that are not included as a submission document in Volume 5.
Fifty thousand dollars is at the high end across components in this cycle. For this topic, the strongest uses are manufacturing and foundry transition consulting, given that the Phase III deliverable is a pilot qualification plan, and intellectual property counsel, since graphene transfer and sensitization processes are commonly licensed.
The 20-page structure
Volume 2 is 20 pages maximum, divided into Part 1, Phase I Justification, at 5 pages maximum, and Part 2, Phase II Technical Proposal, at 15 pages maximum. Within the 15 pages, the Technology Transition and Commercialization Strategy is not to exceed 2 pages and counts toward the limit.
All figures, tables, charts, and references must be included within the page count. Any pages past the limit will not be considered, and no separate appendices will be evaluated.
So: 5 pages of feasibility, 13 pages of technical proposal, 2 pages of commercialization. For a topic with four named risks, a two-task program, a foundry integration flow, and a characterization campaign, that is tight. Plan the page budget before you draft.
What the Phase II technical proposal must contain
The Phase II Technical Objectives and Approach section must list specific technical objectives and provide a detailed technical approach, and include these named subsections.
Phase II Work Plan, with an explicit, detailed description of the approach, indicating what is planned, how and where the work will be carried out, a schedule of major events, and the final product to be developed.
Related Work, describing significant activities directly related to the effort including those of the Principal Investigator, the firm, consultants, or others, and demonstrating awareness of the state of the art.
Relationship with Future Research or Research and Development, stating anticipated results and the significance of the Phase II effort as a foundation for Phase III.
Technology Transition and Commercialization Strategy, at no more than 2 pages counting toward the 15-page limit, addressing five specific questions: what is the first product this technology will go into; who will be your customers and what is your estimate of the market size; how much funding will you need to bring the technology to market and how will you raise those funds; does your company contain marketing expertise and if not how do you intend to bring it in; and who are your competitors and what is your price or quality advantage.
Key Personnel, including the Principal Investigator, with directly related education, experience, and relevant publications, and a concise resume of the PI.
Facilities and Equipment, describing available instrumentation and physical facilities, justifying equipment purchases including Government Furnished Equipment, and stating whether facilities meet federal, state, and local environmental laws across the named groupings.
Consultants, describing in detail any involvement of universities, academic institutions, or other consultants and identifying them in the Cost Volume.
Answer the five commercialization questions as five distinct answers. They are enumerated and a reviewer will look for each one.
The Company Commercialization Report, and a contradiction
Completion of the CCR as Volume 4 is required.
The Direct to Phase II section of this release states that the information contained in the CCR will not be considered by "SCO" during proposal evaluations. The Phase I section of the same document states that CCR information will be considered by OSW-Reliance 21 during proposal evaluations.
The two statements conflict, and the reference to SCO appears to be residual text from another organization's instructions, which weakens the DP2 statement as authority. Complete the CCR carefully and completely, and if the answer materially affects your proposal, raise it through DSIP Topic Q&A.
Note that the commercialization strategy in Volume 2 is separate from the CCR. The strategy addresses how you propose to commercialize this research; the CCR covers what you have done to commercialize the results of past Phase II awards.
Supporting documents that are optional but encouraged
Letters of Support from prospective transition stakeholders within DEVCOM C5ISR Center, PAE Maneuver Ground, PAE Maneuver Air, CPE Autonomy, the Naval Research Laboratory, or the Air Force Research Laboratory.
A Data Management Plan addressing provenance, licensing, and protection of pre-training data and government-furnished data.
For this topic the Naval Research Laboratory is the most natural fit, since the topic is written around naval seekers and naval ISR. NRL is also among the named evaluating organizations. DEVCOM C5ISR Center is relevant for the handheld and weapon-mounted night vision applications, and CPE Autonomy for the unmanned platform cases.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW solicitation.
Government technical evaluators from Army DEVCOM C5ISR Center, Army DEVCOM Army Research Lab, the Naval Research Laboratory, and the Air Force Research Laboratory may participate in the evaluation. Non-government support contractors may assist in administrative handling of proposals if the individual has signed a non-disclosure agreement, and they will not participate in selection decisions.
Proposing firms will be notified of selection or non-selection status within 90 days of the closing date of the topic, which is approximately January 19, 2027. Notifications will be issued through DSIP to both the Corporate Official and the Principal Investigator listed on the proposal.
Protests after award should be submitted, as prescribed in FAR 33.106(b) and FAR 52.233-3, to osd.ncr.ousd-r-e.mbx.SBIR-STTR-Protest@mail.mil.
Tri-service coordination and the TPOC question
This topic is of joint interest to the U.S. Army, the U.S. Navy, and the U.S. Air Force and Space Force through the organizations named above. Proposers are strongly encouraged to engage the Technical Point of Contact listed in the topic description during the pre-release period to discuss technical scope and transition opportunities across the Services.
No Technical Point of Contact appears in the DV027 topic description, or in any of the four topic descriptions in this release. Use DSIP Topic Q&A, and send administrative questions to osd.pentagon.ousd-atl.mbx.communities-of-interest@mail.mil.
Note that Phase II efforts under this release shall include a transition plan addressing at least two of the three Services. The topic is written from a Navy perspective, but its Phase III list already spans naval platforms, handheld and helmet-mounted night vision, and space-adjacent applications, so a two-service transition case is available. Write it explicitly rather than assuming it is obvious.
Classification
Classified proposals are not accepted. Including classified data in an unclassified proposal may be grounds for the Agency to determine the proposal non-responsive and not evaluate it.
In some instances, work being performed on Phase II contracts will require security clearances. If a Phase II contract requires classified work, the offeror must have a facility clearance and appropriate personnel clearances.
Note that although this topic carries no ITAR restriction paragraph, infrared focal plane array technology is frequently export controlled in practice, and specific seeker performance requirements are often classified. Keep the unclassified proposal unclassified and be careful about claims regarding specific naval seeker systems.
The References
Only three, and together they define the topic's premise almost exactly.
Goossens, Navickaite, Monasterio, and colleagues, "Broadband image sensor array based on graphene-CMOS integration," Nature Photonics 11, 366 to 371, 2017. This is the demonstration the topic is building on: a broadband image sensor array made by integrating graphene with CMOS. Read it as the baseline, and be explicit about what your work adds, since the topic's whole framing is that demonstrations exist and the remaining risks are manufacturability and performance at naval requirements.
Neumaier, Pindl, and Lemme, "Integrating graphene into semiconductor fabrication lines," Nature Materials 18, 525 to 529, 2019. This is risk two, back-end-of-line integration, stated as a review. If your BEOL transfer and patterning approach does not engage with the issues this paper raises, a reviewer will notice.
Koppens, Mueller, Avouris, and colleagues, "Photodetectors based on graphene, other two-dimensional materials and hybrid systems," Nature Nanotechnology 9, 780 to 793, 2014. This is the device physics foundation, including the sensitized and hybrid architectures that risk three concerns.
Three references, three of the four risks. The fourth risk, focal plane array system performance against naval ISR requirements, has no cited reference, which is consistent with it being the part no one has published because no one has built it. That is also where your proposal has the most room to differentiate.
Bring additional literature yourself, particularly on colloidal quantum dot and metal oxide nanocrystal sensitization stability, since the keywords name those approaches while the reference list does not cover them, and stability is the property most likely to be questioned.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
DSIP Topic Q&A closes: October 7, 2026, two weeks before the topic closes, per the DoW SBIR Program BAA
Proposal deadline: October 21, 2026
Selection notification: within 90 days of the closing date, approximately January 19, 2027
Period of performance: 24 months from award
A working backward plan
Before September 23. Audit the funding provenance of every feasibility result you intend to cite, because work based upon or logically extending from prior or ongoing federally funded SBIR or STTR work is excluded and failing the feasibility bar means the proposal is not evaluated. Confirm the work was substantially performed by your firm or your PI. Resolve intellectual property ownership or licensing for graphene transfer, encapsulation, and sensitization processes, since license rights must be in place before submission. Pull your measured integrated-device carrier mobility, Dirac-point distribution, hysteresis, and Raman D-to-G data, and be ready to present them as distributions across wafers and lots rather than best-coupon values. Quantify your die- or wafer-scale integration yield and uniformity. Secure or confirm your foundry CMOS readout relationship and process design kit access, since the Base task requires a BEOL integration flow for a foundry ROIC. Identify your low-light and obscurant test approach, including whether you need a dark field site, a calibrated low-light chamber, or a fog chamber. Model your Percentage of Work before finalizing foundry, supplier, university, and integrator arrangements. Approach the Naval Research Laboratory and other named organizations about letters of support. Confirm SAM registration and your CMMC Level 1 posture.
September 23 through October 5. Draft the 5-page Phase I justification around measured broadband photoresponse on coupon or graphene-on-CMOS structures and quantified die- or wafer-scale integration yield and uniformity, remembering that requirements definition, FPA modeling, and pixel and BEOL design are explicitly not required at proposal. Draft the 13-page technical proposal organized around the four named risks and the Base and Prototype task structure. Address the long-wave infrared end of the band explicitly rather than extrapolating a short-wave sensitization story. Draft the 2-page commercialization strategy answering all five enumerated questions. Draft the 3,000 character cover sheet abstract and the 3,000 character anticipated benefits and commercial applications discussion.
October 6 through October 7. Submit questions through DSIP Topic Q&A before it closes. Worth asking: the broadband versus short-wave infrared scope question if it affects your architecture, the CCR evaluation discrepancy, and whether any government-furnished readout wafers or test support are available.
October 8 through October 14. Build the cost volume in the DSIP online webform following the Cost Breakdown Guidance in the DoW 2026 SBIR BAA. Price the foundry CMOS run, graphene growth and transfer at wafer scale, encapsulation and patterning, sensitizer deposition, camera integration electronics and optics, the characterization campaign including nightglow and obscurant testing, and the environmental and reliability assessment. Justify equipment purchases and identify any Government Furnished Equipment needs, remembering those are settled in contract negotiations. Add supplementary cost detail as a Volume 3 PDF if useful. Complete the SBIR/STTR TABA Request Form and place it in Volume 5.
October 15 through October 18. Complete Volume 4, the Company Commercialization Report, carefully. Assemble Volume 5 with the TABA form, letters of support, and a Data Management Plan if applicable. Complete Volume 6 training and the Volume 7 foreign affiliations webform, remembering that Volume 7 must be the webform and will not be accepted as a PDF in Volume 5, and that no previous versions should be uploaded to Volume 5. Run compliance: 5 plus 15 pages with the 2-page commercialization strategy inside the 15, all figures, tables, charts, and references counted inside, no appendices, unclassified or CUI only.
October 19 through October 20. Submit and certify in DSIP.
Frequently Asked Questions
What is OSW-Reliance 21 SBIR topic OSW26BZ06-DV027?
OSW26BZ06-DV027 is a Direct to Phase II SBIR topic titled "Monolithic Graphene-CMOS Broadband (UV-LWIR) Focal Plane Arrays for target detection in low-light conditions," released under the Office of the Secretary of War, Reliance 21, 2026 SBIR Broad Agency Announcement, Release 6. The objective is to develop, fabricate, and demonstrate a wafer-scale, monolithically integrated graphene-on-CMOS image sensor providing broadband imaging from 300 nanometers to 10,000 nanometers in a compact, low-power, uncooled focal plane array.
How much funding is available?
$2,000,000 over 24 months. Phase II awardees may also request up to $50,000 in Technical and Business Assistance, in addition to the cost ceiling and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5.
When is the proposal deadline?
The topic opens September 23, 2026 and proposals are due October 21, 2026 through the Defense SBIR/STTR Innovation Portal at dodsbirsttr.mil.
Can I submit a Phase I proposal?
No. This topic solicits Direct to Phase II proposals only, and Phase I proposals will not be accepted.
What feasibility evidence does the topic accept?
Fabrication and measured characterization of graphene phototransistor or graphene-on-CMOS coupon test structures exhibiting broadband ultraviolet-through-long-wave-infrared photoresponse, and demonstration of die- or wafer-scale graphene integration with quantified device yield and uniformity. Documentation may comprise technical reports, peer-reviewed publications, measured data, and design artifacts.
What do I not need to have done yet?
Requirements definition for the naval use case, detailed focal plane array performance modeling, and pixel-architecture and back-end-of-line integration-flow design need not be complete at proposal. Those activities are carried into the funded Phase II Base effort.
Can my feasibility evidence come from a prior SBIR award?
No. The OSW-Reliance 21 Direct to Phase II guidelines state that feasibility documentation cannot be based upon or logically extend from any prior or ongoing federally funded SBIR or STTR work, and that the work must have been substantially performed by the proposer or the Principal Investigator. The Volume 2 instruction uses the weaker phrasing "must not be solely based on"; plan against the stricter reading. This matters here because graphene photodetector work in the United States has been heavily SBIR funded.
What if my feasibility documentation falls short?
If the proposer fails to demonstrate technical merit and feasibility equivalent to the Phase I level as described in the topic, the related Phase II proposal will not be evaluated.
Do I need to own the intellectual property?
Yes, or hold license rights. If technology in the feasibility documentation is subject to intellectual property, the proposer must either own the IP or have obtained license rights prior to proposal submission, sufficient to enable it and its subcontractors to legally carry out the proposed work. Graphene transfer, encapsulation, and sensitization processes are commonly licensed, so resolve this before submitting.
What is the spectral requirement?
Broadband imaging from 300 nanometers to 10,000 nanometers, spanning ultraviolet through long-wave infrared, in an uncooled focal plane array.
Why is monolithic integration better than the current approach?
The incumbent approach bonds InGaAs photodiode arrays to a silicon readout integrated circuit by indium-bump flip-chip bonding. The topic names five resulting constraints: high unit cost, a limited domestic foundry base, pixel pitches difficult to scale below roughly 15 micrometers, array formats limited by hybridization yield, and spectral response typically cutting off near 1,700 nanometers. Monolithic integration removes the bonding step that drives most of that.
What are the four risks the topic wants retired?
Graphene material quality and uniformity at wafer scale, covering carrier mobility, Dirac-point control, low hysteresis, and low defect density such as Raman D-to-G ratio, held tight across full wafers and lot to lot. Repeatable, high-yield back-end-of-line transfer, encapsulation, patterning, and contacting of graphene on CMOS readout wafers. Absorber sensitization and passivation delivering uniform, stable, low-noise broadband response. And a focal plane array meeting the noise-equivalent irradiance, dynamic range, frame rate, operability, and stability required for naval ISR.
What mobility should I target?
A substantial increase in integrated-device carrier mobility, with a goal approaching 10,000 square centimeters per volt-second. Note "integrated-device," meaning mobility in the finished device after transfer, encapsulation, patterning, and contacting, not mobility measured on a pristine film.
How is Phase II structured?
Into a Base task and a Prototype task. The Base task defines imager requirements against a representative naval ISR use case, develops and validates FPA performance models, completes the pixel-architecture and BEOL integration-flow design for a foundry CMOS ROIC, fabricates and characterizes coupon and small-array test structures, and finalizes the wafer-scale process definition. The Prototype task matures the wafer-scale process for high yield and uniformity across full wafers, builds and demonstrates a functioning broadband camera, characterizes it against Navy-relevant targets, and conducts an initial environmental and reliability assessment.
What demonstration scenarios are required?
Characterization against Navy-relevant targets including passive low-light imaging under atmospheric nightglow and imaging through obscurants such as fog and haze. Both require real test infrastructure, either a dark field site or a calibrated low-light chamber, and either a fog chamber or measured field conditions.
What are the Phase II deliverables?
Validated performance models and final FPA design; prototype focal plane arrays and cameras; the documented wafer-scale process flow with measured yield and uniformity; a full test and evaluation data package; and a Phase III transition and manufacturing plan including a path to a pilot qualification lot on the order of tens of wafers with defined yield and performance metrics.
Is the requirement full ultraviolet to long-wave infrared, or short-wave infrared?
The title and objective specify 300 to 10,000 nanometers. The Phase II and Phase III sections repeatedly say "broadband and short-wave infrared," and both named demonstration scenarios are short-wave infrared applications. The reasonable reading is that the full band is the objective while short-wave infrared is the near-term naval application driving the demonstration. Raise it through DSIP Topic Q&A if it affects your architecture.
How long can my technical volume be?
Twenty pages maximum, divided into Part 1 Phase I Justification at 5 pages maximum and Part 2 Phase II Technical Proposal at 15 pages maximum. The Technology Transition and Commercialization Strategy is limited to 2 pages and counts toward the 15. All figures, tables, charts, and references count inside the limits, and no separate appendices will be evaluated.
What must the commercialization strategy address?
Five specific questions. What is the first product this technology will go into. Who will be your customers and what is your estimate of the market size. How much funding will you need to bring the technology to market and how will you raise those funds. Does your company contain marketing expertise and if not how do you intend to bring it in. Who are your competitors and what is your price or quality advantage over them.
Is this topic ITAR restricted?
No topic-level ITAR or EAR restriction paragraph appears on DV027, unlike NV024 and DV025 in the same release. Note that infrared focal plane array technology is frequently export controlled in practice, and the release's Additional Information section still addresses foreign national disclosure, so treat export control as a live consideration in your business planning.
What CMMC level applies?
The projected requirement for this topic is CMMC Level 1.
What cost volume format do I use?
The DSIP online Cost Volume webform. OSW-Reliance 21 does not require a separate Excel template. Supplementary cost detail may be uploaded as a PDF attachment within Volume 3. Follow the Cost Breakdown Guidance in the DoW 2026 SBIR BAA.
Are there Percentage of Work restrictions?
Yes. OSW-Reliance 21 will not accept any deviation to the Percentage of Work requirements described in the DoW solicitation. This is a real risk on a topic that naturally involves a CMOS foundry, a graphene supplier, a university, and possibly a camera integrator. Model your POW before assembling the team.
Is the Company Commercialization Report evaluated?
The document conflicts with itself. The Direct to Phase II section states that CCR information will not be considered by "SCO," which appears to be residual text from another organization's instructions. The Phase I section of the same document states that CCR information will be considered by OSW-Reliance 21 during proposal evaluations. Complete it carefully either way and raise the discrepancy through DSIP Topic Q&A if it matters.
How do I request TABA?
Using the SBIR/STTR TABA Request Form, included in Volume 5 of the DSIP submission. OSW will not accept TABA requests that do not use the form or that are not submitted in Volume 5. Phase II is up to $50,000 per project, in addition to the cost ceiling and not subject to profit or fee.
What optional documents help?
Letters of support from prospective transition stakeholders within DEVCOM C5ISR Center, PAE Maneuver Ground, PAE Maneuver Air, CPE Autonomy, the Naval Research Laboratory, or the Air Force Research Laboratory. For this topic the Naval Research Laboratory is the most natural fit, and NRL is also among the named evaluating organizations. A Data Management Plan addressing provenance, licensing, and protection of pre-training and government-furnished data is also encouraged.
Who evaluates my proposal?
Government technical evaluators from Army DEVCOM C5ISR Center, Army DEVCOM Army Research Lab, the Naval Research Laboratory, and the Air Force Research Laboratory may participate. Non-government support contractors may assist with administrative handling under a non-disclosure agreement but do not participate in selection decisions.
When will I hear back?
Within 90 days of the closing date of the topic, which is approximately January 19, 2027. Notifications go through DSIP to both the Corporate Official and the Principal Investigator listed on the proposal.
Does Phase II require a multi-service transition plan?
Yes. Phase II efforts under this release shall include a transition plan addressing at least two of the three Services. Phase II contracting actions are anticipated to be firm-fixed-price or cost-plus-fixed-fee at the discretion of the Contracting Officer.
What is the commercial market?
Automotive and autonomous-vehicle vision through fog, smoke, and darkness; machine vision and semiconductor and solar wafer inspection; agricultural and food-quality sorting; medical and biometric imaging; and environmental and industrial process monitoring. Automotive short-wave infrared vision is the largest of these and is driven by the same cost and pixel pitch constraints the topic identifies.
Who is the technical point of contact?
The release strongly encourages engaging the Technical Point of Contact listed in the topic description during pre-release, but no TPOC appears in the DV027 description or in any of the four topic descriptions in this release. Use DSIP Topic Q&A, and send administrative questions to osd.pentagon.ousd-atl.mbx.communities-of-interest@mail.mil.
Positioning Advice for Companies Considering This Topic
Audit your feasibility provenance first, before anything else. Feasibility documentation cannot be based upon or logically extend from any prior or ongoing federally funded SBIR or STTR work, and a proposal that fails the feasibility bar is not evaluated at all. Graphene photodetector development in the United States is heavily SBIR-funded, so this is a real risk for exactly the companies most likely to bid. Trace every result. Internally funded work, privately funded work, non-SBIR government contract work, foundry-partnered development, and your own published academic work under other funding are all cleaner.
Report distributions, not champion devices. The first named risk is material quality and uniformity "held tight across full wafers and lot to lot." That is a statistical process control claim. Give mobility, Dirac point, hysteresis, and Raman D-to-G as distributions with wafer maps and lot-to-lot comparison. A single spectacular coupon answers a question the topic did not ask.
Quote integrated-device mobility, honestly. The goal approaching 10,000 square centimeters per volt-second is specified for the integrated device, after transfer, encapsulation, patterning, and contacting. Citing a film-level number and hoping no one notices is the fastest way to lose credibility with a reviewer who works on this. State your current integrated-device value, the measurement conditions, and your path to the goal.
Treat wafer-scale BEOL transfer as the central manufacturing problem. It is the second named risk and it is the thing that has kept graphene out of production for a decade. The topic cites Neumaier, Pindl, and Lemme specifically on integrating graphene into semiconductor fabrication lines. Engage with that paper's issues directly, and give measured yield across full wafers rather than a process description.
Address the long-wave infrared end separately. Quantum dot sensitization gets you visible through short-wave infrared comfortably. Uncooled response at 10 micrometers is a different physical problem and a much bigger claim. A proposal that presents one sensitization story for 300 to 10,000 nanometers will read as not having thought it through. Say what mechanism carries the long-wave end, or be explicit about how far you actually reach and why that still meets the naval need.
Make sensitizer stability a first-class topic. Uniform, stable, and low-noise are the three qualities demanded of the absorber, and stability is where colloidal quantum dot and nanocrystal films usually fail. The reference list does not cover sensitization, so bring that literature yourself and address photodegradation, ambient sensitivity, encapsulation, and drift over the operating life.
Name your foundry. The Base task requires completing the BEOL integration flow for a foundry CMOS readout integrated circuit, and Phase III asks you to establish and qualify a domestic CMOS-foundry-based manufacturing capability. An existing foundry relationship with process design kit access is worth more to this proposal than almost any technical claim. If you have one, put it on page one. If you do not, explain how you get one and when.
Answer the five incumbent constraints one at a time. High unit cost, limited domestic foundry base, pixel pitch below 15 micrometers, array format limited by hybridization yield, and the 1,700 nanometer cutoff. Walk each and state your number. That is the clearest possible articulation of why monolithic integration is worth funding, and it uses the Navy's own framing.
Lead with operability. It is one of the five system metrics and it is exactly where monolithic integration should beat hybridization, because you are not limited by bump-bond yield. If your yield numbers are good, operability is where you convert a process advantage into a system advantage a focal plane engineer will recognize immediately.
Budget the demonstration campaign properly. Atmospheric nightglow imaging needs a genuinely dark site on a moonless night or a calibrated low-light chamber with the right spectral distribution. Obscurant testing needs a fog chamber or measured field conditions. Both are named requirements, both cost money and schedule, and both are easy to underprice. Say how you will measure the obscurant condition rather than showing a qualitative image pair.
Price the pilot lot plan as a deliverable. The Phase III transition and manufacturing plan must include a path to a pilot qualification lot on the order of tens of wafers with defined yield and performance metrics. That is manufacturing engineering, not a roadmap slide. It is also the deliverable that makes this program credible to a program office.
Lead the commercialization strategy with automotive. Short-wave infrared automotive vision is the largest market on the topic's own list and it is bounded by the same cost and pixel pitch limits the topic identifies. A cost model that closes at automotive volumes is also the thing that funds the domestic pilot line the Navy wants, which makes the defense and commercial cases mutually reinforcing rather than parallel.
Get a Naval Research Laboratory letter. The topic is written around naval seekers and naval ISR, NRL is on the encouraged letters list, and NRL is among the named evaluating organizations. It is the highest-leverage optional document here.
Model Percentage of Work before you build the team. A foundry, a graphene supplier, a university, and a camera integrator is a natural team for this work and a fast route to a POW violation that no deviation will fix.
Plan the page budget before drafting. Five pages of feasibility, thirteen of technical proposal, two of commercialization, everything counted inside, no appendices. Four risks, two tasks, a foundry integration flow, and a characterization campaign do not fit unless you decide the allocation up front.
OSW-Reliance 21 SBIR OSW26BZ06-NV026: Defect Metrology and Charge Trapping Dynamics in Transfer-Doped Diamond Transistors
Deadline: October 21, 2026
Funding Award Size: $300k
Description: Complete guide to OSW-Reliance 21 SBIR Phase I topic OSW26BZ06-NV026, defect metrology for charge trapping in hydrogen-terminated diamond transistors. $314K over 6 months. Closes October 21, 2026.
Quick Answer
OSW26BZ06-NV026 is a Phase I SBIR topic under the Office of the Secretary of War, Reliance 21, 2026 SBIR Broad Agency Announcement, Release 6. Hydrogen-terminated diamond transistors could be a step change in high-frequency, high-power RF electronics, but they suffer from current collapse and knee walkout, and nobody has the right instrument to find out why. This topic funds building that instrument. The award is $314,363 over 6 months, with a 20-page technical volume. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
Note the period of performance. Six months is the shortest Phase I in this release and unusually short for a metrology development effort, which shapes what you can credibly promise.
The gap is stated plainly. Commercial device-characterization tools, built primarily for silicon or traditional compound semiconductor materials, lack the specialized physics and sensitivity required to isolate and characterize traps in ultra-wide bandgap diamond devices. And because hydrogen-terminated diamond relies on surface-channel p-type conduction, the topic warns that validation of new techniques cannot necessarily rely on prior validation from conventional n-type materials such as gallium nitride, gallium oxide, or aluminum nitride, because the fundamentally different physical mechanisms and device architectures may not be acceptable surrogates.
The most important practical fact about this topic: the Government supplies the devices. All primary experimental validations and core metrology capability demonstrations must be performed on Government-provided hydrogen-terminated diamond material and RF transistor devices, furnished as Government Furnished Property. That is true in Phase I as well as Phase II. Your program schedule depends on someone else's delivery.
Topic At a Glance
Topic number: OSW26BZ06-NV026
Title: Defect Metrology and Charge Trapping Dynamics in Transfer-Doped Diamond Transistors
Agency: Office of the Secretary of War, Reliance 21, administered by the OUSW(R&E) SBIR Program
Solicitation: OSW-Reliance 21, 2026 SBIR Broad Agency Announcement, Release 6, Proposal Submission Instructions
Program type: Phase I
Base award: $314,363
Base period of performance: 6 months, the shortest in this release
Technical volume limit: 20 pages, with all figures, tables, charts, and references counted inside that limit
OUSW (R&E) Critical Technology Area: Microelectronics
Component Technology Priority Areas: Microelectronics, Quantum Science
Projected CMMC level requirement: Level 2 (Self)
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic, unlike NV024 and DV025 in the same release. Note that the release's Additional Information section still refers to foreign national disclosure "per the ITAR notice in the topic description," and the Phase III section notes that some technologies evaluated fall under ITAR or Commerce Control List export controls
Government Furnished Property: the Government will provide hydrogen-terminated diamond material and RF transistor devices, and representative diamond test structures in Phase I
Target defects: charge traps causing current collapse and knee walkout in surface-channel diamond transistors
Required outputs from the technique: trap density, energy levels, physical location within the device stack, and time constants
Candidate trap locations: the gate dielectric, the dielectric-to-diamond interface, the diamond epitaxial layer, the epitaxial-to-substrate interface, and the diamond bulk substrate
Method constraint: non-destructive
Phase II end state: the complete, operational prototype measurement system, hardware and software, is delivered to the Government, with an SOP manual and hands-on training
Technical and Business Assistance: Phase I up to $6,500, Phase II up to $50,000 per project, in addition to the cost ceilings and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5
Cost volume: the DSIP online Cost Volume webform is required. No separate Excel template
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026
Submission portal: DSIP at dodsbirsttr.mil
Keywords: diamond transistors, charge trapping, current collapse, semiconductor metrology, ultra-wide bandgap, RF power electronics, defect dynamics, surface-channel devices, hydrogen-terminated diamond
The Objective, Read Carefully
Develop an innovative, non-destructive diamond semiconductor defect metrology technique and laboratory-scale apparatus to identify, quantify, determine location in the device, and energetically characterize charge traps that cause current collapse in surface-channel diamond transistors.
Four verbs, and each is a separate capability your instrument must have. Identify, meaning distinguish one trap population from another. Quantify, meaning produce a density. Determine location in the device, meaning attribute a trap to a specific layer or interface. Energetically characterize, meaning extract an energy level. The description adds a fifth quantity, time constants.
"Non-destructive" is a constraint, not a preference, and it rules out approaches that require sectioning, delayering, or otherwise consuming the device. That matters because the devices are Government Furnished Property in limited supply and because a metrology tool that destroys the part cannot be used for production screening, which is the eventual commercial application.
"Laboratory-scale apparatus" sets the form factor expectation. You are building a benchtop instrument, not a fab tool.
Why This Problem Exists
The promise
Hydrogen-terminated, meaning surface-channel, diamond transistors hold great promise for next-generation, high-frequency, high-power RF electronics and advanced communications due to diamond's high breakdown field and superior thermal conductivity.
The obstacle
Wide-scale operational deployment is constrained by current collapse and knee walkout, phenomena where transient charges trapped at defect sites in the semiconductor prevent the transistor from operating at its full, high-frequency RF power.
Both terms are worth knowing precisely, because the topic uses them as the observable behavior your technique must explain. Current collapse is the reduction in drain current under RF or pulsed operation relative to DC characteristics, caused by trapped charge modulating the channel. Knee walkout is the shift of the knee voltage, the boundary between the linear and saturation regions, toward higher drain voltage under stress, which reduces the usable voltage swing and therefore the achievable RF power. Both are classic trapping signatures familiar from gallium nitride, and both cost you the power density that made the wide bandgap material attractive in the first place.
Why existing tools fail
Commercial device-characterization tools, built primarily for silicon or traditional compound semiconductor materials, lack the specialized physics and sensitivity required to isolate and characterize traps in ultra-wide bandgap diamond devices.
Standard capacitance-voltage or simple transient electrical analyses do not provide sufficient physical insight into trap dynamics.
Two named inadequate approaches, CV and simple transient electrical analysis. If your technique is a variation on either, you need to explain what you add that produces physical insight rather than a signature.
The five candidate locations
These performance-limiting charge traps can reside in multiple distinct, critical locations within the device stack, including the gate dielectric, the dielectric-to-diamond interface, the diamond epitaxial layer, the epitaxial-to-substrate interface, and the diamond bulk substrate.
Five locations, and location attribution is the capability the topic emphasizes most. The Phase II validation definition names it explicitly: isolating and distinguishing between different trap locations, for example the gate-dielectric interface versus the bulk epitaxial layer.
This is the technically hardest requirement in the topic. An electrical measurement at the terminals integrates contributions from everywhere in the stack. Separating five spatially distinct populations from terminal measurements requires either a physical discriminant, such as depth-sensitive optical excitation or a frequency-dependent coupling argument, or a modeling framework that inverts the measurement, or both. State which and defend it.
What the system must extract
This topic seeks the development of a comprehensive measurement system and associated methodology capable of extracting trap density, energy levels, physical location, specifically identifying the layers or interfaces where the traps reside, and time constants characterizing these defects, so that design teams can efficiently work to address them.
Specifically, a successful technique and system must be capable of resolving multiple trap populations and correlating these distinct signatures with observed current collapse behavior in diamond.
Note the last clause. It is not enough to find traps. You must correlate the trap signatures with the observed current collapse behavior, which means your instrument has to measure the RF or pulsed degradation and the trap properties on the same device and connect them causally.
The p-type surrogate warning
Because hydrogen-terminated diamond relies on unique surface-channel p-type conduction, validation of these new techniques cannot necessarily rely on prior validation from conventional n-type semiconductor materials, for example gallium nitride, gallium oxide, or aluminum nitride, as the fundamentally different physical mechanisms and device architectures may not be acceptable surrogates for the diamond device behavior.
This paragraph is aimed squarely at the most likely proposal in the pile: a company with a proven gallium nitride trap characterization tool proposing to point it at diamond. The topic is telling you in advance that a GaN validation history is not sufficient evidence, and that your physics argument has to account for surface-channel p-type conduction rather than assuming the n-type framework transfers.
If you do have GaN or gallium oxide heritage, use it, but use it as capability evidence while making an explicit argument about what changes for hydrogen-terminated diamond: the two-dimensional hole gas at the hydrogen-terminated surface, the role of the surface acceptor layer, the p-type band alignment, and why your discriminant still works.
Government Furnished Property, Which Governs Your Schedule
To meet operational requirements, all primary experimental validations and core metrology capability demonstrations under this effort must be performed specifically on Government-provided hydrogen-terminated diamond material and RF transistor devices.
The Government will provide these verified working devices and test articles as Government Furnished Property during the execution phase to anchor the research.
The public literature contains examples of hydrogen-terminated diamond RF transistors that exhibit comparable baseline characteristics. While the GFP will vary in specific design, architecture, and properties from the literature, they can provide a baseline for concept development.
It is expected that preliminary or intermediate metrology demonstrations on other wide or ultra-wide bandgap semiconductor devices may be used for initial development, but the main and final capability demonstrations must be validated directly on the specific diamond device architectures provided.
In Phase I specifically: while initial benchtop calibration may utilize other wide bandgap materials, the core Phase I feasibility demonstration must be applied specifically to hydrogen-terminated diamond architectures. To facilitate this early validation, the Government will provide representative diamond test structures as Government Furnished Property.
What this means for a 6-month Phase I
Read the two constraints together. Your core Phase I feasibility demonstration must be on hydrogen-terminated diamond. The diamond comes from the Government as GFP. And you have six months.
That is a real schedule dependency on an external party. Three practical consequences.
First, ask about GFP timing through DSIP Topic Q&A before the topic closes. When the test structures become available, what they consist of, and how they are shipped and handled are all questions that materially affect whether a 6-month plan is executable. This is the single most valuable question you can ask on this topic.
Second, structure your Phase I so the non-diamond work happens first and productively. Initial benchtop calibration on other wide bandgap materials is explicitly permitted, so the sensible plan front-loads instrument setup, calibration, and modeling on available GaN or gallium oxide parts while the diamond GFP is in transit, then applies the technique to diamond as soon as it arrives.
Third, note that all requirements for Government Furnished Equipment or other assets, and associated costs, must be determined and agreed to during Phase II contract negotiations, per the release's Phase II instructions. State your GFP requirements explicitly in your proposal, including quantity, device type, and any test structure features you need, so the negotiation has a starting point.
Also useful: the topic points you at the literature for baseline characteristics. The two cited papers, Yu and colleagues 2022 on hydrogen-terminated diamond MOSFETs with state-of-the-art high RF power density, and Yu and colleagues 2021 on 1.26 watts per millimeter at 10 gigahertz for silicon nitride passivated hydrogen-terminated diamond MOSFETs, describe device classes comparable to what you will receive. Read both and design your instrument against those device geometries and impedance levels.
What Phase I Requires
Conduct a 6-month study to establish the scientific and technical feasibility of the proposed defect metrology technique.
The performer shall describe the measurement process, provide sound scientific arguments justifying the approach's ability to meet the project objectives of trap identification and measurement, and describe the hardware equipment requirement needed to perform the measurement.
Expected means to demonstrate feasibility include preliminary experimental demonstrations, a review of relevant existing techniques used on other types of semiconductors, and supporting modeling or simulations.
Phase I deliverables, three of them
A comprehensive feasibility study report detailing the proposed metrology process, scientific justification, and measurement equipment used for the approach.
Initial feasibility data, derived from preliminary experimental demonstrations, literature review of relevant semiconductor techniques, or modeling, supporting the viability of the technique for surface-channel p-type architectures.
A detailed Phase II transition plan outlining the schedule, testing procedures, and integration strategy for validating the prototype instrument on the provided diamond devices.
Reading the Phase I scope
The three permitted sources of feasibility evidence are generous: preliminary experimental demonstrations, literature review, or modeling. The second deliverable says "or," which means a well-constructed modeling and literature argument can satisfy it without new diamond data. That is a sensible accommodation given the six-month window and the GFP dependency, and it tells you where to put your effort.
But note that the second deliverable specifies feasibility for surface-channel p-type architectures. Whatever your evidence source, it has to speak to p-type surface-channel diamond, not to wide bandgap devices generally. That is the same warning as the surrogate paragraph, restated as a deliverable requirement.
The third deliverable, a detailed Phase II transition plan with schedule, testing procedures, and integration strategy, is worth more attention than proposers usually give a planning deliverable. Since Phase II ends with delivering an instrument to the Government, the transition plan is where you show you understand what building and handing over a validated tool involves.
Phase II, For Planning Purposes
Phase II proposals under this release may only be submitted by Phase I awardees, by invitation. The Phase II scope still matters now, because your Phase I proposal is evaluated partly on whether it sets up a credible Phase II, and because Phase II here is unusual.
Develop, construct, and validate a fully functional prototype, laboratory-scale, trap metrology instrument based on the Phase I design.
The performer is expected to demonstrate the system's ability to successfully isolate and distinguish between different trap locations, for example gate-dielectric interface versus bulk epitaxial layer, and correlate these signatures with measured RF current collapse on hydrogen-terminated diamond transistors, which will continue to be provided as Government Furnished Property.
The performer is expected to continue refining the technique developed in Phase I, incorporating those refinements into the prototype.
How validation is defined
The principal milestone of the Phase II effort is the validation of the developed hardware and measurement methodology at the performer's facility. For this effort, validation is explicitly defined as the successful identification, characterization, and physical location attribution of the specific charge traps responsible for current collapse and similar detrimental effects in the provided diamond transistors.
That is a specific and demanding definition. Not "the instrument works." Not "we measured trap signatures." You must identify, characterize, and locate the specific traps responsible for the current collapse in the devices you were given. Success is defined by explaining a particular device's particular degradation.
You hand over the instrument
Following this successful demonstration, the complete, operational prototype measurement system, including all associated hardware, specialized instrumentation, and software, will be provided as a deliverable to the Government.
This is the most consequential business fact about the topic and it should shape your Phase I proposal. Phase II is not a technology development contract that leaves you owning a tool. You build the instrument and deliver it, hardware and software.
Three Phase II deliverables are named.
Prototype system validation and hardware transfer: successful demonstration of the technique and hardware's ability to identify, locate, and quantitatively characterize, specifically providing numerical values for trap density, energy levels, and time constants, the specific traps responsible for current collapse and other detrimental effects in the GFP diamond devices, plus delivery of the fully operational prototype defect metrology test instrumentation to the Government.
Documentation and training: a comprehensive Standard Operating Procedure manual containing step-by-step measurement instructions tailored specifically to the provided diamond RF devices, accompanied by hands-on training for Government personnel.
Final report: a technical report detailing the identified trap signatures on the GFP devices, the methodology used to correlate these measurements to observed RF performance loss, detailed procedures by which the prototype tool can be used to reproduce the validation results on GFP devices, and details of any identified limitations of the technique or system related to the desired function, including any identified material or processing mitigation options related to those limitations.
What the delivery model implies for your business plan
Since the instrument goes to the Government, your commercial position rests on the design, the methodology, the software, and your ability to replicate and sell the system elsewhere, not on the delivered unit. That makes intellectual property strategy and data rights the center of your commercialization thinking, and it makes the Phase III commercial application the topic describes, commercializing the diagnostic system design, measurement software, and procedural methodology, the actual business.
It also means the SOP manual and the training are real deliverables with real cost. Writing a step-by-step measurement manual tailored to specific device architectures, and delivering hands-on training to Government personnel, is technical writing and instruction time that belongs in a Phase II budget.
Note as well the final report's requirement to identify material or processing mitigation options related to the technique's limitations. You are being asked to go one step past measurement and suggest what the device fabricators should do differently, which means understanding diamond device processing, not only metrology.
Phase III Dual Use
Military application
The delivered measurement technique will be used to analyze, optimize, and validate the reliability of high-power diamond transistors.
US Army DEVCOM is presently developing diamond transistor devices for the next generation of advanced military sensing, high-frequency communications, and broad-spectrum electromagnetic platforms. Successful outcomes of this research and development program will result in laboratory techniques and supporting equipment necessary to continue development of this technology.
It is expected that the technology will become a standard metrology technique within DEVCOM electronic device laboratories and among the Defense Industrial Base partners with semiconductor foundries presently being engaged in this development.
The named customer is specific and current: DEVCOM is developing diamond transistors now, which is why the GFP exists. That is a strong transition story, and "become a standard metrology technique within DEVCOM electronic device laboratories and among Defense Industrial Base partners" describes a multi-unit market rather than a single delivered instrument.
Commercial application
The performing small business can commercialize the diagnostic system design, measurement software, and procedural methodology to support the emerging domestic commercial diamond semiconductor industry.
This highly specialized metrology capability will be valuable to domestic and allied commercial foundries manufacturing advanced diamond electronics to improve device reliability and manufacturing yield for commercial 5G and 6G telecommunications, commercial satellite communications, commercial radar systems, and high-frequency communication systems requiring advanced thermal management.
Because this metrology system is designed to evaluate high-power semiconductor technologies, some of which fall under ITAR or Commerce Control List export controls, all commercialization, licensing, and replication services will need to comply with U.S. export control laws.
While this may limit certain commercial markets, it also has the potential to establish higher exclusivity within controlled markets.
That final observation is unusually candid for a solicitation and it is a good frame for your commercialization strategy. Export control narrows the market and raises the barrier to entry, which for a specialized instrument vendor can be a net advantage. Say so, rather than pretending the constraint does not exist or treating it as purely a cost.
Note that although this topic carries no ITAR restriction paragraph of its own, the Phase III text acknowledges ITAR and CCL exposure in the commercialization path. Plan for export control compliance in your business model even though the topic-level notice is absent.
Funding, Cost Structure, and OSW-Reliance 21 Mechanics
The award
$314,363 over 6 months. The Phase I base amount must not exceed the base limit set by the topic.
The short duration is the thing to plan around. Roughly $314,000 in six months is a high burn rate, which suits a focused instrument-development effort with existing hardware to build on and does not suit a program that has to buy and commission major capital equipment first.
Cost volume mechanics
OSW-Reliance 21 requires the use of the DSIP online Cost Volume webform. No separate Excel template is required. If supplementary cost detail is desired, it may be uploaded as a PDF attachment within Volume 3.
For this topic, be explicit about Government Furnished Property in your cost and facilities discussion. Requirements for government furnished equipment or other assets, and associated costs, must be determined and agreed to during Phase II contract negotiations, so stating what you need early serves you.
Percentage of Work, with no exceptions
Review the updated Percentage of Work calculation details included in the DoW solicitation. OSW-Reliance 21 will not accept any deviation to the POW requirements.
Metrology development invites university collaboration, particularly for trap physics modeling. Model your POW before you build the team.
Technical and Business Assistance
Phase I awardees may request up to $6,500 in TABA funding. Phase II awardees may request up to $50,000 per Phase II project. TABA funding is in addition to the Phase I and Phase II cost ceilings and is not subject to profit or fee.
All requests for TABA must be completed using the SBIR/STTR TABA Request Form, and the completed form must be included in Volume 5 of the proposal submission in DSIP. OSW will not accept requests for TABA that do not utilize the form or that are not included as a submission document in Volume 5.
For this topic, intellectual property counsel is the standout use, because the delivery model hands the physical instrument to the Government and your commercial position depends entirely on protecting the design, methodology, and software. Export control counsel is second, given the ITAR and CCL exposure the Phase III section describes.
Page limits and the no-appendix rule
The technical volume for this topic is limited to 20 pages. All figures, tables, charts, and references must be included within the page count limit listed in the topic index. Any pages past the limit will not be considered, and no separate appendices will be evaluated.
Follow all instructions under the Phase I Proposal Instructions section in the DoW SBIR Program BAA and use the Phase I technical volume template provided as Appendix A in that BAA.
The Company Commercialization Report, and a contradiction
Completion of the CCR as Volume 4 is required.
The Phase I Proposal Guidelines in this release state that information contained in the CCR will be considered by OSW-Reliance 21 during proposal evaluations. That is notable, since many components exclude it. The Direct to Phase II section of the same document states the opposite and attributes the statement to "SCO," which appears to be residual text from another organization's instructions.
For a Phase I proposal to NV026, the applicable statement is the one in the Phase I section: the CCR will be considered. Complete it carefully rather than perfunctorily, and raise the discrepancy through DSIP Topic Q&A if it matters to you.
Supporting documents that are optional but encouraged
Letters of Support from prospective transition stakeholders within DEVCOM C5ISR Center, PAE Maneuver Ground, PAE Maneuver Air, CPE Autonomy, the Naval Research Laboratory, or the Air Force Research Laboratory.
A Data Management Plan addressing provenance, licensing, and protection of pre-training data and government-furnished data.
The Data Management Plan is directly relevant here, and more so than on the other topics in this release, because your entire experimental program runs on government-furnished devices. A plan addressing how you handle, protect, and report data derived from GFP is a sensible inclusion.
On letters of support, note that the topic names US Army DEVCOM as the organization developing the diamond transistors. DEVCOM C5ISR Center is on the encouraged list, and DEVCOM Army Research Lab is named among the evaluating organizations. A letter from the DEVCOM element actually building these devices would be the strongest possible support document, since that element is presumably also the GFP source.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW solicitation.
Government technical evaluators from Army DEVCOM C5ISR Center, Army DEVCOM Army Research Lab, the Naval Research Laboratory, and the Air Force Research Laboratory may participate in the evaluation. Non-government support contractors may assist in administrative handling of proposals if the individual has signed a non-disclosure agreement, and they will not participate in selection decisions.
For this topic, DEVCOM Army Research Lab is the evaluator to write for. ARL is where diamond electronics work of this kind lives, and your reviewer may well be someone who has personally measured current collapse on one of the devices you will receive. Write with that level of specificity.
Proposing firms will be notified of selection or non-selection status within 90 days of the closing date of the topic, which is approximately January 19, 2027. Notifications will be issued through DSIP to both the Corporate Official and the Principal Investigator listed on the proposal.
Protests after award should be submitted, as prescribed in FAR 33.106(b) and FAR 52.233-3, to osd.ncr.ousd-r-e.mbx.SBIR-STTR-Protest@mail.mil.
Tri-service coordination and the TPOC question
This topic is of joint interest to the U.S. Army, the U.S. Navy, and the U.S. Air Force and Space Force through the organizations named above. Proposers are strongly encouraged to engage the Technical Point of Contact listed in the topic description during the pre-release period to discuss technical scope and transition opportunities across the Services.
No Technical Point of Contact appears in the NV026 topic description, or in any of the four topic descriptions in this release. Use DSIP Topic Q&A, and send administrative questions to osd.pentagon.ousd-atl.mbx.communities-of-interest@mail.mil. Given the GFP dependency, using the Q&A channel before it closes on October 7 is particularly important on this topic.
Note also that Phase II efforts under this release shall include a transition plan addressing at least two of the three Services, and that Phase II contracting actions are anticipated to be firm-fixed-price or cost-plus-fixed-fee at the Contracting Officer's discretion. For a metrology tool intended to become a standard technique across DEVCOM laboratories and Defense Industrial Base partners, the multi-service transition case is straightforward, but it must be written.
Classification
Phase I efforts are expected to be performed at the Unclassified and CUI level. Classified proposals are not accepted, and including classified data in an unclassified proposal may be grounds for the Agency to determine the proposal non-responsive and not evaluate it.
In some instances, work being performed on Phase II contracts will require security clearances. If a Phase II contract requires classified work, the offeror must have a facility clearance and appropriate personnel clearances.
A note on the Critical Technology Area designation
The OUSW (R&E) Critical Technology Area listed for this topic is Microelectronics. The release's introduction lists six Critical Technology Areas that the program prioritizes: Applied Artificial Intelligence, Biomanufacturing, Contested Logistics Technologies, Quantum and Battlefield Information Dominance, Scaled Directed Energy, and Scaled Hypersonics. Microelectronics is not among those six.
The Component Technology Priority Areas listed for the topic are Microelectronics and Quantum Science, which are standard Component Technology Priority Area names. The most plausible reading is that the OUSW (R&E) Critical Technology Area field was populated with a Component Technology Priority Area value. It is unlikely to affect your proposal, but if you are aligning your narrative to a Critical Technology Area, note that Quantum and Battlefield Information Dominance is the nearest of the six to this work, given diamond's role in quantum-adjacent and advanced RF electronics.
The References
Only two, and both are device papers rather than metrology papers.
Yu, Zhou, Guo, He, Ma, Yu, Song, Bu, and Feng, "Hydrogen-terminated diamond MOSFETs on (001) single crystal diamond with state of the art high RF power density," Functional Diamond, 2022.
Yu, Hu, Zhou, and colleagues, "1.26 W/mm Output Power Density at 10 GHz for Si3N4 Passivated H-Terminated Diamond MOSFETs," IEEE Transactions on Electron Devices, 2021.
The topic cites these specifically as examples of hydrogen-terminated diamond RF transistors exhibiting comparable baseline characteristics to the Government Furnished Property you will receive. That makes them functional specifications rather than background reading.
Read them for the things that determine whether your instrument can measure these devices: the gate dielectric and passivation stack, silicon nitride in the 2021 paper, the substrate orientation, (001) single crystal in the 2022 paper, the device geometry and gate dimensions, the operating voltages and currents, the frequency of operation, and the impedance environment. Then design your measurement around those parameters.
The absence of metrology references is itself informative in the same way as on DV025. The Government is stating a capability requirement and leaving the technique open, which means you carry the full burden of establishing awareness of the state of the art in trap characterization. Bring that literature yourself: deep level transient spectroscopy and its variants, drain current transient spectroscopy, pulsed IV and gate lag measurements, low-frequency noise spectroscopy, and the substantial gallium nitride trapping literature, while being explicit about what changes for p-type surface-channel diamond.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
DSIP Topic Q&A closes: October 7, 2026, two weeks before the topic closes, per the DoW SBIR Program BAA
Proposal deadline: October 21, 2026
Selection notification: within 90 days of the closing date, approximately January 19, 2027
Period of performance: 6 months from award
A working backward plan
Before September 23. Decide your physical discriminant for location attribution, because that is the hardest requirement and the one a reviewer will probe. Read both cited device papers and design your measurement around those device geometries, dielectric stacks, and operating conditions. Inventory your existing instrumentation honestly against a 6-month, $314,000 effort, since there is no time to specify, purchase, and commission major capital equipment. Line up access to wide bandgap devices for initial benchtop calibration, which is explicitly permitted, so the diamond GFP is not on your critical path from day one. Prepare your Government Furnished Property request specifically: quantity, device type, test structure features, and timing. Model your Percentage of Work before committing to university collaboration on trap physics. Approach DEVCOM about a letter of support, ideally the element developing the diamond devices. Confirm SAM registration and CMMC Level 2 self-assessment in SPRS. Download the DoW SBIR Program BAA Appendix A Phase I template.
September 23 through October 5. Draft the 20-page technical volume. Structure it around the measurement physics, the location discrimination mechanism, the modeling framework that inverts the measurement, the correlation with current collapse, the hardware requirement, and the three Phase I deliverables. Make the p-type surface-channel argument explicitly rather than leaning on n-type heritage. Write the Phase II transition plan deliverable as a real plan, since Phase II ends with delivering an instrument. Draft the 3,000 character cover sheet abstract and the 3,000 character anticipated benefits and commercial applications discussion.
October 6 through October 7. Submit questions through DSIP Topic Q&A before it closes. On this topic, the highest-value questions are about Government Furnished Property: when the Phase I diamond test structures become available, what they consist of, how many, and what the handling and shipping arrangements are. Also worth asking: the CCR evaluation discrepancy, and the Critical Technology Area designation if you are aligning your narrative to it.
October 8 through October 14. Build the cost volume in the DSIP online webform. Price instrumentation, any hardware you must build, modeling and simulation effort, device handling and probing fixtures suited to the GFP geometry, and the technical writing effort for the feasibility report and transition plan. State GFP requirements clearly. Add supplementary cost detail as a Volume 3 PDF if useful. Complete the SBIR/STTR TABA Request Form and place it in Volume 5.
October 15 through October 18. Complete Volume 4, the Company Commercialization Report, carefully, since the Phase I instructions state it will be considered. Assemble Volume 5 with the TABA form, letters of support, and a Data Management Plan addressing government-furnished data, which is particularly apt here. Complete Volume 6 training and the Volume 7 foreign affiliations webform, remembering that Volume 7 must be the webform and will not be accepted as a PDF in Volume 5, and that no previous versions should be uploaded to Volume 5. Run compliance: 20 pages with figures, tables, charts, and references counted inside, no appendices, unclassified or CUI only.
October 19 through October 20. Submit and certify in DSIP.
Frequently Asked Questions
What is OSW-Reliance 21 SBIR topic OSW26BZ06-NV026?
OSW26BZ06-NV026 is a Phase I SBIR topic titled "Defect Metrology and Charge Trapping Dynamics in Transfer-Doped Diamond Transistors," released under the Office of the Secretary of War, Reliance 21, 2026 SBIR Broad Agency Announcement, Release 6. The objective is to develop an innovative, non-destructive diamond semiconductor defect metrology technique and laboratory-scale apparatus to identify, quantify, determine location in the device, and energetically characterize charge traps that cause current collapse in surface-channel diamond transistors.
How much funding is available and for how long?
$314,363 over 6 months. That is the shortest period of performance in this release. Phase I awardees may also request up to $6,500 in TABA, in addition to the cost ceiling and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5.
When is the proposal deadline?
The topic opens September 23, 2026 and proposals are due October 21, 2026 through the Defense SBIR/STTR Innovation Portal at dodsbirsttr.mil.
How long can my technical volume be?
Twenty pages. All figures, tables, charts, and references must be included within that limit. Any pages past the limit will not be considered, and no separate appendices will be evaluated.
What is current collapse and knee walkout?
Current collapse is the reduction in drain current under RF or pulsed operation relative to DC characteristics, caused by transient charge trapped at defect sites modulating the channel. Knee walkout is the shift of the knee voltage toward higher drain voltage under stress, reducing usable voltage swing and achievable RF power. Both prevent hydrogen-terminated diamond transistors from operating at their full high-frequency RF power.
What must the technique measure?
Trap density, energy levels, physical location within the device stack, specifically identifying the layers or interfaces where the traps reside, and time constants. It must also resolve multiple trap populations and correlate those signatures with observed current collapse behavior in diamond.
Where can the traps be located?
Five named locations: the gate dielectric, the dielectric-to-diamond interface, the diamond epitaxial layer, the epitaxial-to-substrate interface, and the diamond bulk substrate. Attributing traps to a specific one of these is the capability the topic emphasizes most, and the hardest requirement in it.
Must the method be non-destructive?
Yes. Non-destructive is stated in the objective. That rules out sectioning or delayering approaches, which matters both because the devices are Government Furnished Property in limited supply and because the eventual commercial application is production screening.
Why won't commercial characterization tools work?
Commercial device-characterization tools are built primarily for silicon or traditional compound semiconductor materials and lack the specialized physics and sensitivity required for ultra-wide bandgap diamond. The topic also states that standard capacitance-voltage or simple transient electrical analyses do not provide sufficient physical insight into trap dynamics.
Can I use my gallium nitride validation history as evidence?
Not on its own. The topic warns that because hydrogen-terminated diamond relies on unique surface-channel p-type conduction, validation cannot necessarily rely on prior validation from conventional n-type materials such as GaN, Ga2O3, or AlN, since the fundamentally different physical mechanisms and device architectures may not be acceptable surrogates. Use GaN heritage as capability evidence, but make an explicit physics argument for p-type surface-channel diamond.
Who provides the diamond devices?
The Government. All primary experimental validations and core metrology capability demonstrations must be performed on Government-provided hydrogen-terminated diamond material and RF transistor devices, furnished as Government Furnished Property. In Phase I, the Government will provide representative diamond test structures as GFP.
Can I do preliminary work on other materials?
Yes. Preliminary or intermediate metrology demonstrations on other wide or ultra-wide bandgap semiconductor devices may be used for initial development, and initial benchtop calibration may utilize other wide bandgap materials. But the core Phase I feasibility demonstration must be applied specifically to hydrogen-terminated diamond architectures, and the main and final capability demonstrations must be validated directly on the provided diamond device architectures.
How do I plan a 6-month Phase I around Government Furnished Property?
Front-load the work that does not need diamond: instrument setup, calibration on other wide bandgap materials, and modeling. Then apply the technique to diamond as soon as the GFP arrives. Ask about GFP availability, quantity, device type, and shipping through DSIP Topic Q&A before it closes on October 7, since that is the single most schedule-relevant unknown on this topic. State your GFP requirements explicitly in the proposal.
What are the Phase I deliverables?
Three. A comprehensive feasibility study report detailing the proposed metrology process, scientific justification, and measurement equipment. Initial feasibility data, from preliminary experimental demonstrations, literature review, or modeling, supporting viability for surface-channel p-type architectures. And a detailed Phase II transition plan outlining the schedule, testing procedures, and integration strategy for validating the prototype instrument on the provided diamond devices.
Do I need new experimental data for Phase I?
Not necessarily. The second deliverable permits feasibility data derived from preliminary experimental demonstrations, literature review of relevant semiconductor techniques, or modeling. A well-constructed modeling and literature argument can satisfy it, provided it speaks specifically to surface-channel p-type architectures.
What happens in Phase II?
Develop, construct, and validate a fully functional prototype, laboratory-scale trap metrology instrument based on the Phase I design, demonstrating the ability to isolate and distinguish between different trap locations and correlate those signatures with measured RF current collapse on hydrogen-terminated diamond transistors provided as GFP.
How is Phase II validation defined?
Explicitly, as the successful identification, characterization, and physical location attribution of the specific charge traps responsible for current collapse and similar detrimental effects in the provided diamond transistors. The principal milestone is validation of the hardware and methodology at the performer's facility.
Do I keep the instrument I build?
No. Following successful demonstration, the complete operational prototype measurement system, including all associated hardware, specialized instrumentation, and software, is provided as a deliverable to the Government. Phase II also requires a Standard Operating Procedure manual tailored to the provided diamond RF devices and hands-on training for Government personnel.
What does the delivery model mean for my business?
Your commercial position rests on the design, methodology, and software rather than on the delivered unit, which makes intellectual property strategy and data rights central. The topic's own Phase III commercial application describes commercializing the diagnostic system design, measurement software, and procedural methodology, which is the actual business.
Is this topic ITAR restricted?
No topic-level ITAR or EAR restriction paragraph appears on NV026, unlike NV024 and DV025 in the same release. However, the release's Additional Information section refers to foreign national disclosure per the ITAR notice in the topic description, and the topic's own Phase III section states that some of the technologies this system evaluates fall under ITAR or Commerce Control List export controls and that all commercialization, licensing, and replication services will need to comply with U.S. export control laws. Plan for export control compliance in your business model.
What CMMC level applies?
The projected requirement for this topic is CMMC Level 2 with self-assessment.
What Critical Technology Area does this fall under?
The topic lists Microelectronics as the OUSW (R&E) Critical Technology Area, which is not among the six Critical Technology Areas named in the release introduction. Its Component Technology Priority Areas are Microelectronics and Quantum Science. The field appears to have been populated with a Component Technology Priority Area value. Of the six named Critical Technology Areas, Quantum and Battlefield Information Dominance is the nearest to this work.
Is the Company Commercialization Report evaluated?
For Phase I under this release, yes. The Phase I Proposal Guidelines state that CCR information will be considered by OSW-Reliance 21 during proposal evaluations. The Direct to Phase II section of the same document says the opposite and attributes it to a different organization, which appears to be residual text. Complete the CCR carefully.
What cost volume format do I use?
The DSIP online Cost Volume webform. No separate Excel template is required. Supplementary cost detail may be uploaded as a PDF attachment within Volume 3.
Are there Percentage of Work restrictions?
Yes. OSW-Reliance 21 will not accept any deviation to the Percentage of Work requirements described in the DoW solicitation. Model your POW before committing to university collaboration on trap physics modeling.
Who evaluates my proposal?
Government technical evaluators from Army DEVCOM C5ISR Center, Army DEVCOM Army Research Lab, the Naval Research Laboratory, and the Air Force Research Laboratory may participate. For this topic, DEVCOM Army Research Lab is the most likely technical reviewer, since ARL is where diamond electronics work of this kind lives.
When will I hear back?
Within 90 days of the closing date of the topic, which is approximately January 19, 2027. Notifications go through DSIP to both the Corporate Official and the Principal Investigator listed on the proposal.
What is the commercial market?
Domestic and allied commercial foundries manufacturing advanced diamond electronics, improving device reliability and manufacturing yield for commercial 5G and 6G telecommunications, satellite communications, commercial radar, and high-frequency communication systems requiring advanced thermal management. On the defense side, the topic expects the technique to become standard within DEVCOM electronic device laboratories and among Defense Industrial Base partners with semiconductor foundries engaged in this development.
Who is the technical point of contact?
The release strongly encourages engaging the Technical Point of Contact listed in the topic description during pre-release, but no TPOC appears in the NV026 description or in any of the four topic descriptions in this release. Use DSIP Topic Q&A, which is especially important here because of the Government Furnished Property dependency, and send administrative questions to osd.pentagon.ousd-atl.mbx.communities-of-interest@mail.mil.
Positioning Advice for Companies Considering This Topic
Lead with your location discrimination mechanism. Five candidate trap locations, terminal measurements that integrate all of them, and a requirement to attribute traps to specific layers and interfaces. That is the hardest thing in the topic and the thing a reviewer at DEVCOM ARL will probe first. Whether your discriminant is depth-sensitive optical excitation, frequency-dependent coupling, a bias-dependent sensitivity argument, or a model inversion, name it in the first page and defend the physics.
Make the p-type surface-channel argument explicitly. The topic warns in advance that n-type wide bandgap validation may not transfer. If you have GaN or gallium oxide heritage, and most credible bidders will, present it as capability while walking through what changes for hydrogen-terminated diamond: the two-dimensional hole gas at the terminated surface, the surface acceptor layer, the band alignment, and why your method still discriminates. Skipping that argument is the most predictable weakness in this pile of proposals.
Ask about the Government Furnished Property before the Q&A closes. Your core Phase I demonstration must be on Government-supplied diamond, you have six months, and you do not control the delivery. Timing, quantity, device type, test structure features, and handling arrangements are all legitimate questions, and asking them signals that you have thought about executing rather than only about physics.
Design your Phase I schedule so diamond is not on the critical path from day one. Initial benchtop calibration on other wide bandgap materials is explicitly permitted. Use that permission structurally: instrument setup, calibration, and modeling first, diamond validation as soon as the GFP arrives. A proposal whose entire plan waits on a shipment is fragile in a way a reviewer will notice.
Design the measurement against the actual devices. The two cited papers describe device classes comparable to the GFP, including a silicon nitride passivated stack and a (001) single crystal substrate. Read them for gate geometry, dielectric stack, operating voltages, currents, frequency, and impedance, and show that your fixturing and sensitivity suit those parameters. That level of specificity is the difference between a general trap metrology pitch and a proposal for this topic.
Connect the trap signature to the current collapse, not just to the trap. The topic requires correlating distinct trap signatures with observed current collapse behavior. That means measuring the RF or pulsed degradation and the trap properties on the same device and closing the causal loop. A technique that produces a beautiful trap spectrum with no link to the device's actual performance loss has not met the requirement.
Be honest about a 6-month, $314,000 envelope. There is no room to specify, buy, and commission major capital equipment. Show what you already have. If your approach needs an instrument you do not own, explain how you access it and what that costs in schedule.
Plan for the fact that you hand over the instrument. Phase II delivers the complete hardware and software to the Government, plus an SOP manual and hands-on training. Start thinking now about what you protect and how: the methodology, the software, the design, data rights assertions. That is your business, and getting it wrong in Phase I framing is expensive later.
Budget the documentation and training as real work. A step-by-step SOP tailored to specific device architectures, plus hands-on training for Government personnel, is technical writing and instruction time. Most proposals will treat it as a line item and underprice it.
Address the mitigation question. The Phase II final report asks you to identify material or processing mitigation options related to the technique's limitations. That means understanding diamond device fabrication, not only measurement. Showing some of that understanding in Phase I positions you as a partner to the device developers rather than a vendor of a black box.
Go after a DEVCOM letter of support. The topic names US Army DEVCOM as the organization presently developing these diamond transistors, which makes it both the likely GFP source and the likely eventual customer for multiple instruments. DEVCOM C5ISR Center is on the encouraged letters list and DEVCOM ARL is among the named evaluators.
Include a Data Management Plan. It is optional but encouraged across this release, and on this topic it is genuinely apt, because your entire experimental program runs on government-furnished devices and the plan is specifically described as addressing protection of government-furnished data.
Bring the trap metrology literature yourself. The topic cites only two device papers and no metrology references, so it gives you no scaffolding and you carry the burden of demonstrating awareness of the state of the art. Deep level transient spectroscopy and its variants, drain current transient spectroscopy, pulsed IV and gate lag, low-frequency noise spectroscopy, and the GaN trapping literature all belong in your related work, with an explicit account of what changes for diamond.
Frame export control as exclusivity, following the topic's own lead. The Phase III text notes that export control may limit certain commercial markets while establishing higher exclusivity within controlled ones. That is a more sophisticated commercialization argument than a large unqualified market estimate, and it comes from the solicitation itself.
OSW-Reliance 21 SBIR OSW26BZ06-DV025: Compact Passive Radar
Deadline: October 21, 2026
Funding Award Size: $2.1m
Description: Complete guide to OSW-Reliance 21 SBIR topic OSW26BZ06-DV025, compact attritable passive radar. Phase I $314K over 12 months or Direct to Phase II $2.1M over 24 months. Closes October 21, 2026.
Quick Answer
OSW26BZ06-DV025 is an OSW-Reliance 21 SBIR topic under the 2026 SBIR Broad Agency Announcement, Release 6, and it is the only topic in this release that accepts both a Phase I proposal and a Direct to Phase II proposal. Phase I is $314,363 over 12 months. Direct to Phase II is $2,095,748 over 24 months. Both cap the technical volume at 20 pages. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The requirement is a radar that never transmits. In contested anti-access and area-denial environments, active radar emissions act as beacons for adversary electronic warfare, so the Department needs covert sensing using illuminators of opportunity. The topic is explicit: this effort requires a 100 percent receive-only payload, and the sensor shall not transmit any RF energy. That sentence appears again in the Phase II deliverable definition, which requires demonstrating detection and imaging "without emitting any RF energy from the passive sensor itself."
What makes this hard is not passive radar as a concept but the combination of capabilities at a price and size the Department is willing to throw away. Current passive systems are bulky, expensive, and primarily optimized for signals intelligence and electronic support measures rather than radar functionality. They lack the tightly synchronized, multi-channel receivers necessary for direct reference signal acquisition, spatial processing, and clutter cancellation. The Department seeks a paradigm shift toward low-cost, multi-channel attritable sensors designed specifically for high-risk deployments.
Concretely: a small form factor passive X-band radar doing both synthetic aperture imaging and ground moving target indication, light enough for a Group 1 or 2 unmanned aircraft, a loitering munition, or an unattended ground sensor, cheap enough that losing it is acceptable.
Topic At a Glance
Topic number: OSW26BZ06-DV025
Title: Compact Passive Radar
Agency: Office of the Secretary of War, Reliance 21, administered by the OUSW(R&E) SBIR Program
Solicitation: OSW-Reliance 21, 2026 SBIR Broad Agency Announcement, Release 6, Proposal Submission Instructions
Program types accepted: Phase I and Direct to Phase II. This is the only topic in the release offering both
Phase I award: $314,363 over 12 months, technical volume limited to 20 pages
Direct to Phase II award: $2,095,748 over 24 months, technical volume limited to 20 pages, structured as 5 pages of Phase I justification plus 15 pages of Phase II technical proposal
OUSW (R&E) Critical Technology Area: Quantum and Battlefield Information Dominance
Component Technology Priority Area: Integrating Sensing and Cyber
Projected CMMC level requirement: Level 2 (Self)
Export control status: ITAR restricted. The technology within this topic is restricted under the International Traffic in Arms Regulation, 22 CFR Parts 120-130, or the Export Administration Regulation, 15 CFR Parts 730-774
Absolute constraint: a 100 percent receive-only payload. The sensor shall not transmit any RF energy
Band: X-band
Required modes: synthetic aperture radar imaging and ground moving target indication
Host platforms: Group 1 or 2 unmanned aircraft systems, loitering munitions, or unattended ground sensors
Design driver: extreme size, weight, power, and cost constraints, with attritability as an explicit requirement
Technical and Business Assistance: Phase I up to $6,500, Phase II up to $50,000 per project, in addition to the cost ceilings and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5
Cost volume: the DSIP online Cost Volume webform is required. No separate Excel template
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026
Submission portal: DSIP at dodsbirsttr.mil
Keywords: passive radar, GMTI, SAR, bistatics
The Two Paths, and How to Choose
This topic appears in both the Phase I table and the Direct to Phase II table of the OSW-Reliance 21 Release 6 topic index. That is unusual and it is the first decision you have to make.
Phase I
$314,363 over 12 months, with a 20-page technical volume. Follow the Phase I Proposal Instructions in the DoW SBIR Program BAA and use the Phase I technical volume template provided as its Appendix A.
The Phase I objective is a feasibility study to determine the scientific, technical, and commercial merit of a small form factor, attritable passive radar architecture. The desired end product is a robust simulation and a basic laboratory prototype that validates the feasibility of detecting targets using signals of opportunity while strictly adhering to the SWaP-C metrics required for expendable systems.
Note that a 20-page Phase I technical volume is generous by SBIR standards, and twice what topic NV024 in this same release allows.
Direct to Phase II
$2,095,748 over 24 months, with a 20-page technical volume divided as follows: Part 1, Phase I Justification, 5 pages maximum, and Part 2, Phase II Technical Proposal, 15 pages maximum.
The statutory basis is 15 U.S.C. 638(cc), which allows the Department to make a Phase II award without regard to whether the concern received a Phase I award for that project.
The feasibility restriction that will disqualify many DP2 proposers
Read this carefully, because it is stricter than what most components impose and stricter than DARPA's equivalent language in the same cycle.
Each eligible topic requires that proposers provide documentation to demonstrate that the feasibility described in the Phase I section of the topic has been met. Feasibility documentation cannot be based upon or logically extend from any prior or ongoing federally funded SBIR or STTR work. Work submitted within the feasibility documentation must have been substantially performed by the proposer or the principal investigator. If technology in the feasibility documentation is subject to intellectual property, the proposer must either own the IP or must have obtained license rights to such technology prior to proposal submission, to enable it and its subcontractors to legally carry out the proposed work.
The Volume 2 instruction repeats it: feasibility documentation must not be solely based on work performed under prior or ongoing federally funded SBIR or STTR work.
Two different formulations appear in the same document, "cannot be based upon or logically extend from" in the DP2 guidelines and "must not be solely based on" in the Volume 2 instruction. The stricter reading is the first, and it is the one to plan against. If your passive radar prototype was built under a prior SBIR or STTR from any federal agency, that work may not be usable as your feasibility basis, and even work that logically extends from it may be excluded.
For a small radar company, much prior work is federally funded, and a great deal of it is SBIR-funded specifically. Before you commit to the DP2 path, audit the provenance of every result you intend to cite. Internally funded development, privately funded development, non-SBIR government contract work, and commercial product development are all cleaner ground.
And the consequence of getting it wrong is severe: if the proposer fails to demonstrate technical merit and feasibility equivalent to the Phase I level as described in the topic, the related Phase II proposal will not be evaluated.
Which path fits you
Choose Direct to Phase II if you have a working multi-channel passive radar receiver with a demonstrated architecture, developed outside prior federal SBIR or STTR funding, and you can substantiate the Phase I feasibility described in the topic in five pages. Choose Phase I if your passive radar capability is at the algorithm and simulation stage, or if your existing hardware traces back to SBIR funding you cannot use as a feasibility basis.
Note that the DP2 award is roughly 6.7 times the Phase I award, which is a large ratio. If you can clear the feasibility bar honestly, the DP2 path is worth the audit effort.
What the Requirement Actually Is
The objective
Develop a low-cost, small form factor, attritable passive radar sensor capable of detecting and tracking targets using external RF signals for tactical unmanned and expendable platforms.
The operational driver
In contested anti-access and area-denial environments, active radar emissions act as beacons for adversary electronic warfare. To ensure survivability, the Department requires covert sensing utilizing illuminators of opportunity.
This effort requires a 100 percent receive-only payload. The sensor shall not transmit any RF energy.
That last requirement is absolute and it is repeated in the Phase II deliverable definition. It has design consequences beyond the obvious. A receive-only system has no control over its illumination geometry, waveform, timing, or power, which means everything conventional radar gets for free must be recovered from the environment. It also means no active calibration, no transmitted reference, and no ability to choose a favorable pulse repetition frequency. And practically, "shall not transmit any RF energy" is worth interpreting conservatively: unintentional emissions from local oscillators, clocks, and digital processing are worth addressing in your design discussion, since an attritable sensor whose digital section radiates is not actually covert.
What is wrong with existing systems
Current passive systems are bulky, expensive, and primarily optimized for signals intelligence and electronic support measures rather than radar functionality. They lack the tightly synchronized, multi-channel receivers necessary for direct reference signal acquisition, spatial processing, and clutter cancellation.
The Department seeks a paradigm shift toward low-cost, multi-channel attritable, meaning expendable, sensors designed specifically for high-risk deployments.
The critique names three specific capabilities that SIGINT-derived hardware lacks: direct reference signal acquisition, spatial processing, and clutter cancellation, all of which depend on tight synchronization across multiple channels. Channel-to-channel phase coherence at low cost is the engineering crux of this topic, and a proposal that treats it casually has missed the point of the paragraph.
The system being sought
This topic seeks the development of a small form factor passive X-band radar featuring both synthetic aperture radar imaging and ground moving target indicator capabilities.
Optimized for extreme size, weight, power, and cost constraints, the payload must be light enough for integration onto Group 1 or 2 unmanned aircraft systems, loitering munitions, or unattended ground sensors.
Three things to notice.
X-band specifically. Passive radar research has historically leaned on VHF and UHF broadcast illuminators, FM radio and digital television, because they are powerful, continuous, and everywhere. X-band illuminators of opportunity are a different and harder problem: the available sources are other radars, satellite downlinks, and communications systems rather than broadcast transmitters. Your proposal has to identify which X-band illuminators you intend to exploit and what their availability and geometry look like in a contested environment. This is arguably the single most important technical judgment in the proposal.
Both SAR and GMTI. These place opposite demands on the collection. Synthetic aperture imaging needs coherent integration over a long aperture with precise platform motion knowledge. Moving target indication needs Doppler discrimination against clutter. Doing both from a bistatic or multistatic passive geometry, on a small platform, is a substantial signal processing burden, and the Phase I language acknowledges it by asking for algorithms that work "under constrained computational limits."
Group 1 or 2 UAS. Group 1 is under 20 pounds gross takeoff weight and Group 2 runs to 55 pounds. That is the payload envelope, and it is severe for a multi-channel coherent receiver plus antennas plus edge processing. Loitering munitions and unattended ground sensors imply the same or tighter.
Phase I Requirements
The objective of Phase I is to conduct a feasibility study to determine the scientific, technical, and commercial merit of a small form factor, attritable passive radar architecture.
Offerors are expected to define a comprehensive system architecture, identifying critical low-cost receiver components, antenna configurations suitable for small form factor platforms, and highly efficient edge-processing hardware.
The effort will include developing proof-of-concept signal processing algorithms for reference signal isolation, clutter cancellation, and target detection and tracking under constrained computational limits.
The desired end product is a robust simulation and a basic laboratory prototype that validates the feasibility of detecting targets using signals of opportunity while strictly adhering to the SWaP-C metrics required for expendable systems.
Reading the Phase I scope
Four architecture elements are named: low-cost receiver components, antenna configurations for small platforms, edge-processing hardware, and three specific algorithms. The three algorithms are reference signal isolation, clutter cancellation, and target detection and tracking, and they correspond directly to the three capabilities the topic says existing SIGINT hardware lacks.
The end product is a robust simulation plus a basic laboratory prototype. Both, not either. And the phrase "strictly adhering to the SWaP-C metrics required for expendable systems" means your Phase I has to commit to numbers: mass, volume, power, and unit cost. A feasibility study that establishes technical performance without bounding cost has not addressed attritability, which is the topic's defining constraint.
Phase II Requirements
Phase II culminates in the delivery of a viable hardware prototype.
Offerors are expected to transition their Phase I concepts into a physical, highly miniaturized sensor prototype suitable for integration onto a small unmanned aircraft system or expendable ground node.
The minimum required deliverable is a functional prototype that successfully demonstrates passive stationary and moving target detection, tracking, and imaging in a realistic outdoor environment utilizing RF illumination, cooperative or non-cooperative signals of opportunity, without emitting any RF energy from the passive sensor itself.
The performer must characterize system performance, demonstrate real-time edge processing that outputs actionable products, and validate that the unit cost and SWaP footprint meet the criteria for a truly attritable tactical asset.
Unpacking the minimum deliverable
Six things are packed into that one sentence, and each is a separate test.
Stationary target detection. Moving target detection. Tracking. Imaging. In a realistic outdoor environment. Without transmitting.
Note "cooperative or non-cooperative signals of opportunity." Cooperative illumination is permitted for the demonstration, which is an important allowance: you may use a known, controlled X-band source rather than depending on whatever happens to be radiating over your test range. That said, the operational value of the system depends on non-cooperative sources, so a demonstration that only works with a cooperative illuminator should be framed honestly as a step, with the non-cooperative case addressed in your risk discussion.
Three additional validation requirements follow. Characterize system performance, which means quantitative detection, resolution, and accuracy figures rather than imagery alone. Demonstrate real-time edge processing that outputs actionable products, meaning the processing happens on the platform and produces something a user can act on, not a raw data dump for later analysis. And validate that the unit cost and SWaP footprint meet the criteria for a truly attritable tactical asset, which is a cost claim you must substantiate, not assert.
That last one deserves emphasis. "Validate that the unit cost meets the criteria for a truly attritable tactical asset" is a bill of materials and manufacturing cost analysis, delivered as part of a technical program. Build the cost model early and update it as the design matures.
Phase III Dual Use
In Phase III, the expendable passive radar sensor is expected to transition into operational military platforms and commercial markets, supported by non-SBIR and non-STTR funding.
For Department applications, the technology will be integrated into launched effects, loitering munitions, and distributed unattended ground sensor networks.
In the commercial sector, potential applications include low-cost air traffic monitoring, counter-unmanned aircraft surveillance for critical infrastructure, and general airspace monitoring where adding active RF spectrum clutter is undesirable or prohibited.
The commercial case here is stronger than it first appears, and the reason is spectrum. A receive-only sensor needs no transmit authorization, which removes the regulatory obstacle that limits deployment of active radar for counter-drone and airspace monitoring at airports, stadiums, prisons, substations, and data centers. The topic names that advantage directly: environments "where adding active RF spectrum clutter is undesirable or prohibited." For a commercialization strategy, that is a concrete, defensible market entry argument rather than a generic dual-use paragraph.
Launched effects is worth noting on the military side. It is a current Army program area with active procurement, and a passive sensing payload for launched effects has an identifiable transition path.
Funding, Cost Structure, and OSW-Reliance 21 Mechanics
The awards
Phase I: $314,363 over 12 months. The Phase I base amount must not exceed the base limit set by the topic.
Direct to Phase II: $2,095,748 over 24 months.
Both figures are oddly precise, which usually indicates a specific funding line rather than a round programmatic allocation. Treat them as ceilings, not targets, and build a budget that fits.
Cost volume mechanics
OSW-Reliance 21 requires the use of the DSIP online Cost Volume webform. No separate Excel template is required. If supplementary cost detail is desired, it may be uploaded as a PDF attachment within Volume 3.
For Direct to Phase II, a detailed cost volume must be submitted online in the proper format shown in the Cost Breakdown Guidance in the DoW 2026 SBIR BAA. Provide enough information to allow evaluators to assess your plans to use the requested funds.
Percentage of Work, with no exceptions
Review the updated Percentage of Work calculation details included in the DoW solicitation. OSW-Reliance 21 will not accept any deviation to the POW requirements.
This matters on a radar topic, where the temptation is to subcontract RF front-end design, antenna design, or embedded processing. Model the POW before you assemble the team.
Technical and Business Assistance
Phase I awardees may request up to $6,500 in TABA funding. Phase II awardees may request up to $50,000 per Phase II project. TABA funding is in addition to the Phase I and Phase II cost ceilings and is not subject to profit or fee.
All requests for TABA must be completed using the SBIR/STTR TABA Request Form, and the completed form must be included in Volume 5 of the proposal submission in DSIP. OSW will not accept requests for TABA that do not utilize the form or that are not included as a submission document in Volume 5.
The $50,000 Phase II figure is at the high end across components in this cycle, and for this topic the highest-value uses are export control counsel, given the ITAR restriction, and manufacturing cost engineering, since validating attritable unit cost is a contract deliverable.
Page limits and the no-appendix rule
For Phase I, the technical volume must follow the DoW SBIR solicitation formatting requirements except that all figures, tables, charts, and references must be included within the page count limit listed in the topic index. Any pages past the technical volume limit will not be considered, and no separate appendices will be evaluated.
For Direct to Phase II, the technical volume is 20 pages maximum: Part 1 Phase I Justification at 5 pages maximum, and Part 2 Phase II Technical Proposal at 15 pages maximum. Within that 15 pages, the Technology Transition and Commercialization Strategy is not to exceed 2 pages and counts toward the 15-page limit.
So a DP2 proposer has 5 pages to establish feasibility, 13 pages of technical proposal, and 2 pages of commercialization strategy. That is tight for a system with two radar modes, a multi-channel receiver, an antenna, edge processing, and a cost model. Plan the page budget before drafting.
What the DP2 technical proposal must contain
The Phase II Technical Objectives and Approach section must list specific technical objectives and provide a detailed technical approach, and it must include the following named subsections.
Phase II Work Plan, with an explicit, detailed description of the approach, indicating what is planned, how and where the work will be carried out, a schedule of major events, and the final product to be developed.
Related Work, describing significant activities directly related to the effort including those of the Principal Investigator, the firm, consultants, or others, and demonstrating the proposers' awareness of the state of the art.
Relationship with Future Research or Research and Development, stating anticipated results and discussing the significance of the Phase II effort as a foundation for Phase III.
Technology Transition and Commercialization Strategy, not to exceed 2 pages and counting toward the 15-page limit, addressing five specific questions: what is the first product this technology will go into; who will be your customers and what is your estimate of the market size; how much funding will you need to bring the technology to market and how will you raise it; does your company contain marketing expertise and if not how will you bring it in; and who are your competitors and what is your price or quality advantage.
Key Personnel, including the Principal Investigator, with directly related education, experience, and relevant publications, and a concise resume of the PI.
Facilities and Equipment, describing available instrumentation and physical facilities, justifying equipment to be purchased including Government Furnished Equipment, and stating whether the facilities meet federal, state, and local environmental laws and regulations across the named groupings. All requirements for government furnished equipment or other assets, and associated costs, must be determined and agreed to during Phase II contract negotiations.
Consultants, describing in detail any involvement of universities, academic institutions, or other consultants and identifying them in the Cost Volume.
Those five commercialization questions are literal and a reviewer will look for all five. Answer them as five answers, not as a narrative that touches on them.
The Company Commercialization Report, and a contradiction
Completion of the CCR as Volume 4 is required.
The Phase I Proposal Guidelines state that information contained in the CCR will be considered by OSW-Reliance 21 during proposal evaluations. The Direct to Phase II Proposal Guidelines in the same document state that the information contained in the CCR will not be considered by "SCO" during proposal evaluations.
The two statements conflict, and the reference to SCO appears to be residual text from another organization's instructions. The safe approach for either path is to complete the CCR carefully and completely, since at least one section of the governing document says it is scored. If the answer materially affects your proposal, raise it through DSIP Topic Q&A before it closes.
Note also that the commercialization strategy in Volume 2 is separate from the CCR. The strategy addresses how you propose to commercialize this research; the CCR covers what you have done to commercialize the results of past Phase II awards.
Supporting documents that are optional but encouraged
Letters of Support from prospective transition stakeholders within DEVCOM C5ISR Center, PAE Maneuver Ground, PAE Maneuver Air, CPE Autonomy, the Naval Research Laboratory, or the Air Force Research Laboratory.
A Data Management Plan addressing provenance, licensing, and protection of pre-training data and government-furnished data.
For this topic the named list is unusually well matched. DEVCOM C5ISR Center is the Army's sensors and electronic warfare organization and is the natural home for a passive radar payload. PAE Maneuver Air and CPE Autonomy connect to the unmanned platform side. A letter from C5ISR is the highest-value optional document here.
If your signal processing uses machine learning trained on RF data, the Data Management Plan is directly relevant rather than boilerplate, and addressing provenance and licensing of your training data is worth doing.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW solicitation.
Government technical evaluators from Army DEVCOM C5ISR Center, Army DEVCOM Army Research Lab, the Naval Research Laboratory, and the Air Force Research Laboratory may participate in the evaluation. Non-government support contractors may assist in administrative handling of proposals if the individual has signed a non-disclosure agreement, and they will not participate in selection decisions.
Proposing firms will be notified of selection or non-selection status within 90 days of the closing date of the topic, which is approximately January 19, 2027. Notifications will be issued through DSIP to both the Corporate Official and the Principal Investigator listed on the proposal.
Protests after award should be submitted, as prescribed in FAR 33.106(b) and FAR 52.233-3, to osd.ncr.ousd-r-e.mbx.SBIR-STTR-Protest@mail.mil. Refer to the DoW solicitation for procedures to protest the announcement itself.
Tri-service coordination and the TPOC question
This topic is of joint interest to the U.S. Army, meaning DEVCOM C5ISR Center and DEVCOM ARL, the U.S. Navy, meaning the Naval Research Laboratory, and the U.S. Air Force and Space Force, meaning the Air Force Research Laboratory. Proposers are strongly encouraged to engage the Technical Point of Contact listed in the topic description during the pre-release period to discuss technical scope and transition opportunities across the Services.
No Technical Point of Contact appears in the DV025 topic description, or in any of the four topic descriptions in this release. Use DSIP Topic Q&A, and send administrative questions to osd.pentagon.ousd-atl.mbx.communities-of-interest@mail.mil.
Note that Phase II efforts under this release shall include a transition plan addressing at least two of the three Services. On this topic that is a natural fit, since passive radar payloads have Army, Navy, and Air Force applications, but it is a stated requirement rather than a suggestion.
Classification
Phase I efforts are expected to be performed at the Unclassified and CUI level. Classified proposals are not accepted, and including classified data in an unclassified proposal may be grounds for the Agency to determine the proposal non-responsive and not evaluate it.
In some instances, work being performed on Phase II contracts will require security clearances. If a Phase II contract requires classified work, the offeror must have a facility clearance and appropriate personnel clearances.
For a passive radar topic touching electronic warfare survivability and specific illuminators of opportunity, this possibility is real. Be careful about what you write in an unclassified proposal regarding adversary systems and specific X-band sources, and understand that a facility clearance may become necessary.
Export control and foreign nationals
The technology within this topic is restricted under ITAR, 22 CFR Parts 120-130, or EAR, 15 CFR Parts 730-774.
Offerors must disclose any proposed use of foreign nationals, their countries of origin, the type of visa or work permit possessed, and the statement of work tasks intended for accomplishment by the foreign nationals, in accordance with the Announcement. Foreign national participation will be evaluated on a case-by-case basis and may be restricted due to U.S. export control laws.
Radar signal processing talent is internationally distributed and small radar firms often employ foreign nationals on visas. Resolve this before submitting, with named individuals, countries, visa types, and specific task assignments. Note that the case-by-case language is more accommodating than a flat bar, but the disclosure has to be specific to be evaluated.
The References
Only two, and both are textbooks rather than papers.
Melvin and Scheer, editors, "Principles of Modern Radar: Radar Applications," Volume 3, SciTech Publishing and IET, 2014.
Moo and Ding, "Adaptive Radar Resource Management," Academic Press and Elsevier, 2015.
The brevity is itself informative. Unlike the DARPA topics in this cycle, which cite recent primary literature to point at a specific technical approach, this topic cites the standard reference works. The Government is not steering you toward a particular passive radar technique. It is stating a system requirement and leaving the architecture open.
That has a practical consequence: you will not gain credit for aligning with a cited approach, and you carry the full burden of justifying your architecture from first principles. It also means the "Related Work" section of a DP2 proposal, where you must demonstrate awareness of the state of the art, is doing more work than usual, since the topic gives you no scaffolding. Cite the real passive radar and bistatic literature yourself.
The Moo and Ding reference on adaptive radar resource management is a slightly curious inclusion for a receive-only system, since resource management usually concerns allocating transmit resources. The plausible reading is that it points at managing limited receive and computational resources under constraint, which is consistent with the Phase I emphasis on algorithms that work under constrained computational limits.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
DSIP Topic Q&A closes: October 7, 2026, two weeks before the topic closes, per the DoW SBIR Program BAA
Proposal deadline: October 21, 2026
Selection notification: within 90 days of the closing date, approximately January 19, 2027
Phase I period: 12 months from award
Direct to Phase II period: 24 months from award
A working backward plan
Before September 23. Decide Phase I or Direct to Phase II. If DP2, audit the funding provenance of every result you intend to cite as feasibility evidence, because work based upon or logically extending from prior or ongoing federally funded SBIR or STTR work is excluded, and failing the feasibility bar means the proposal is not evaluated at all. Confirm the work was substantially performed by your firm or your PI, and resolve IP ownership or licensing. Identify which X-band illuminators of opportunity you intend to exploit and be ready to defend their availability and geometry. Build a preliminary bill of materials and unit cost model, since attritability is the defining requirement. Model your Percentage of Work before finalizing subcontracts. Reach out to DEVCOM C5ISR Center and the other named organizations about a letter of support. Resolve foreign national participation questions with named individuals, countries, visas, and tasks. Confirm SAM registration and CMMC Level 2 self-assessment in SPRS. Download the DoW SBIR Program BAA and its Appendix A Phase I template if bidding Phase I.
September 23 through October 5. Draft the technical volume. For Phase I, 20 pages against the DoW Appendix A template covering the system architecture, low-cost receiver components, antenna configurations, edge processing hardware, the three named algorithms, the simulation plan, the laboratory prototype plan, and the SWaP-C metrics. For DP2, 5 pages of feasibility justification, 13 pages of technical proposal covering the work plan, related work, and future research relationship, and 2 pages of commercialization strategy answering all five named questions. Address the multi-channel synchronization problem explicitly, since it is the capability the topic says existing systems lack. Draft the 3,000 character cover sheet abstract and the 3,000 character anticipated benefits and commercial applications discussion.
October 6 through October 7. Submit remaining questions through DSIP Topic Q&A before it closes, including the CCR evaluation discrepancy if it affects you.
October 8 through October 14. Build the cost volume in the DSIP online webform, following the Cost Breakdown Guidance in the DoW 2026 SBIR BAA. Price receiver hardware, antenna fabrication, edge processing hardware, anechoic or outdoor test range access, and for DP2 the realistic outdoor demonstration campaign. Identify any Government Furnished Equipment requirements, remembering those must be determined and agreed during contract negotiations. Add supplementary cost detail as a Volume 3 PDF if useful. Complete the SBIR/STTR TABA Request Form and place it in Volume 5.
October 15 through October 18. Complete Volume 4, the Company Commercialization Report, carefully. Assemble Volume 5 with the TABA form, letters of support, and a Data Management Plan if your processing uses trained models. Complete Volume 6 training and the Volume 7 foreign affiliations webform, remembering that Volume 7 must be the webform and will not be accepted as a PDF in Volume 5, and that no previous versions should be uploaded to Volume 5. Run compliance: page limits with figures, tables, charts, and references counted inside, no appendices, unclassified or CUI only, no classified data.
October 19 through October 20. Submit and certify in DSIP.
Frequently Asked Questions
What is OSW-Reliance 21 SBIR topic OSW26BZ06-DV025?
OSW26BZ06-DV025 is an OSW-Reliance 21 SBIR topic titled "Compact Passive Radar," released under the 2026 SBIR Broad Agency Announcement, Release 6. The objective is to develop a low-cost, small form factor, attritable passive radar sensor capable of detecting and tracking targets using external RF signals for tactical unmanned and expendable platforms.
Can I submit either a Phase I or a Direct to Phase II proposal?
Yes. DV025 is the only topic in this release that appears in both the Phase I table and the Direct to Phase II table of the topic index. Phase I is $314,363 over 12 months. Direct to Phase II is $2,095,748 over 24 months. Both have a 20-page technical volume limit.
How much funding is available?
Phase I is $314,363 over 12 months. Direct to Phase II is $2,095,748 over 24 months. Phase I awardees may request up to $6,500 in TABA and Phase II awardees up to $50,000 per project, both in addition to the cost ceilings and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5.
When is the proposal deadline?
The topic opens September 23, 2026 and proposals are due October 21, 2026 through the Defense SBIR/STTR Innovation Portal at dodsbirsttr.mil.
Can my feasibility evidence come from a prior SBIR award?
No, and this is the most consequential restriction on the DP2 path. Feasibility documentation cannot be based upon or logically extend from any prior or ongoing federally funded SBIR or STTR work. Work submitted must have been substantially performed by the proposer or the Principal Investigator. If the technology is subject to intellectual property, you must own the IP or have obtained license rights prior to proposal submission. Note that the Volume 2 instruction phrases this as "must not be solely based on" prior SBIR or STTR work, which is a weaker formulation than the DP2 guidelines language; plan against the stricter reading.
What happens if my feasibility documentation is inadequate?
If the proposer fails to demonstrate technical merit and feasibility equivalent to the Phase I level as described in the topic, the related Phase II proposal will not be evaluated.
How is the Direct to Phase II technical volume structured?
Twenty pages maximum, divided into Part 1 Phase I Justification at 5 pages maximum and Part 2 Phase II Technical Proposal at 15 pages maximum. Within the 15 pages, the Technology Transition and Commercialization Strategy is not to exceed 2 pages and counts toward that limit.
Can the sensor transmit at all?
No. The topic states that this effort requires a 100 percent receive-only payload and that the sensor shall not transmit any RF energy. The Phase II deliverable definition repeats the constraint, requiring demonstration without emitting any RF energy from the passive sensor itself.
What band is required?
X-band. This is a meaningful constraint, since much passive radar work relies on VHF and UHF broadcast illuminators. X-band illuminators of opportunity are other radars, satellite downlinks, and communications systems rather than broadcast transmitters, so identifying and justifying your illuminator set is a central part of the proposal.
What radar modes are required?
Both synthetic aperture radar imaging and ground moving target indication. These place different demands on the collection and processing, and the Phase I language asks for algorithms that work under constrained computational limits.
What platforms must the payload fit?
Group 1 or 2 unmanned aircraft systems, loitering munitions, or unattended ground sensors. Group 1 is under 20 pounds gross takeoff weight and Group 2 runs to 55 pounds, which sets a severe envelope for a multi-channel coherent receiver plus antennas plus edge processing.
What is wrong with existing passive radar systems?
They are bulky, expensive, and primarily optimized for SIGINT and electronic support measures rather than radar functionality. Specifically, they lack the tightly synchronized, multi-channel receivers necessary for direct reference signal acquisition, spatial processing, and clutter cancellation. Low-cost channel-to-channel phase coherence is the engineering crux of the topic.
What does Phase I have to deliver?
A comprehensive system architecture identifying critical low-cost receiver components, antenna configurations suitable for small form factor platforms, and highly efficient edge-processing hardware. Proof-of-concept signal processing algorithms for reference signal isolation, clutter cancellation, and target detection and tracking under constrained computational limits. And an end product consisting of a robust simulation plus a basic laboratory prototype that validates feasibility while strictly adhering to the SWaP-C metrics required for expendable systems.
What does Phase II have to deliver?
A functional, highly miniaturized hardware prototype demonstrating passive stationary and moving target detection, tracking, and imaging in a realistic outdoor environment using cooperative or non-cooperative signals of opportunity, without emitting any RF energy. The performer must also characterize system performance, demonstrate real-time edge processing that outputs actionable products, and validate that the unit cost and SWaP footprint meet the criteria for a truly attritable tactical asset.
Can I use a cooperative illuminator for the demonstration?
Yes. The Phase II deliverable permits cooperative or non-cooperative signals of opportunity. Since operational value depends on non-cooperative sources, a demonstration relying on cooperative illumination should be framed as a step, with the non-cooperative case addressed in your risk discussion.
Do I have to prove the cost?
Yes. Validating that the unit cost and SWaP footprint meet the criteria for a truly attritable tactical asset is a stated Phase II requirement. That is a bill of materials and manufacturing cost analysis delivered as part of a technical program, so build the cost model early.
Is this topic ITAR restricted?
Yes. The technology is restricted under ITAR, 22 CFR Parts 120-130, or EAR, 15 CFR Parts 730-774. Offerors must disclose any proposed use of foreign nationals, their countries of origin, visa or work permit type, and the specific statement of work tasks assigned to each. Foreign national participation is evaluated case by case and may be restricted.
What CMMC level applies?
The projected requirement for this topic is CMMC Level 2 with self-assessment.
What cost volume format do I use?
The DSIP online Cost Volume webform. OSW-Reliance 21 does not require a separate Excel template. Supplementary cost detail may be uploaded as a PDF attachment within Volume 3. For Direct to Phase II, follow the Cost Breakdown Guidance in the DoW 2026 SBIR BAA.
Are there Percentage of Work restrictions?
Yes. OSW-Reliance 21 will not accept any deviation to the Percentage of Work requirements described in the DoW solicitation. Model your POW before subcontracting RF front-end, antenna, or embedded processing work.
Is the Company Commercialization Report evaluated?
The document conflicts with itself. The Phase I Proposal Guidelines state that CCR information will be considered by OSW-Reliance 21 during proposal evaluations. The Direct to Phase II Proposal Guidelines state that it will not be considered by "SCO," which appears to be residual text from another organization's instructions. Complete the CCR carefully either way, and raise the discrepancy through DSIP Topic Q&A if it matters to you.
What must the commercialization strategy address?
Five specific questions, in no more than 2 pages counting toward the 15-page Phase II technical proposal limit. What is the first product this technology will go into. Who will be your customers and what is your estimate of the market size. How much funding will you need to bring the technology to market and how will you raise those funds. Does your company contain marketing expertise and if not how do you intend to bring it in. Who are your competitors and what is your price or quality advantage.
How do I request TABA?
Using the SBIR/STTR TABA Request Form, included in Volume 5 of the DSIP submission. OSW will not accept TABA requests that do not use the form or that are not submitted in Volume 5.
What optional documents help?
Letters of support from prospective transition stakeholders within DEVCOM C5ISR Center, PAE Maneuver Ground, PAE Maneuver Air, CPE Autonomy, the Naval Research Laboratory, or the Air Force Research Laboratory. For this topic DEVCOM C5ISR Center is the most natural fit. A Data Management Plan addressing provenance, licensing, and protection of pre-training and government-furnished data is also encouraged and is directly relevant if your signal processing uses trained models.
Who evaluates my proposal?
Government technical evaluators from Army DEVCOM C5ISR Center, Army DEVCOM Army Research Lab, the Naval Research Laboratory, and the Air Force Research Laboratory may participate. Non-government support contractors may assist with administrative handling under a non-disclosure agreement but do not participate in selection decisions.
When will I hear back?
Within 90 days of the closing date of the topic, which is approximately January 19, 2027. Notifications go through DSIP to both the Corporate Official and the Principal Investigator listed on the proposal.
Does Phase II require a multi-service transition plan?
Yes. Phase II efforts under this release shall include a transition plan addressing at least two of the three Services. Phase II contracting actions are anticipated to be firm-fixed-price or cost-plus-fixed-fee at the discretion of the Contracting Officer.
Who is the technical point of contact?
The release strongly encourages engaging the Technical Point of Contact listed in the topic description during pre-release, but no TPOC appears in the DV025 description or in any of the four topic descriptions in this release. Use DSIP Topic Q&A, and send administrative questions to osd.pentagon.ousd-atl.mbx.communities-of-interest@mail.mil.
What is the commercial market?
Low-cost air traffic monitoring, counter-unmanned aircraft surveillance for critical infrastructure, and general airspace monitoring where adding active RF spectrum clutter is undesirable or prohibited. The regulatory advantage is real: a receive-only sensor needs no transmit authorization, which removes the main obstacle to deploying radar at airports, stadiums, substations, and similar sites.
Positioning Advice for Companies Considering This Topic
Audit your feasibility provenance before you decide on the DP2 path. This is the single most important thing on this topic. Feasibility documentation cannot be based upon or logically extend from any prior or ongoing federally funded SBIR or STTR work, and a proposal that fails the feasibility bar is not evaluated at all. Trace every result you intend to cite. Internally funded, privately funded, non-SBIR government contract, and commercial product work are all cleaner ground. If your passive radar hardware came out of a prior SBIR, seriously consider the Phase I path instead.
Make multi-channel synchronization the centerpiece. The topic tells you exactly what existing systems lack: tightly synchronized, multi-channel receivers for direct reference signal acquisition, spatial processing, and clutter cancellation. Achieving channel-to-channel phase coherence at attritable cost is the hard problem. Lead with your approach to it, with numbers on phase stability and the cost of the components that deliver it.
Name your X-band illuminators and defend them. This is where most proposals will be weakest. X-band is not the broadcast-illuminator regime, so you must say what you intend to exploit, whether other radars, satellite downlinks, or communications systems, and address availability, geometry, bandwidth, and what happens in a contested environment where the adversary's emissions are both your illuminator and a thing they may turn off. Treating illuminator availability as an assumption rather than an analysis is the most likely fatal weakness.
Take "shall not transmit any RF energy" literally and completely. Address unintentional emissions from local oscillators, clocks, and digital processing, not just the absence of a transmitter. An attritable sensor whose digital section radiates is not covert, and a reviewer thinking about survivability in an electronic warfare environment will notice if you only address the obvious reading.
Show the cost model, not a cost claim. Validating attritable unit cost is a stated Phase II requirement. A bill of materials at projected volume, with the receiver, antenna, and processing broken out, is far more persuasive than a target price. It also forces the design discipline the topic is asking for.
Budget the SAR and GMTI processing honestly against the platform. Both modes on Group 1 or 2 power and thermal budgets, in real time, producing actionable outputs, is a demanding computational claim. State your processing architecture, the operations per second, the power draw, and what you trade to fit. The Phase I language explicitly asks for algorithms that work under constrained computational limits, so acknowledging the constraint is expected rather than a weakness.
Write the platform integration story concretely. Group 1 or 2 UAS, loitering munitions, and unattended ground sensors are three different mechanical, thermal, and interface problems. Pick a primary and name it, ideally an actual platform, and show the mass, volume, and power budget against it.
Lead the commercial case with spectrum regulation. A receive-only sensor needs no transmit authorization. That single fact is why counter-drone and airspace monitoring at airports, stadiums, prisons, substations, and data centers is a reachable market for this technology and not for active radar. The topic names the advantage itself. Build the market size estimate on it.
Get a DEVCOM C5ISR letter of support. Letters are optional but encouraged, C5ISR is on the named list, it is the Army's sensors and electronic warfare organization, and its evaluators may be reviewing your proposal. Launched effects is a named Phase III transition target with an actual program behind it.
Plan the DP2 page budget before drafting. Five pages of feasibility, thirteen pages of technical proposal, two pages of commercialization, with all figures, tables, charts, and references inside the limits and no appendices evaluated. For a system with two radar modes, a coherent receiver, an antenna, edge processing, and a cost model, page discipline is a real design task.
Answer the five commercialization questions as five answers. They are enumerated in the instructions and a reviewer will look for each. First product, customers and market size, funding required and how raised, marketing expertise, competitors and your advantage. Do not bury them in a narrative.
Cite the real passive radar literature yourself. The topic cites only two textbooks, which means it gives you no scaffolding and you carry the full burden of demonstrating awareness of the state of the art, which the Related Work section explicitly requires. Bring the bistatic and passive coherent location literature to the proposal.
Be careful what you write about adversary systems. Classified proposals are not accepted and including classified data may make your proposal non-responsive. On a topic about surviving adversary electronic warfare and exploiting specific X-band sources, keep the unclassified proposal unclassified and note where a classified annex or a cleared discussion would follow in Phase II.
Model Percentage of Work first. No deviations are accepted, and radar development invites subcontracting the RF front end, the antenna, or the embedded software. Run the calculation before you commit to a team structure.
OSW-Reliance 21 SBIR OSW26BZ06-NV024: Engineered Microstructures for Enhanced IR Aperture Performance
Deadline: October 21, 2026
Funding Award Size: $300k
Description: Complete guide to OSW-Reliance 21 SBIR Phase I topic OSW26BZ06-NV024, engineered microstructures for IR-transparent composite ceramic apertures. $300K over 12 months. Closes October 21, 2026.
Quick Answer
OSW26BZ06-NV024 is a Phase I SBIR topic under the Office of the Secretary of War, Reliance 21, 2026 SBIR Broad Agency Announcement, Release 6. The Navy has already made an IR-transparent composite ceramic with sub-100 nanometer features in every phase and porosity below 0.1 percent, but only on parts about 0.4 inch across. This topic funds the scale-up: the same material properties on samples larger than 2 by 4 by 0.5 inches, with an industrial path to 3 by 9 inches or larger. The award is $300,000 over 12 months, and the technical volume is capped at 10 pages. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The core difficulty is stated plainly in the topic. Conventional ceramic processing routes that rely on mechanical mixing or co-precipitation of constituent phases typically result in microstructural coarsening during densification that precludes achieving sub-100 nm feature sizes in all phases. In other words, the standard way to make a dense ceramic destroys the nanostructure that makes this material valuable. Innovative powder synthesis and consolidation approaches are what the Navy is buying.
Note the 10-page technical volume. That is the tightest page limit across the four topics in this release and among the tightest in the entire 2026 cycle, and figures, tables, charts, and references all count inside it. No separate appendices will be evaluated. This is a materials science pitch that has to fit in ten pages including your micrographs.
Topic At a Glance
Topic number: OSW26BZ06-NV024
Title: Engineered Microstructures for Enhanced IR Aperture Performance
Agency: Office of the Secretary of War, Reliance 21, administered by the OUSW(R&E) SBIR Program
Solicitation: OSW-Reliance 21, 2026 SBIR Broad Agency Announcement, Release 6, Proposal Submission Instructions
Program type: Phase I
Base award: $300,000
Base period of performance: 12 months
Technical volume limit: 10 pages, with all figures, tables, charts, and references counted inside that limit
OUSW (R&E) Critical Technology Area: Scaled Hypersonics
Component Technology Priority Area: Hypersonics
Projected CMMC level requirement: Level 1
Export control status: ITAR restricted. The technology within this topic is restricted under the International Traffic in Arms Regulation, 22 CFR Parts 120-130, or the Export Administration Regulation, 15 CFR Parts 730-774
Target sample size: exceeding 2 by 4 by 0.5 inches, with a path toward 3 by 9 inches or larger
Microstructure requirement: sub-100 nanometer features in all phases
Porosity requirement: below 0.1 percent
Optical requirement: SWIR and MWIR transparency of at least 80 percent
Thermal stability requirement: retained microstructure and IR transparency after exposure to 1200 degrees C for 10 minutes
Thermal figure of merit: alpha over k at or below 1.0 micrometers per watt at 1000 degrees C, threshold, with a target of 0.5
Technical and Business Assistance: Phase I awardees may request up to $6,500, in addition to the cost ceiling and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5
Cost volume: the DSIP online Cost Volume webform is required. No separate Excel template
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026
Submission portal: DSIP at dodsbirsttr.mil
Keywords: composite ceramic, infrared transparent, nanoscale microstructure, ceramic scale-up, IR window, nanocomposite, thermal stability, thermal shock resistance
The Technical Problem, Stated Precisely
What the Navy has already done
The Navy has demonstrated within its laboratories that a composite ceramic exhibiting IR transparency can be produced with distinct phases, microstructural feature sizes below 100 nanometers in all phases, and porosity content below 0.1 percent on component parts as large as 0.4 inch in lateral dimension.
Phase identity, crystal structure, phase chemistry, microstructure feature size, and porosity content were established via X-ray diffraction and scanning electron microscopy cross-sections with energy dispersive spectroscopy analysis.
Two things follow from that paragraph. First, this is not a speculative material. It exists at small scale in a Navy laboratory, which means a reviewer has a physical reference point for what "good" looks like. Second, the characterization suite is named: XRD, plus SEM cross-sections with EDS. Use the same methods and report the same quantities, because that is how your result will be compared to theirs.
Why the properties matter
Retention of sub-100 nanometer feature sizes across all phases is critical to achieving the desired mechanical and thermal properties of the composite. Porosity above 0.1 percent similarly degrades mechanical performance.
All three specifications, meaning sub-100 nanometer microstructural features in all phases, porosity below 0.1 percent, and short-wave to mid-wave infrared transparency of at least 80 percent, must be achieved at scale for the material to be suitable for transition.
Read "all three" as a conjunction with no partial credit. A proposal that achieves the nanostructure but at 0.5 percent porosity, or achieves density but with coarsened grains in the second phase, has not met the requirement. The phrase "in all phases" appears repeatedly and is the specific thing that makes this hard: it is comparatively easy to keep one phase fine and much harder to keep every phase fine through densification.
The central obstacle
Producing IR-transparent composite ceramics meeting these specifications at sample sizes relevant to aperture applications is a significant technical challenge.
Conventional ceramic processing routes that rely on mechanical mixing or co-precipitation of constituent phases typically result in microstructural coarsening during densification that precludes achieving sub-100 nanometer feature sizes in all phases.
Innovative powder synthesis and consolidation approaches capable of satisfying all microstructural, porosity, and optical transparency requirements at sample sizes exceeding 2 by 4 by 0.5 inches are sought.
Two named villains, mechanical mixing and co-precipitation, and one named failure mechanism, coarsening during densification. Your proposal should say explicitly which of those you avoid and how. The reference list points at two routes worth knowing: Chaim and colleagues on sintering and densification of nanocrystalline ceramic oxide powders, which is the coarsening physics, and Katsui and Goto on coating ceramic powders by rotary chemical vapor deposition and sintering the coated powders, which is a specific synthesis strategy for keeping phases separated and fine.
Scalability, stated as a requirement rather than a hope
Furthermore, the proposed processes should be industrially scalable in both capacity and final component size, with a path towards producing components in the 3 by 9 inch range or larger.
Note both dimensions of scalability: capacity, meaning throughput, and final component size. A process that makes one good 2 by 4 inch plate per month is not industrially scalable in capacity, even if the part passes every specification. Address both.
Thermal requirements
In addition to these requirements, the composite ceramic must retain its microstructure and IR transparency after exposure to temperatures up to 1200 degrees C for 10 minutes. Phase transformation or grain growth during high-temperature service would degrade both mechanical and optical performance.
The composite must also exhibit a low ratio of thermal expansion coefficient, alpha, to thermal conductivity, k. The threshold requirement for alpha over k is 1.0 micrometers per watt and the target is 0.5 micrometers per watt at 1000 degrees C.
The alpha over k figure of merit is the thermal shock parameter in disguise. A low ratio means the material expands little for the heat it carries away, which is what keeps an aperture from fracturing when it heats rapidly. Since the Critical Technology Area is Scaled Hypersonics, the operative scenario is a window or dome on a vehicle experiencing severe aerothermal heating. Framing your material's alpha over k in those terms, rather than as an abstract ratio, connects your chemistry to the mission.
Note also that the 1200 degrees C exposure is specified as 10 minutes. That is a short-duration thermal excursion, not a soak, which again is consistent with a hypersonic flight profile rather than a furnace application.
What Phase I Requires
Phase I efforts should focus on research to establish the feasibility of a proposed material system and processing approach.
Proposals should outline a plan to investigate and demonstrate the potential to produce the target IR-transparent composite ceramic with sub-100 nanometer microstructural features in all phases and porosity below 0.1 percent.
The research should address initial process tuning, explore pathways to scalability, demonstrating the approach can plausibly be extended to industrially relevant sample sizes, and provide a scientific basis for meeting the optical and thermal property goals.
The applicant should also investigate the anticipated thermal stability of the proposed composition to 1200 degrees C and provide an estimate or measurement of the alpha over k ratio.
Reading the Phase I scope realistically
Note the verbs. Phase I asks you to investigate, explore, and provide a scientific basis. It asks for an estimate or measurement of alpha over k, which explicitly permits a calculated or literature-based estimate rather than a measured value. It asks you to demonstrate that scalability is plausible, not to demonstrate scale.
That is a well-scoped 12-month, $300,000 effort, and it tells you what not to over-promise. A Phase I proposal claiming it will deliver a 2 by 4 inch part meeting all specifications is not more competitive, it is less credible. The deliverable Phase I is buying is a defensible material system and process route with evidence that the nanostructure survives densification and a scientific argument that it scales.
One inconsistency in the source document
The Phase I section closes with this sentence: "For Direct to Phase II (D2P2) proposals, the applicant is required to provide details and documentation demonstrating accomplishments of a Phase I-type effort."
However, OSW26BZ06-NV024 appears only in the Phase I table of the topic index, at $300,000 over 12 months with a 10-page technical volume. It does not appear in the Direct to Phase II table, which lists only DV025 and DV027. The topic number itself carries the NV prefix that this release uses for Phase I topics.
The most likely explanation is that the D2P2 sentence is residual boilerplate. Anyone who believes they have Phase I-equivalent accomplishments and wants to pursue a Direct to Phase II award on this scope should confirm with the program office through DSIP Topic Q&A before investing in that path, because the topic index does not establish a DP2 award amount, duration, or page limit for this topic.
Phase II, For Planning Purposes
Phase II proposals under this release may only be submitted by Phase I awardees, by invitation. Understanding the Phase II scope now still matters, because your Phase I proposal is evaluated partly on whether it sets up a credible Phase II.
The Phase II effort should mature the proposed technology through focused research and development aimed at scaling the process and optimizing the material. Seven key research objectives are named, explicitly not limited to these.
Developing and refining a novel composition and process route with the goal of achieving sub-100 nanometer microstructural features in all phases, porosity below 0.1 percent, and SWIR and MWIR transparency of at least 80 percent on samples exceeding 2 by 4 by 0.5 inches.
Investigating and verifying the retention of nanoscale microstructural features, critical to the target mechanical and thermal properties, via SEM and EDS cross-section analysis.
Researching and demonstrating the industrial scalability of the process, including establishing a clear path towards producing components in the 3 by 9 inch range or larger.
Validating microstructural and optical stability, including maintained SWIR and MWIR transparency and retained sub-100 nanometer features without deleterious phase transformation, after exposure to 1200 degrees C.
Measuring key thermomechanical properties, specifically targeting an alpha over k ratio at or below the 1.0 micrometers per watt threshold, with a target of 0.5, at 1000 degrees C.
Performing and reporting initial mechanical property characterization, including hardness, flexural strength, and fracture toughness testing at ambient temperature.
Providing representative samples to the Navy for independent characterization, alongside a final cost estimate for continued scale-up and per-unit material cost.
Two Phase II items worth planning for now
Independent Navy characterization is the seventh objective and it changes how you should think about the whole program. Your samples will be measured by someone else with their own instruments. That argues for conservative claims, well-documented measurement conditions, and enough sample volume to send parts away without exhausting your inventory.
The cost estimate is also a named deliverable: a final cost estimate for continued scale-up and per-unit material cost. Aperture materials live or die on cost per part, and being asked for a per-unit number as a formal deliverable tells you the Navy is thinking about procurement, not just properties. Start tracking process cost in Phase I.
Note also that Phase II efforts under this release shall include a transition plan addressing at least two of the three Services, and that Phase II contracting actions are anticipated to be firm-fixed-price or cost-plus-fixed-fee at the discretion of the Contracting Officer.
Phase III Dual Use
The contractor will pursue commercialization of the materials and processes developed during Phase II with ITAR and CUI-eligible organizations.
IR-transparent composite ceramics with refined nanostructure have broad potential for dual-use applications including electro-optical and infrared sensor windows and domes for airborne and maritime platforms, hypersonic vehicle apertures, and commercial thermal imaging systems.
The technologies may be transitioned by expanding mission capabilities across a broad range of government users including directed energy, EO and IR sensing, and hypersonic systems programs.
Direct procurement of IR-transparent composite ceramic components in coordination with the government program manager may be part of a Phase III program.
Two observations. First, "with ITAR and CUI-eligible organizations" is a real constraint on your commercialization narrative. The natural customers for this material are export controlled, which narrows the addressable market but also raises barriers to entry, and a candid treatment of that tradeoff is more persuasive than a market size claim that ignores it.
Second, directed energy appears in the list of government users. Beam director windows for high-energy laser systems have the same combination of optical transmission, thermal shock, and large-aperture requirements, and Scaled Directed Energy is one of the six OSW-Reliance 21 Critical Technology Areas. If your material transmits usefully at directed energy wavelengths as well as SWIR and MWIR, that is a second transition path worth a sentence.
The Ten-Page Problem
This deserves its own section because it is the practical constraint that will shape your proposal more than anything else.
The technical volume for this topic is limited to 10 pages. Under the OSW-Reliance 21 instructions, all figures, tables, charts, and references must be included within the page count limits listed in the topic index. Any pages past the technical volume limit will not be considered, and no separate appendices will be evaluated.
Ten pages including micrographs, XRD patterns, property tables, and citations, for a materials proposal that must cover a composition, a synthesis route, a consolidation route, an optical property argument, a thermal property argument, a scalability argument, key personnel, and facilities.
Practical consequences. Every figure has to earn its space, which in practice means composite figures with multiple panels rather than one micrograph per page. Reference lists should be short and load-bearing rather than comprehensive. The scientific basis for optical transparency and for alpha over k can often be made in a paragraph with a citation rather than a derivation. And the Phase I work plan, which is what the Government is actually buying, should not be the section you compress to make room for background.
Follow all instructions under the Phase I Proposal Instructions section in the DoW SBIR Program BAA, and use the Phase I technical volume template provided as Appendix A in that BAA, since the OSW-Reliance 21 instructions defer to it on formatting apart from the figures-count-inside rule.
Funding, Cost Structure, and OSW-Reliance 21 Mechanics
The award
$300,000 over 12 months. The Phase I base amount must not exceed the base limit set by the topic.
Worth noting that the other two Phase I topics in this release are funded at $314,363, and NV024 at $300,000 is slightly lower. There is no stated reason for the difference. Do not read anything into it beyond the ceiling itself.
Cost volume mechanics, which differ from most components
OSW-Reliance 21 requires the use of the DSIP online Cost Volume webform. No separate Excel template is required. If supplementary cost detail is desired, it may be uploaded as a PDF attachment within Volume 3.
This is a genuine difference from components that mandate a downloadable spreadsheet, and it simplifies preparation. Provide enough information to allow evaluators to assess your plans to use the requested funds, following the Cost Breakdown Guidance in the DoW 2026 SBIR BAA.
Percentage of Work, with no exceptions
Review the updated Percentage of Work calculation details included in the DoW solicitation. OSW-Reliance 21 will not accept any deviation to the POW requirements.
That sentence is unusually flat. For a ceramics topic, the temptation is to subcontract powder synthesis, hot pressing or spark plasma sintering, or optical characterization to a university or a commercial lab. Model your POW before you build the team, because a plan that puts too much of the work outside your firm cannot be fixed by negotiation.
Technical and Business Assistance, which is unusually generous
The OSW SBIR/STTR Program will consider TABA requests in accordance with 15 U.S.C. 638(q). Phase I awardees may request up to $6,500 in TABA funding. Phase II awardees may request up to $50,000 per Phase II project. TABA funding is in addition to the Phase I and Phase II cost ceilings and is not subject to profit or fee.
The $50,000 Phase II figure is double what DARPA offers in the same cycle and is at the high end across components. If you expect to reach Phase II, that is a meaningful amount of non-dilutive assistance funding to plan around.
All requests for TABA must be completed using the SBIR/STTR TABA Request Form, and the completed form must be included in Volume 5 of the proposal submission in DSIP. OSW will not accept requests for TABA that do not utilize the form or that are not included as a submission document in Volume 5.
Read that twice if you intend to request TABA. The form is mandatory and it must be in Volume 5. A TABA request made any other way is not accepted.
Supporting documents that are optional but encouraged
Letters of Support from prospective transition stakeholders within DEVCOM C5ISR Center, PAE Maneuver Ground, PAE Maneuver Air, CPE Autonomy, the Naval Research Laboratory, or the Air Force Research Laboratory.
A Data Management Plan addressing provenance, licensing, and protection of pre-training data and government-furnished data.
The named organization list is worth reading as a map of who this program office talks to. For an IR aperture material, the Naval Research Laboratory is the obvious fit, and the topic description itself refers to Navy laboratory work and Navy independent characterization. A letter from an NRL stakeholder is the single highest-value optional document available on this topic.
The Data Management Plan language reads as written for artificial intelligence topics rather than for ceramics, but the underlying concerns still apply if your effort will handle government-furnished data or materials data with licensing constraints.
The Company Commercialization Report, and a contradiction worth knowing about
Completion of the CCR as Volume 4 is required.
The Phase I Proposal Guidelines section of this document states that information contained in the CCR will be considered by OSW-Reliance 21 during proposal evaluations. That is unusual. Many components, DARPA among them, explicitly exclude the CCR from evaluation. Here it counts, which means your prior Phase II commercialization history is part of your Phase I score.
Note that the Direct to Phase II Proposal Guidelines section of the same document says the opposite, stating that CCR information will not be considered by "SCO" during proposal evaluations. The reference to SCO appears to be residual text from another organization's instructions. For a Phase I proposal to NV024, the applicable statement is the one in the Phase I section: the CCR will be considered. Fill it out carefully and completely rather than treating it as a formality, and if the discrepancy matters to you, raise it through DSIP Topic Q&A.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW solicitation.
Government technical evaluators from Army DEVCOM C5ISR Center, Army DEVCOM Army Research Lab, the Naval Research Laboratory, and the Air Force Research Laboratory may participate in the evaluation. Non-government support contractors may assist in administrative handling of proposals if the individual has signed a non-disclosure agreement, and they will not participate in selection decisions.
Proposing firms will be notified of selection or non-selection status within 90 days of the closing date of the topic. Notifications will be issued through DSIP to both the Corporate Official and the Principal Investigator listed on the proposal.
Ninety days from October 21, 2026 is approximately January 19, 2027.
That evaluator list is tri-service and it tells you something about how to write. Your reviewer may be an Army materials scientist or an Air Force sensor engineer rather than a Navy ceramist, even though the topic originates in Navy work. Do not assume familiarity with the specific Navy laboratory result, and do not write only to a naval use case.
Protests after award should be submitted, as prescribed in FAR 33.106(b) and FAR 52.233-3, to osd.ncr.ousd-r-e.mbx.SBIR-STTR-Protest@mail.mil. Refer to the DoW solicitation for procedures to protest the announcement itself.
Tri-service coordination and the TPOC question
This topic is of joint interest to the U.S. Army, meaning DEVCOM C5ISR Center and DEVCOM ARL, the U.S. Navy, meaning the Naval Research Laboratory, and the U.S. Air Force and Space Force, meaning the Air Force Research Laboratory. Proposers are strongly encouraged to engage the Technical Point of Contact listed in the topic description during the pre-release period to discuss technical scope and transition opportunities across the Services.
One practical problem: no Technical Point of Contact appears in the NV024 topic description, or in any of the four topic descriptions in this release. Use DSIP Topic Q&A during the pre-release and open periods, and direct administrative questions about the program and these instructions to osd.pentagon.ousd-atl.mbx.communities-of-interest@mail.mil.
Classification
Phase I efforts are expected to be performed at the Unclassified and CUI level. Classified proposals are not accepted. The inclusion of classified data in an unclassified proposal may be grounds for the Agency to determine the proposal non-responsive and not evaluate it.
In some instances, work being performed on Phase II contracts will require security clearances. If a Phase II contract requires classified work, the offeror must have a facility clearance and appropriate personnel clearances to perform the classified work.
For a hypersonic aperture material, that Phase II possibility is not remote. If you do not hold a facility clearance, understand that it is a long lead item and start thinking about sponsorship early.
Export control and foreign nationals
The technology within this topic is restricted under ITAR, 22 CFR Parts 120-130, which controls the export and import of defense-related material and services including export of sensitive technical data, or EAR, 15 CFR Parts 730-774, which controls dual use items.
Offerors must disclose any proposed use of foreign nationals, their countries of origin, the type of visa or work permit possessed, and the statement of work tasks intended for accomplishment by the foreign nationals, in accordance with the Announcement.
Foreign national participation will be evaluated on a case-by-case basis and may be restricted due to U.S. export control laws.
Note the "case-by-case basis" language in the release's Additional Information section. That is more accommodating than a flat prohibition, but it means disclosure has to be specific: which person, which country, which visa, which tasks. If your ceramics expertise sits with a foreign national researcher or a university collaborator, resolve this before submitting rather than hoping it goes unexamined.
The References
Only four, and each maps to a distinct part of the technical problem.
Chaim, Levin, Shlayer, and Estournes, "Sintering and densification of nanocrystalline ceramic oxide powders: a current understanding," Advances in Applied Ceramics, 2008. This is the coarsening problem itself, and it is the paper that explains why the topic exists.
Goldstein, "Correlation between MgAl2O4-spinel structure, properties and scale-up product performance in IR-windows and domes," Optical Materials, 2012. Spinel is the incumbent transparent ceramic for IR windows and domes, and this reference is about scale-up performance specifically. Treat spinel as the baseline your material must beat, and say how.
Nordahl, Hartnett, Gattuso, and Gentilman, "Optical and Mechanical Properties of Nano-Composite Optical Ceramics," Raytheon Integrated Defense Systems, DTIC accession ADA527006, 2009. This is prior nanocomposite optical ceramic work from a defense prime, retrievable from DTIC, and it is the closest published precedent to what the Navy has demonstrated.
Katsui and Goto, "Coatings on ceramic powders by rotary chemical vapor deposition and sintering of the coated powders," Journal of the Ceramic Society of Japan, 2018. This is a specific synthesis strategy for the exact problem: coat the powder so the phases stay separated and fine through densification.
The set is short enough that all four are worth reading before you write. The Katsui and Goto route in particular is a strong hint about the kind of powder synthesis approach that would satisfy the "innovative powder synthesis and consolidation" language, and positioning your route relative to it, whether you use it, improve on it, or reject it, shows a reviewer you know the space.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
DSIP Topic Q&A closes: October 7, 2026, two weeks before the topic closes, per the DoW SBIR Program BAA
Proposal deadline: October 21, 2026
Selection notification: within 90 days of the closing date, approximately January 19, 2027
Period of performance: 12 months from award
A working backward plan
Before September 23. Identify your material system and be able to defend the composition choice against spinel as the incumbent. Assemble your existing micrographs, XRD patterns, and any porosity or transparency data, and select the few figures that will fit in ten pages. Confirm access to XRD and to SEM with EDS, since those are the named characterization methods. Estimate or find alpha and k for your composition and compute the ratio at 1000 degrees C, since Phase I explicitly accepts an estimate. Model your Percentage of Work before finalizing subcontracts, because no POW deviations are accepted. Read the four references. Reach out to Naval Research Laboratory and other named stakeholder organizations about a letter of support, which is optional but is the highest-value optional document here. Resolve any foreign national participation questions. Confirm SAM registration and your CMMC Level 1 posture. Download the DoW SBIR Program BAA Appendix A Phase I technical volume template.
September 23 through October 5. Draft the 10-page technical volume against the DoW Appendix A template. Structure it around the material system, the synthesis and consolidation route that avoids coarsening, the scientific basis for 80 percent SWIR and MWIR transmission, the thermal stability argument to 1200 degrees C, the alpha over k estimate, and the scalability pathway addressing both capacity and component size. Keep the Phase I work plan substantial rather than compressing it for background. Draft the 3,000 character cover sheet abstract and the 3,000 character anticipated benefits and commercial applications discussion.
October 6 through October 7. Submit any remaining questions through DSIP Topic Q&A before it closes, including the Direct to Phase II ambiguity if that path interests you and the CCR evaluation discrepancy if it affects you.
October 8 through October 14. Build the cost volume in the DSIP online webform. Price powder synthesis, consolidation runs, XRD and SEM with EDS characterization time, optical transmission measurement, thermal property measurement or modeling, and high-temperature exposure testing. Add supplementary cost detail as a PDF in Volume 3 if useful. Complete the SBIR/STTR TABA Request Form if you want the $6,500 and put it in Volume 5.
October 15 through October 18. Complete Volume 4, the Company Commercialization Report, carefully rather than perfunctorily, since the Phase I instructions state it will be considered during evaluations. Assemble Volume 5 with the TABA form, any letters of support, and a Data Management Plan if applicable. Complete Volume 6 training and the Volume 7 foreign affiliations webform, remembering that Volume 7 must be the webform and will not be accepted as a PDF in Volume 5, and that no previous versions of the form should be uploaded to Volume 5. Run compliance: 10 pages maximum with figures, tables, charts, and references counted inside, unclassified or CUI only, no classified data, no separate appendices.
October 19 through October 20. Submit and certify in DSIP.
Frequently Asked Questions
What is OSW-Reliance 21 SBIR topic OSW26BZ06-NV024?
OSW26BZ06-NV024 is a Phase I SBIR topic titled "Engineered Microstructures for Enhanced IR Aperture Performance," released under the Office of the Secretary of War, Reliance 21, 2026 SBIR Broad Agency Announcement, Release 6. The objective is to spur research into industrially scalable production methods for an IR-transparent composite ceramic exhibiting sub-100 nanometer microstructural features in all phases on samples larger than 2 by 4 by 0.5 inches, porosity below 0.1 percent, microstructural and optical stability to 1200 degrees C, and a thermal expansion to thermal conductivity ratio at or below 1.0 micrometers per watt at 1000 degrees C.
How much funding is available?
$300,000 over 12 months for Phase I. Phase I awardees may also request up to $6,500 in Technical and Business Assistance, which is in addition to the cost ceiling and not subject to profit or fee, using the mandatory SBIR/STTR TABA Request Form in Volume 5.
When is the proposal deadline?
The topic opens September 23, 2026 and proposals are due October 21, 2026 through the Defense SBIR/STTR Innovation Portal at dodsbirsttr.mil.
How long can my technical volume be?
Ten pages. That is the shortest limit of the four topics in this release. All figures, tables, charts, and references must be included within the page count. Any pages past the limit will not be considered, and no separate appendices will be evaluated.
What are the four hard specifications?
Sub-100 nanometer microstructural features in all phases. Porosity below 0.1 percent. Short-wave to mid-wave infrared transparency of at least 80 percent. And samples exceeding 2 by 4 by 0.5 inches. The topic states that all three of the microstructure, porosity, and transparency specifications must be achieved at scale for the material to be suitable for transition.
What has the Navy already demonstrated?
A composite ceramic with IR transparency, distinct phases, microstructural feature sizes below 100 nanometers in all phases, and porosity below 0.1 percent, on component parts as large as 0.4 inch in lateral dimension. Characterization was by X-ray diffraction and SEM cross-sections with EDS analysis. The gap this topic addresses is scale, not existence.
Why is scale-up hard?
Conventional ceramic processing routes relying on mechanical mixing or co-precipitation of constituent phases typically cause microstructural coarsening during densification, which precludes achieving sub-100 nanometer feature sizes in all phases. The topic seeks innovative powder synthesis and consolidation approaches that avoid that mechanism.
What size components does the Navy ultimately want?
The proposed processes should be industrially scalable in both capacity and final component size, with a path towards producing components in the 3 by 9 inch range or larger. Phase II targets samples exceeding 2 by 4 by 0.5 inches.
What is the thermal stability requirement?
The composite must retain its microstructure and IR transparency after exposure to temperatures up to 1200 degrees C for 10 minutes, without phase transformation or grain growth.
What is the alpha over k requirement?
A low ratio of thermal expansion coefficient to thermal conductivity. The threshold is 1.0 micrometers per watt and the target is 0.5 micrometers per watt at 1000 degrees C. This is effectively a thermal shock figure of merit for an aperture experiencing rapid aerothermal heating.
Do I have to measure alpha over k in Phase I?
No. Phase I asks the applicant to provide an estimate or measurement of the alpha over k ratio, which explicitly permits a calculated or literature-based estimate.
What is Phase I actually buying?
Feasibility of a material system and processing approach. Phase I asks you to investigate and demonstrate the potential to hit the microstructure and porosity targets, address initial process tuning, explore pathways to scalability by showing the approach can plausibly be extended to industrially relevant sizes, and provide a scientific basis for the optical and thermal property goals. It does not ask you to deliver a full-size part.
Can I submit a Direct to Phase II proposal for this topic?
The Phase I section of the topic contains a sentence referring to Direct to Phase II proposals, but NV024 appears only in the Phase I table of the topic index and not in the Direct to Phase II table, which lists only DV025 and DV027. No DP2 award amount, duration, or page limit is established for this topic. The D2P2 sentence appears to be residual boilerplate. Confirm through DSIP Topic Q&A before pursuing that path.
What does Phase II involve?
Scaling the process and optimizing the material, across seven named objectives: refining the composition and process route to hit all specifications on samples exceeding 2 by 4 by 0.5 inches; verifying nanoscale feature retention via SEM and EDS cross-sections; demonstrating industrial scalability with a path to 3 by 9 inches or larger; validating microstructural and optical stability after 1200 degrees C exposure; measuring alpha over k against the 1.0 threshold and 0.5 target at 1000 degrees C; initial mechanical characterization including hardness, flexural strength, and fracture toughness at ambient temperature; and providing representative samples to the Navy for independent characterization along with a final cost estimate for scale-up and per-unit material cost.
How do I get a Phase II award?
Phase II proposals may only be submitted by Phase I awardees. Submission notices are issued by the OSW SBIR/STTR Program Office to eligible Phase I performers through DSIP, with submission windows announced individually following Phase I final review. Notices go to the Corporate Official and Principal Investigator listed on the Phase I award. Phase II efforts shall include a transition plan addressing at least two of the three Services, and contracting actions are anticipated to be firm-fixed-price or cost-plus-fixed-fee at the Contracting Officer's discretion.
Will the Navy test my samples themselves?
Yes, in Phase II. Providing representative samples to the Navy for independent characterization is a named Phase II objective. Plan for conservative claims, well-documented measurement conditions, and enough sample volume to send parts out.
Is this topic ITAR restricted?
Yes. The technology is restricted under ITAR, 22 CFR Parts 120-130, or EAR, 15 CFR Parts 730-774. Offerors must disclose any proposed use of foreign nationals, their countries of origin, visa or work permit type, and the specific statement of work tasks assigned to each. Foreign national participation is evaluated case by case and may be restricted.
What CMMC level applies?
The projected requirement for this topic is CMMC Level 1.
Is the Company Commercialization Report evaluated?
For Phase I under this release, yes. The Phase I Proposal Guidelines state that information contained in the CCR will be considered by OSW-Reliance 21 during proposal evaluations. Note that the Direct to Phase II section of the same document states the opposite and refers to a different organization, which appears to be residual text. Treat the CCR as scored and complete it carefully.
What cost volume format do I use?
The DSIP online Cost Volume webform. OSW-Reliance 21 does not require a separate Excel template. Supplementary cost detail may be uploaded as a PDF attachment within Volume 3.
Are there Percentage of Work restrictions?
Yes, and they are strict. OSW-Reliance 21 will not accept any deviation to the Percentage of Work requirements described in the DoW solicitation. Model your POW before finalizing subcontracts for powder synthesis, consolidation, or characterization.
How do I request TABA?
Using the SBIR/STTR TABA Request Form, included in Volume 5 of the DSIP submission. OSW will not accept TABA requests that do not use the form or that are not submitted in Volume 5. Phase I is up to $6,500 and Phase II is up to $50,000 per project, both in addition to the cost ceilings and not subject to profit or fee.
What optional documents help?
Letters of support from prospective transition stakeholders within DEVCOM C5ISR Center, PAE Maneuver Ground, PAE Maneuver Air, CPE Autonomy, the Naval Research Laboratory, or the Air Force Research Laboratory. For this topic the Naval Research Laboratory is the most natural fit. A Data Management Plan addressing provenance, licensing, and protection of pre-training and government-furnished data is also encouraged.
Who evaluates my proposal?
Government technical evaluators from Army DEVCOM C5ISR Center, Army DEVCOM Army Research Lab, the Naval Research Laboratory, and the Air Force Research Laboratory may participate. Non-government support contractors may assist with administrative handling under a non-disclosure agreement but do not participate in selection decisions.
When will I hear back?
Within 90 days of the closing date of the topic, which is approximately January 19, 2027. Notifications go through DSIP to both the Corporate Official and the Principal Investigator listed on the proposal.
Can I submit a classified proposal?
No. Phase I efforts are expected to be performed at the Unclassified and CUI level, and classified proposals are not accepted. Including classified data in an unclassified proposal may make the proposal non-responsive. Some Phase II contracts may require facility and personnel clearances.
Who is the technical point of contact?
The release strongly encourages engaging the Technical Point of Contact listed in the topic description during the pre-release period, but no TPOC appears in the NV024 description or in any of the four topic descriptions in this release. Use DSIP Topic Q&A, and send administrative questions to osd.pentagon.ousd-atl.mbx.communities-of-interest@mail.mil.
What is the commercial market?
Electro-optical and infrared sensor windows and domes for airborne and maritime platforms, hypersonic vehicle apertures, and commercial thermal imaging systems. Government users named include directed energy, EO and IR sensing, and hypersonic systems programs. Note that commercialization is to be pursued with ITAR and CUI-eligible organizations, which constrains the market but also raises barriers to entry.
Positioning Advice for Companies Considering This Topic
Beat spinel explicitly. Magnesium aluminate spinel is the incumbent transparent ceramic for IR windows and domes, and the topic cites a paper specifically about spinel scale-up product performance. A reviewer will be holding spinel in mind whether you mention it or not. Say what your composite does that spinel cannot, in the specific terms the topic uses: feature size, porosity, transmission, thermal stability, and alpha over k.
Name your coarsening defense in the first page. The topic identifies mechanical mixing and co-precipitation as the routes that fail and coarsening during densification as the mechanism. Whatever your approach is, powder coating, in situ phase formation, field-assisted sintering, a second phase that pins boundaries, state it early and explain the physics. That single argument is the topic's whole technical premise.
Say "in all phases" as often as the topic does. Keeping one phase below 100 nanometers is achievable. Keeping every phase there through full densification is the hard requirement, and it is stated four separate times. A proposal that presents an average grain size rather than per-phase data has answered a different question.
Report porosity like it is a specification, not a byproduct. Below 0.1 percent is a demanding density target for a nanostructured multiphase ceramic. State the measurement method, whether Archimedes, image analysis, or another approach, and be candid about its resolution near 0.1 percent, because that is exactly where measurement uncertainty starts to matter.
Address scalability in both dimensions. Capacity and final component size are both named. A single-part-per-month process that makes beautiful plates fails the capacity half. Give a throughput estimate and a route to 3 by 9 inches or larger, including what equipment scale-up requires and what it costs.
Connect alpha over k to the mission. The Critical Technology Area is Scaled Hypersonics and the thermal exposure is 1200 degrees C for 10 minutes. That is a flight thermal transient, not a furnace soak. Framing your thermal shock argument against an aerothermal profile rather than as an abstract ratio shows you understand why the requirement exists.
Fit in ten pages by cutting background, not the work plan. Every figure, table, chart, and reference counts inside the limit and no appendices are evaluated. Use multi-panel composite figures, keep the reference list short and load-bearing, and protect the space for the Phase I research plan, which is what the Government is buying.
Get a Naval Research Laboratory letter if you can. Letters of support are optional but encouraged, NRL is on the named list, and this topic originates in Navy laboratory work with Navy independent characterization built into Phase II. It is the single highest-leverage optional document available here.
Write for a tri-service reviewer. Evaluators may come from Army DEVCOM C5ISR, DEVCOM ARL, NRL, or AFRL. Do not assume your reviewer knows the specific Navy result, and do not write only to a naval application when directed energy and hypersonic systems programs are both named as government users.
Fill out the Company Commercialization Report properly. The Phase I instructions say it will be considered during evaluations, which is not true of every component. If you have prior Phase II awards, their commercialization outcomes are part of your score here.
Model Percentage of Work before you build the team. No deviations are accepted, and ceramics work invites subcontracting synthesis, consolidation, and characterization. Run the calculation first.
Use the TABA form or lose the TABA. Six thousand five hundred dollars in Phase I and fifty thousand in Phase II are both meaningful, and both require the SBIR/STTR TABA Request Form in Volume 5. A request made any other way is not accepted.
Do not over-promise Phase I. The Phase I verbs are investigate, explore, and provide a scientific basis, and an estimate of alpha over k is acceptable. Claiming you will deliver a full-size qualified part in 12 months on $300,000 reads as inexperience rather than ambition.
Start the clearance conversation if you might need one. Phase II contracts under this release may require classified work with a facility clearance and cleared personnel. For a hypersonic aperture material, that is a plausible outcome, and a facility clearance is a long lead item.
DARPA STTR DPA26TZ06-DV006: Localization, Characterization, and Modeling of Freestream Disturbances in Hypersonic Wind Tunnels
Deadline: October 21, 2026
Funding Award Size: $2m
Description: Complete guide to DARPA STTR Direct to Phase II topic DPA26TZ06-DV006, localizing and modeling freestream disturbances in hypersonic wind tunnels. $750K plus $1.25M option. Closes October 21, 2026.
Quick Answer
DPA26TZ06-DV006 is a DARPA STTR Direct to Phase II topic under the DoW 2026 STTR Broad Agency Announcement, Release 6. DARPA wants diagnostics that can measure flow and acoustic disturbances in the parts of a hypersonic wind tunnel where you cannot put a window: the driver, the reservoir, and the nozzle throat. Then it wants those measurements tied, through multi-fidelity simulation, to the freestream noise that contaminates the test section downstream. The award is $750,000 over 12 months with a $1,250,000 option over 12 months. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The scientific problem is old and specific. Laufer established in 1964 that nozzle-wall turbulent boundary layers radiate downstream-directed Mach waves, making them a principal source of freestream noise. But in high-enthalpy and hypervelocity impulse facilities the noise field may be further enhanced by complex upstream dynamics that are not optically accessible. Standard optical techniques such as FLEET, focused laser differential interferometry, and Rayleigh scattering all require optical-grade line-of-sight window access, which may not withstand the extreme pressures, temperatures, or geometric constraints of the driver, reservoir, or nozzle throat.
Why it matters operationally: hypersonic ground test data is only as good as the tunnel's noise floor, and boundary-layer transition on a test article is exquisitely sensitive to freestream disturbance. If you cannot say where the noise comes from, you cannot correct for it, and every transition measurement carries an unquantified facility signature.
This topic is unusual in the release for two reasons. It is the only one with an OUSW Research and Engineering Critical Technology Area designation, Scaled Hypersonics. And its option period is larger than its base, $1,250,000 against $750,000, which means the majority of the money is behind a facility integration campaign that DARPA can decline to fund.
Topic At a Glance
Topic number: DPA26TZ06-DV006
Title: Localization, Characterization, and Modeling of Freestream Disturbances in Hypersonic Wind Tunnels
Agency: Defense Advanced Research Projects Agency (DARPA)
Solicitation: DoW 2026 Small Business Technology Transfer Broad Agency Announcement, Release 6, DARPA Proposal Submission Instructions
Program type: Direct to Phase II (DP2). This topic is soliciting Direct to Phase II proposals only
Technical volume format: Standard, 35 pages. Feasibility documentation is a 10-page volume, the technical proposal shall not exceed 20 pages, and the Phase II commercialization strategy shall not exceed 5 pages
Base award: $750,000
Base period of performance: 12 months
Option: $1,250,000 over 12 months
OUSW (R&E) Critical Technology Area: Scaled Hypersonics
Component Technology Priority Area: Hypersonics
Projected CMMC level requirement: Level 1
Export control status: ITAR restricted. The technology within this topic is restricted under the International Traffic in Arms Regulation, 22 CFR Parts 120-130, or the Export Administration Regulation, 15 CFR Parts 730-774
Feasibility gate: a 10-page volume providing written evidence of upstream flow and acoustic modeling, plus a proof-of-concept or benchtop demonstration of a confined-space diagnostic
Target regions: driver, reservoir, nozzle throat and walls, meaning the locations where optical access is restricted or unavailable
Base end state: Critical Design Review, with a benchtop diagnostic demonstrated at 100 kHz frequency response and structural and thermal certification by the target facility operators
Option end state: diagnostic suite integrated in a live facility at minimum Mach 5, with upstream fluctuations correlated to downstream freestream noise and a notional mitigation plan
Technical and Business Assistance: DARPA will provide up to $25,000 for the Direct to Phase II
Topic Q&A: DSIP Topic Q&A is not available for DARPA topics. Technical questions go to SBIR_BAA@darpa.mil by October 14, 2026
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Submission portal: DSIP at dodsbirsttr.mil
Keywords: hypersonics, wind tunnel noise, upstream disturbances, nozzle throat, driver-gas acoustics, non-optical diagnostics, high-frequency sensors, acoustic mitigation, fiber-optic probes, reflected-shock tunnels, expansion tubes
The Feasibility Bar, Which Is the First Thing to Check
This topic is soliciting Direct to Phase II proposals only. Proposers must submit a 10-page feasibility volume providing written evidence of two things.
Upstream flow and acoustic modeling
Successful modeling of upstream disturbance generation and propagation, for example driver-gas acoustic focusing, shock-tube boundary layer interactions, or nozzle throat shear layer acoustics.
Three example phenomena are named, and they correspond to three different facility classes. Driver-gas acoustic focusing is an expansion tube and tunnel problem. Shock-tube boundary layer interaction is a reflected-shock tunnel problem. Nozzle throat shear layer acoustics applies to all tunnels. Pick the phenomenon you have actually modeled, and be clear which facility class your model addresses, because that choice constrains the facility you will target in Phase II.
Confined-space diagnostic feasibility
Proof-of-concept or benchtop demonstration of a diagnostic method, for example fiber-optic-coupled probes, ultra-high-frequency flush-mounted sensors, acoustic emissions arrays, or shear sensors, demonstrating a frequency response and the physical ruggedness required to operate in high-pressure and high-temperature or non-optically accessible environments.
Note the two attributes: frequency response and physical ruggedness. Both must be demonstrated, not asserted. The Month 8 base milestone sets the specific bar at 100 kHz, which is a reasonable indication of what "high-frequency" means here, so a diagnostic with a few tens of kilohertz response has a gap to close.
What the gate screens for, and what makes this topic different
Notice what the gate does not require. There is no facility integration, no Mach number, no tunnel data. It asks for a model and a benchtop diagnostic. Compared to the other two topics in this STTR release, this is the most accessible feasibility bar, and it is well matched to a small diagnostics company partnered with a university hypersonics group.
The document also says the modeling and diagnostics work should come from "Collaborative proposals from Small Business Concerns," which is consistent with the STTR structure: the disturbance physics and simulation naturally sit with the research institution, and the sensor hardware with the small business.
Two Appendix A rules apply. Work submitted within the feasibility documentation must have been substantially performed by the proposer or the Principal Investigator, which on a university-partnered proposal means being careful about whose modeling results you are submitting and confirming that the arrangement satisfies the rule. And if the technology is subject to intellectual property, you must own the IP or have obtained license rights prior to proposal submission, with documentation in the Technical Volume.
What DARPA Is Actually Looking For
The objective
Develop and demonstrate a robust methodology to localize, characterize, and model upstream noise and flow disturbances in hypersonic wind tunnels, specifically targeting high-risk, hard-to-access regions upstream of the test section, for example the driver, reservoir, nozzle throat and walls, where traditional optical access is restricted or unavailable.
Performers must implement novel non-intrusive or minimally intrusive diagnostics coupled with multi-fidelity simulation strategies to accurately trace the evolution of unwanted thermo-fluid-dynamic disturbances from their point of origin to the test section.
The phrase "from their point of origin to the test section" is the whole topic in six words. This is a source localization and propagation tracing problem, not a noise measurement problem. Measuring test section noise is already routine. Saying which upstream feature produced it is not.
The problem
Hypersonic ground-test facilities are critical for evaluating aerodynamic forces, aerothermodynamic heating, and boundary-layer transition behavior on high-speed flight vehicles. However, conventional hypersonic wind tunnels are plagued by high-intensity freestream noise present in the test section.
Because these disturbances build, focus, and propagate downstream, the fundamental understanding of their evolution may require a potentially deep analysis within the upstream components of the facility.
Historically, Laufer established that nozzle-wall turbulent boundary layers radiate downstream-directed Mach waves, making them a principal source of freestream noise. However, especially in high-enthalpy and hypervelocity impulse facilities, the noise field may be further enhanced by complex upstream dynamics that are not optically accessible.
The three named focus areas
DARPA lists examples of focus areas, explicitly not limited to these, organized by facility type.
Nozzle throats and walls, in all tunnels. The extreme thermal and velocity gradients in the nozzle throat pose an opportunity for disturbance generation and initiation of turbulence over the nozzle walls. However, severely restricted throat geometry and high heat flux make optical diagnostic access virtually impossible. Diagnostics on tunnel walls are also very challenging, although there are examples of previous successful attempts.
Reflected-shock tunnels. Reservoir entropy, pressure fluctuations, and driver-gas contamination originate from the complex interaction between the reflected shock wave and the shock tube wall boundary layer. Furthermore, upstream diaphragm particulate-laden flow acts as a continuous source of premature model transition downstream.
Expansion tubes and tunnels. These are subject to driver-gas acoustic focusing, where upstream driver unsteadiness is focused directly into the test gas when sound speed ratios fall into unfavorable ranges. This is exacerbated by secondary diaphragm rupture wave systems propagating from upstream.
Read this list as a menu with consequences. Each focus area implies a different target facility, a different sensor environment, and a different partner. The nozzle throat is the universal option and the most brutal environment. The reflected-shock reservoir problem is the most well-characterized in the literature and the one with the most existing collaborators. The expansion tube acoustic focusing problem is the most specialized and has the fewest candidate facilities in the country.
Note also the particulate point in the reflected-shock bullet. Diaphragm particulate-laden flow causing premature model transition is a contamination problem rather than an acoustic one, and it is cited to a specific paper on shock-tube cleanliness. If your diagnostic can detect or characterize particulate, that is a differentiated capability worth calling out, since the topic frames it as a continuous source of downstream transition.
Why optical diagnostics do not solve this
Characterizing these regions is exceptionally difficult because standard optical diagnostic techniques such as Femtosecond Laser Electronic Excitation Tagging, Focused Laser Differential Interferometry, and Rayleigh scattering require optical-grade line-of-sight window access, which may not withstand the extreme pressures, temperatures, or geometric constraints of regions such as the driver, reservoir, or nozzle throat.
This is the market gap and you should state it in these terms. The hypersonics diagnostics community has excellent optical tools and no way to use them where the noise is born.
The two technical pillars
To bridge this gap, this STTR topic focuses on the development of specialized diagnostics, informed by multi-fidelity computational flow models ranging in scope from system-scale to component-specific. Collaborative proposals from Small Business Concerns must address the following technical pillars.
Upstream noise characterization and modeling. Modeling facility-specific noise generation phenomena such as transient or statistically steady boundary-layer effects, driver-gas acoustic focusing, and diaphragm bursting, starting from the most upstream location ultimately responsible for noise contamination of the test section.
Diagnostics for non-optical zones. Developing and demonstrating novel non-intrusive or minimally intrusive diagnostic systems for flow and acoustic characterization, capable of operating without traditional optical windows. Promising approaches include but are not limited to flush-mounted high-frequency pressure and thermal sensor arrays, micro-bore fiber-optic probes, laser-based acoustic sensing, and hybrid data-assimilation techniques that computationally reconstruct upstream flow states from sparse wall measurements.
The last item on that list is worth attention. Hybrid data-assimilation techniques that reconstruct upstream flow states from sparse wall measurements is a fundamentally different approach from putting a better sensor in a harder place: it accepts that you can only measure at the wall and uses the model to infer the interior. Both routes are invited. A proposal that combines them, using sparse rugged wall sensors plus assimilation into a component-scale model, sits exactly where the two pillars meet, and the topic's framing of diagnostics "informed by multi-fidelity computational flow models" suggests that is the intended shape.
Phase II Requirements
The Phase II effort will be divided into three key tasks.
Task 1: Upstream Noise Modeling and Source Characterization. Refine computational models of the targeted wind tunnel's upstream environment, focusing on detailing how disturbances generate in the driver and reservoir and focus and propagate through the throat.
Task 2: Upstream Diagnostic Suite Development. Build and calibrate a rugged, high-frequency flow and acoustic diagnostic system tailored for non-optical or highly confined spaces, for example micro-fiberoptic probes or flush-mounted acoustic arrays. Compare the novel diagnostic system developed for this task against low-order system-scale or component-specific high-fidelity predictive models.
Task 3: Facility Integration and Baseline Upstream Characterization. Deploy the diagnostic suite developed in Task 2 on the targeted wind tunnel to map baseline fluctuations directly within the upstream driver, reservoir, or throat. The measurements will correlate and support the refinement of Task 1 efforts.
Note that "the targeted wind tunnel" appears in Tasks 1 and 3, definite article. This is a facility-specific program. You are not building a general-purpose product in Phase II, you are instrumenting one tunnel, and the selection of that tunnel is a Month 4 milestone criterion. Which facility you name, and whether its operators have agreed, is likely the single most consequential decision in your proposal.
The Milestone Schedule
Phase II Base Period, 12 months
The Phase II base effort will focus on designing and validating the laboratory-scale prototype of the upstream diagnostic system and completing the engineering and integration plans for the upstream noise-mitigation hardware.
Month 4: Diagnostic System Architecture and Preliminary Design. Deliverable: Preliminary Design Review document and Upstream Diagnostic Specification Report. Criteria: selection of target wind tunnel facility completed; integration layout for upstream diagnostics, for example sensor ports and fiber-optic bypasses, finalized; preliminary numerical models of upstream mitigation hardware and observed environment completed. Associated tasks: 1 and 2.
Month 8: Diagnostic Calibration and Component Assembly. Deliverable: Confined-Space Diagnostic Calibration and Benchtop Testing Report. Criteria: successful benchtop demonstration of the diagnostic tool under simulated high-pressure and high-temperature conditions, showing a frequency response of at least 100 kHz and the ability to capture fluctuations through restricted-access ports. Associated task: 2.
Month 12: Baseline Diagnostic System Critical Design Review. Deliverable: CDR document, Upstream Structural and Thermal Safety Analysis, and Final Base Report. Criteria: final engineering drawings for upstream noise-mitigation hardware approved; structural and thermal safety and tunnel compatibility certified by target facility operators for high-pressure zones; integration interface finalized. Associated tasks: 1 and 2.
Phase II Option Period, 12 months
The Phase II option will focus on facility integration, mapping the baseline noise directly from the upstream components, and demonstrating a quantifiable reduction in freestream noise downstream.
Month 18: Facility Installation and Baseline Mapping. Deliverable: Baseline Upstream Characterization and Integration Report. Criteria: successful integration of the diagnostic suite into the target facility's upstream components, meaning reservoir, throat, or driver; baseline fluctuation measurements completed under nominal run conditions at minimum Mach 5, with and without existing noise mitigation strategies if present in the selected tunnel. Associated task: 3.
Month 24: Upstream-to-Downstream Noise Source Correlation. Deliverable: Upstream Source Localization and Propagation Analysis Report. Criteria: detailed spatial and temporal mapping of the flow path completed, successfully correlating fluctuations measured upstream, for example driver-gas acoustic focusing or throat shear-layer noise, with the resulting freestream acoustic noise in the downstream test section. Notional mitigation plan developed for observed flow characteristics.
Reading the schedule, including the thing that will decide your fate
The Month 12 criterion is the one to organize your entire base year around: structural and thermal safety and tunnel compatibility certified by target facility operators for high-pressure zones.
That is a third-party approval you do not control, for hardware you propose to install in a high-pressure section of someone else's national test asset. Facility operators at a major range or a university hypersonics laboratory are appropriately conservative about penetrations and instrumentation in the driver or reservoir, and the certification depends on their structural analysis review cycle, their safety board, and their run schedule.
The practical implication is that your target facility relationship must exist before you submit, not be developed during the base year. An existing collaboration agreement, a letter of intent from the facility, or a research institution partner who operates the tunnel is worth more than any technical claim in your proposal. If your research institution partner owns and operates the target facility, you have solved the hardest problem in the program on day one, and you should say so in the first page of your technical volume.
Note also a scope item that appears in the milestones but not in the three tasks: "upstream noise-mitigation hardware." Month 4 requires preliminary numerical models of it, and Month 12 requires final engineering drawings for it approved. The three Phase II tasks describe modeling, diagnostics, and facility integration, with no mitigation hardware task, and the option period objective mentions "demonstrating a quantifiable reduction in freestream noise downstream" while the Month 18 and Month 24 criteria ask only for measurement, correlation, and a notional mitigation plan.
So the document points in two directions on mitigation: engineering drawings for mitigation hardware are a base deliverable, but no task funds its design and no milestone requires demonstrating that it reduces noise. This is worth a question to SBIR_BAA@darpa.mil before October 14. In the meantime, the defensible approach is to scope mitigation as design and analysis only, consistent with the Month 12 drawings deliverable and the Month 24 notional plan, and to state that reading explicitly so a reviewer knows you saw the tension rather than missed it.
One more detail: the base period milestones fall at Months 4, 8, and 12, and the option milestones at Months 18 and 24. There is no Month 15 milestone, so the option period effectively runs Months 13 through 24 with reporting at its midpoint and end.
Phase III Dual Use
Military applications
High-fidelity aerodynamic and aerothermodynamic characterization of hypersonic weapons systems, glide vehicles, and interceptors across Major Range and Test Facility Bases, such as the Arnold Engineering Development Complex.
Understanding the mechanisms for tunnel noise generation directly translates to higher-fidelity ground-test data, more accurate boundary-layer transition prediction, and accelerated flight-test qualification.
The named customer is useful. AEDC and the broader MRTFB enterprise are identifiable organizations with test and evaluation budgets, and "accelerated flight-test qualification" is the benefit statement a program office responds to, because flight test is the expensive alternative to trustworthy ground test.
Commercial applications
Commercial space launch vehicle design, thermal protection system testing, and academic and commercial aerospace wind tunnel facility diagnostic upgrades.
The rugged, non-optical high-frequency diagnostic systems developed under this topic have immediate commercial application in monitoring turbulent combustion chambers, gas turbine engines, and high-pressure chemical reactors where optical access is similarly restricted.
That second paragraph is the more valuable commercial story and it is easy to skim past. The transferable product is not the hypersonics methodology, it is the sensor: a rugged, high-frequency, non-optical diagnostic that works in hot, high-pressure, optically inaccessible volumes. Combustion chambers, gas turbines, and chemical reactors are large, non-defense, recurring-revenue markets with the same physical constraint. If your commercialization strategy leads with tunnel diagnostic upgrades alone, you are describing a market of a few dozen facilities worldwide. Leading with industrial high-temperature sensing is a materially bigger case.
Export Control and the University Partnership
The technology within this topic is restricted under the International Traffic in Arms Regulation, 22 CFR Parts 120-130, which controls the export and import of defense-related material and services including export of sensitive technical data, or the Export Administration Regulation, 15 CFR Parts 730-774, which controls dual use items.
Offerors must disclose any proposed use of foreign nationals, their countries of origin, the type of visa or work permit possessed, and the statement of work tasks intended for accomplishment by the foreign nationals, in accordance with the Announcement.
Offerors are advised foreign nationals proposed to perform on this topic may be restricted due to the technical data under US export control laws.
Why this needs early attention
Hypersonics is among the most export-controlled technical areas in aerospace, and university hypersonics laboratories are staffed substantially by international graduate students and postdoctoral researchers. This is an STTR, so a single partnering research institution must perform at least 30 percent of the work.
Resolve this before you submit. Identify which institution personnel will perform which tasks and their citizenship or visa status, since Appendix A requires that disclosure in the technical volume. Engage the institution's export control and research compliance office in the first week. Settle the Fundamental Research determination, since DARPA requires you to either separate Fundamental Research tasks into their own statement of work or identify them within the prime statement of work, and that determination interacts directly with the ITAR restriction and with DARPA's stated right to impose publication restrictions.
A note on the CMMC designation
The projected CMMC requirement for this topic is Level 1, which sits oddly beside the ITAR restriction. The DARPA STTR front matter states that firms engaging in Controlled Unclassified Information, Export Controlled, or ITAR work for DARPA must have CMMC Level 2 self-assessment certification, and Appendix A states that those engaging in ITAR or CUI work must have Level 2 CMMC certification.
Compare topic DV005 in this same release, which is also ITAR restricted and carries a projected CMMC Level 2 (Self) requirement. The Level 1 projection on an ITAR-restricted topic is worth clarifying with DARPA. The prudent planning assumption is Level 2 self-assessment, since that is what the general CMMC language requires for ITAR work regardless of the topic-level projection, and a Level 2 posture satisfies both readings.
Funding, Cost Structure, and DARPA Mechanics
The award
$750,000 over a 12 month base, plus a $1,250,000 option over 12 months, for $2,000,000 across 24 months if the option is exercised.
This is the only topic in either DARPA Release 6 document where the option is larger than the base, and it is a 62 percent share. The structure follows the technical logic: the base is design and benchtop work, the option is a live facility campaign that costs real tunnel run time. But it also means the Government reserves the right to award all, some, one, or none of the options based on available funding and the performer's technical performance, and in this case that decision governs most of the program value and all of the interesting data.
Two implications. Design the base year so the option decision is easy, which here means hitting the 100 kHz benchtop demonstration cleanly and, above all, delivering the facility operators' structural and thermal certification on time. And be realistic that $750,000 over 12 months for two tasks plus a facility relationship is a modest budget, so the base year plan should be lean and focused rather than exploratory.
The resources made available under each topic will depend on the quality of the proposals received and the availability of funds.
Contract type
Multiple awards are anticipated. DARPA may award FAR-based Government contracts, firm-fixed-price or cost-plus reimbursement, or Other Transactions for Prototypes agreements under the authority of 10 U.S.C. 4022, subject to approval of the Contracting Officer or Agreements Officer respectively. Note that the companion DARPA SBIR Release 6 instructions cite 10 U.S.C. 4021 for the same instrument type.
The Government Contracting Officer reserves the right to select award instrument type regardless of what was proposed and to negotiate all terms. DARPA reserves the right to remove a proposal from award consideration if the parties fail to reach agreement within a reasonable time or if the proposer fails to provide requested additional information within three business days. Complete the DARPA SBIR/STTR Pre-Award Checklist before selection.
DARPA will apply publication or other restrictions if it determines the research presents a high likelihood of disclosing performance characteristics of military systems or manufacturing technologies unique and critical to defense, and any such award will require DARPA permission before publishing. On a hypersonics topic with a university partner, treat this as likely rather than hypothetical and raise it with your institution early.
Templates are mandatory
Templates for Volume 2 Technical Volume and Volume 3 Cost Volume are provided as attachments to the announcement at dodsbirsttr.mil, and use of these templates is mandatory. The Volume 3 Direct to Phase II Cost Proposal Template is an Excel spreadsheet on the DARPA Small Business site.
Cost substantiation
All proposed costs should be accompanied by documentation substantiating how the cost was derived: paystubs or a DCMA rate agreement for direct labor, historical invoices or a current contract for consultants, and historical invoices, current quotes, or market research for materials and equipment. You need not propose the cheapest supplier but should explain the choice.
All subcontractor and consultant costs must be detailed at the same level as prime contractor costs and substantiated with Subcontractor Pricing Considerations under FAR 15.404-3(b), entered in the Explanatory Material section of the cost proposal form. Subcontractors should send unsanitized cost proposals directly to SBIR_BAA@darpa.mil.
Three cost centers on this topic deserve documented rates. The research institution subaward, since it performs at least 30 percent of the work. Tunnel run time in the option period, which is billed by the shot or by the day depending on the facility and needs a rate basis from the operator. And high-bandwidth data acquisition, since 100 kHz-class measurements across a sensor array generate a real instrumentation and storage requirement.
If subcontractors will be performing Fundamental Research, you must either provide a separate statement of work for that work or identify those tasks within the prime statement of work. Cost sharing is permitted but not required and is not an evaluation factor. Title to property acquired with Government funds vests with DARPA unless transfer is determined more cost effective, which matters if your plan includes purchasing significant sensor or acquisition hardware.
Technical and Business Assistance
The Small Business Innovation and Economic Security Act Section 7 mandates agencies to offer TABA. DARPA will provide up to $25,000 for the Direct to Phase II.
Note what this document does not say. Unlike the companion DARPA SBIR Release 6 instructions, it does not state that TABA is in addition to the cost ceiling and not subject to profit or fee. Do not assume the SBIR language transfers. TABA requests will be reviewed by the respective contracting office or specialist at time of award.
For this topic, export control counsel and industrial market development are the two highest-value uses, the latter because the commercial case runs through combustion, turbine, and reactor sensing rather than through wind tunnels.
Questions and the FAQ
DSIP Topic Q&A will not be available for these DARPA topics. Technical questions must be submitted by October 14, 2026, by email to SBIR_BAA@darpa.mil with the topic number in the subject line, including the name, email address, and telephone number of a point of contact. All questions must be in English.
Questions submitted within seven calendar days of the proposal due date may not be answered. DARPA posts a consolidated Frequently Asked Questions document under the topic number summary on its Small Business site, updated on an ongoing basis until one week prior to the proposal due date.
DSIP technical support is available Monday through Friday, 9:00 a.m. to 5:00 p.m. Eastern, at DoDSBIRSupport@reisystems.com with a copy to SBIR_BAA@darpa.mil.
DARPA will not accept any late proposals.
Proposal format details
The Technical Volume must be a single PDF including graphics. Virus check before uploading. Do not lock or encrypt. Do not embed active graphics such as videos or moving pictures. Number all pages consecutively. Font no smaller than 10-point on 8.5 by 11 inch paper with one-inch margins. The header on each page should contain your company name, the topic number, and the DSIP-assigned proposal number, and may sit in the one-inch margin.
The Proposal Cover Sheet must include a technical abstract of no more than 3000 characters. Do not include marketing material, which will not be evaluated.
Classification, marking, and registrations
All proposals must be UNCLASSIFIED or CUI. No classified information. No proprietary information on the Proposal Coversheet in Volume 1, which may be released publicly if selected for award. Proprietary or CUI content may go in the Technical Volume, marked with the appropriate CUI Control Block on the first page. The Cost Volume should be marked CUI for PROPIN. Volumes 4 through 7 marked as appropriate based on content.
Titles, abstracts, anticipated benefits, and keywords of selected proposals undergo DARPA Policy and Security Review and may be revised or redacted, with final versions potentially appearing on the DoW SBIR/STTR awards website and sbir.gov/awards.
Maintain an accurate and active SAM.gov entity registration. Given the ITAR restriction, confirm your CMMC posture in SPRS as discussed above. DARPA points to sprs.csd.disa.mil/nistsp.htm and Project Spectrum at projectspectrum.io.
On venture capital ownership
The DARPA STTR Release 6 instructions contain no provision addressing majority ownership by venture capital operating companies, hedge funds, or private equity firms. The companion DARPA SBIR Release 6 instructions do include such a provision, explicitly permitting it under three conditions.
Do not read the SBIR provision across to this document. Eligibility for STTR awards is governed by the DoW STTR Program BAA and the SBA SBIR/STTR Policy Directive. If your ownership structure raises the question, resolve it before investing in a proposal.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW STTR Program BAA. Proposals that do not comply with the requirements detailed in this BAA and the research objectives of the corresponding topic are considered non-conforming and will not be evaluated nor considered for award.
Appendix A adds a second trap: proposals that do not adequately substantiate prior Phase I-equivalent feasibility for the components addressed will be deemed non-responsive and will not be evaluated for award.
The Government will evaluate each proposal in its entirety, documenting strengths and weaknesses against each criterion, and determine overall selectability. Proposals are not evaluated against each other but on their own individual merit. A selectable proposal is one where strengths outweigh weaknesses with no accumulated weaknesses requiring extensive negotiations or a resubmitted proposal.
Awards will be made to proposers whose proposals are most advantageous to the Government, consistent with the DoW STTR Program BAA criteria and availability of funding.
Notification of selection or non-selection within 90 calendar days of BAA close, by email to the Corporate Official on the Proposal Cover Sheet. DARPA will provide a technical evaluation narrative for each proposal, and an informal feedback session may be requested at sbir@darpa.mil at DARPA's sole discretion.
Company Commercialization Report information will not be considered during evaluations.
Protests regarding the selection decision go, as prescribed in FAR 33.106(b) and FAR 52.233-3, to DARPA Contracts Management Office, 675 N. Randolph Street, Arlington, VA 22203, by email to CMO_SBIRProtests@darpa.mil and sbir@darpa.mil.
Post-award support
DARPA provides Transition and Commercialization Support Program services to Phase II awardees upon contract execution at no cost. Awardees may also be eligible for the Embedded Entrepreneurship Initiative, invitation-only at DARPA's sole discretion, typically no more than $310,000 per awardee over the duration of the award, supporting a Senior Commercialization Advisor relationship, investor working group connections, and hiring an embedded entrepreneur to execute a Go-to-Market strategy. Your commercialization strategy section is used to assess EEI suitability, and EEI selection happens independently after award selection.
The References
Twelve, and they are the most demanding reading list in either DARPA Release 6 document. They also sort neatly into the three focus areas, which tells you how to read them.
The foundation. Laufer, "Some Statistical Properties of the Fluctuations in the Boundary Layer of a Supersonic Nozzle," Physics of Fluids, 1964. This is the nozzle-wall Mach wave radiation result that the topic description builds on.
Quiet tunnels and the state of the art. Schneider, "The Development of Hypersonic Quiet Tunnels," Journal of Spacecraft and Rockets, 2008. Hornung, "Performance of and Noise in High-Enthalpy Wind Tunnels," AIAA Paper 93-0185, 1993.
Transition in impulse facilities. Parziale, Shepherd, and Hornung, "Observations of Hypervelocity Boundary Layer Transition on a Cone in a Shock Tunnel," Journal of Fluid Mechanics, 2014.
Reflected-shock tunnel physics. Davies, "The Interaction of the Reflected Shock with the Boundary Layer in a Shock Tube and its Influence on the Duration of Hot Flow in the Reservoir," Aeronautical Research Council CP-881, 1966. Hannemann, Schnieder, Reimann, and Martine Schramm, "The influence and the delay of driver gas contamination in HEG," AIAA 2000-2593.
Particulate contamination. Jewell, Parziale, Leyva, and Shepherd, "Effects of Shock-Tube Cleanliness on Hypersonic Boundary Layer Transition at High Enthalpy," AIAA Journal, 2017.
Expansion tube physics. Trimpi, "A Preliminary Theoretical Study of the Expansion Tube," NASA Technical Report, 1962. Paull and Stalker, "Acoustic waves in shock tunnels and expansion tubes," 18th International Symposium on Shock Waves, 1991. Dufrene, Sharma, and Austin, "Design and Characterization of a Hypervelocity Expansion Tube Facility," Journal of Propulsion and Power, 2007. Furukawa et al., "Visualizing the Secondary Diaphragm Rupture in an Expansion Tube," Shock Waves, 2007.
Wall diagnostics precedent. Kasper et al., "Pressure fluctuations beneath instability wavepackets and turbulent spots in a hypersonic boundary layer," Journal of Fluid Mechanics.
Two observations. First, the Kasper reference is the one the topic points to when it says diagnostics on tunnel walls are challenging "although there are examples of previous successful attempts." That is your precedent for flush-mounted wall sensing, and citing it while explaining what you add is a strong opening for the diagnostics pillar. Second, Paull and Stalker 1991 is the acoustic focusing mechanism, and the sound speed ratio condition it identifies is the physics behind the expansion tube focus area. If you target that focus area, that paper is load-bearing.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
Technical question deadline: October 14, 2026, to SBIR_BAA@darpa.mil with the topic number in the subject line
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Selection notification: within 90 calendar days of BAA close
Base period: 12 months from award
Option: 12 additional months if exercised
A working backward plan
Before September 23. Choose your focus area and your target facility, in that order, and secure the facility relationship in writing. This is the highest-value pre-submission action available on this topic, because the Month 12 criterion requires certification by the facility operators and the option period requires physical integration into their tunnel. Commit your research institution partner, ideally one that operates the target facility. Work the ITAR and export control question with their compliance office immediately, identifying personnel, citizenship or visa status, and assigned tasks. Settle the Fundamental Research determination and raise DARPA's publication restriction language. Verify your benchtop diagnostic's frequency response against the 100 kHz Month 8 bar and its ruggedness under simulated high-pressure and high-temperature conditions. Assemble your upstream modeling evidence for the specific phenomenon you will target. Resolve IP ownership or licensing, since documentation goes in the Technical Volume. Confirm your CMMC posture, planning to Level 2 self-assessment given the ITAR restriction. Download the mandatory Volume 2 and Volume 3 templates. Read the FAQ and keep rechecking it. Read Laufer 1964, Kasper, and the references for your chosen focus area. Decide your contract type. Confirm SAM registration.
September 23 through October 5. Draft the 10 page feasibility volume first, covering both required elements: the upstream modeling evidence and the confined-space diagnostic demonstration. Include the reference list on the last page, counting toward the limit, and the one-page commercialization potential summary. Then draft the 20 page technical proposal against Appendix A's sections, with the statement of work as a substantial portion, organized around Tasks 1 through 3 and the five milestones. Name your target facility and its operator agreement early and prominently. Address the mitigation hardware scope question explicitly. Include the foreign citizens disclosure. Send your questions to SBIR_BAA@darpa.mil, including the mitigation scope question and the CMMC level question.
October 6 through October 14. Build the cost volume in the mandatory Excel template across the 12 month base and 12 month option. Get the research institution's budget at prime-level detail with Subcontractor Pricing Considerations. Get a documented rate basis for tunnel run time in the option period from the facility operator. Price high-bandwidth data acquisition, sensor fabrication, the benchtop high-pressure and high-temperature test rig, structural and thermal analysis labor for the certification package, and computational resources for the multi-fidelity modeling. Draft the 5 page transition and commercialization strategy against Appendix A's nine elements, leading with industrial high-temperature sensing rather than tunnel upgrades alone.
October 15 through October 18. Assemble Volume 5 with data rights assertions, IP documentation, CVs, subcontractor pricing considerations, and any optional letters of intent, particularly from the target facility operator, AEDC, or an MRTFB organization, to substantiate specific transition claims. Complete Volume 6 training and the Volume 7 foreign affiliations webform, which the Corporate Official must submit before certification is possible. Run compliance: single unlocked PDF, no embedded video, 10-point minimum font, consecutive page numbers, correct header, 3000 character abstract, CUI marking, mandatory Excel cost template, no marketing material.
October 19 through October 20. Submit and certify in DSIP, and confirm the mandatory supporting documents actually uploaded, since a completed submission in DSIP does not indicate that they did.
Frequently Asked Questions
What is DARPA STTR topic DPA26TZ06-DV006?
DPA26TZ06-DV006 is a DARPA STTR Direct to Phase II topic titled "Localization, Characterization, and Modeling of Freestream Disturbances in Hypersonic Wind Tunnels," released under the DoW 2026 STTR Broad Agency Announcement, Release 6. The objective is to develop and demonstrate a robust methodology to localize, characterize, and model upstream noise and flow disturbances in hypersonic wind tunnels, targeting hard-to-access regions upstream of the test section such as the driver, reservoir, and nozzle throat and walls, where traditional optical access is restricted or unavailable.
How much funding is available?
The base award is $750,000 over 12 months, with a $1,250,000 option over 12 months, for a maximum of $2,000,000 across 24 months if the option is exercised. This is the only topic in either DARPA Release 6 document where the option is larger than the base. DARPA will also provide up to $25,000 in Technical and Business Assistance for the Direct to Phase II.
When is the proposal deadline?
The topic opens September 23, 2026 and proposals are due October 21, 2026 through the Defense SBIR/STTR Innovation Portal at dodsbirsttr.mil. DARPA will not accept late proposals.
Can I submit a Phase I proposal?
No. This topic is soliciting Direct to Phase II proposals only.
What must my feasibility volume show?
Two things, in a 10-page volume. Successful modeling of upstream disturbance generation and propagation, for example driver-gas acoustic focusing, shock-tube boundary layer interactions, or nozzle throat shear layer acoustics. And a proof-of-concept or benchtop demonstration of a diagnostic method, for example fiber-optic-coupled probes, ultra-high-frequency flush-mounted sensors, acoustic emissions arrays, or shear sensors, demonstrating the frequency response and physical ruggedness required for high-pressure, high-temperature, or non-optically accessible environments.
What frequency response do I need?
The feasibility requirement does not state a number, but the Month 8 base milestone requires a benchtop demonstration showing a frequency response of at least 100 kHz along with the ability to capture fluctuations through restricted-access ports. Plan against 100 kHz.
Why can't optical diagnostics do this already?
Standard techniques such as FLEET, focused laser differential interferometry, and Rayleigh scattering all require optical-grade line-of-sight window access, which may not withstand the extreme pressures, temperatures, or geometric constraints of the driver, reservoir, or nozzle throat. That is the gap the topic exists to fill.
Which upstream regions does DARPA care about?
Three focus areas are named as examples, not limits. Nozzle throats and walls in all tunnels, where extreme thermal and velocity gradients generate disturbance and initiate wall turbulence but restricted geometry and high heat flux make optical access virtually impossible. Reflected-shock tunnels, where reservoir entropy, pressure fluctuations, and driver-gas contamination arise from reflected shock interaction with the shock tube wall boundary layer, and where diaphragm particulate-laden flow causes premature model transition. And expansion tubes and tunnels, subject to driver-gas acoustic focusing when sound speed ratios fall into unfavorable ranges, exacerbated by secondary diaphragm rupture wave systems.
What are the three Phase II tasks?
Task 1, upstream noise modeling and source characterization, refining computational models of the targeted tunnel's upstream environment. Task 2, upstream diagnostic suite development, building and calibrating a rugged high-frequency diagnostic for non-optical confined spaces and comparing it against low-order system-scale or component-specific high-fidelity predictive models. Task 3, facility integration and baseline upstream characterization, deploying the suite on the targeted tunnel to map baseline fluctuations in the driver, reservoir, or throat.
Do I have to name a specific wind tunnel?
Effectively yes. Selection of the target wind tunnel facility is a Month 4 milestone criterion, and Tasks 1 and 3 both refer to "the targeted wind tunnel." More importantly, the Month 12 criterion requires structural and thermal safety and tunnel compatibility certified by the target facility operators for high-pressure zones, which is a third-party approval you cannot obtain without an existing relationship.
What is the hardest milestone?
Month 12, because it requires certification by facility operators of structural and thermal safety and tunnel compatibility for high-pressure zones. That approval depends on someone else's safety review process and run schedule. A proposal whose research institution partner operates the target facility has a decisive advantage here.
What does the option period require?
Integration of the diagnostic suite into the target facility's upstream components with baseline fluctuation measurements at minimum Mach 5, with and without existing noise mitigation strategies if present in the selected tunnel, by Month 18. Then by Month 24, detailed spatial and temporal mapping correlating upstream fluctuations with the resulting downstream freestream acoustic noise, plus a notional mitigation plan.
Am I expected to build noise mitigation hardware?
The document is not fully consistent on this. The Month 4 criterion requires preliminary numerical models of upstream mitigation hardware, and Month 12 requires final engineering drawings for it approved. But none of the three Phase II tasks covers mitigation hardware design, and the Month 18 and Month 24 criteria ask only for measurement, correlation, and a notional mitigation plan, even though the option period objective mentions demonstrating a quantifiable reduction in freestream noise. Ask DARPA at SBIR_BAA@darpa.mil before October 14. The defensible reading is design and analysis only, consistent with the drawings deliverable and the notional plan.
Do I need a research institution partner?
Yes. STTR awards require a formal partnership with a single partnering research institution, with the small business performing at least 40 percent of the work and the research institution at least 30 percent. The DARPA instructions direct proposers to the DoW STTR Program BAA for these general requirements. This topic does not prescribe the role split, but the natural division puts the disturbance physics and multi-fidelity modeling with the institution and the sensor hardware with the small business.
Is this topic ITAR restricted?
Yes. The topic states the technology is restricted under ITAR, 22 CFR Parts 120-130, or EAR, 15 CFR Parts 730-774. Offerors must disclose any proposed use of foreign nationals, their countries of origin, visa or work permit type, and the statement of work tasks they will perform. Foreign nationals proposed to perform on this topic may be restricted.
How does ITAR interact with a university hypersonics lab?
This is the practical problem to solve first. Hypersonics is heavily export controlled and university hypersonics groups are typically staffed substantially by international students and postdocs, while the institution must perform at least 30 percent of the work. Identify personnel and status, assign tasks accordingly, engage the institution's export control office immediately, and settle the Fundamental Research determination, which interacts with both the ITAR restriction and DARPA's stated right to impose publication restrictions.
What CMMC level applies?
The projected requirement stated for this topic is Level 1, which sits oddly beside the ITAR restriction, since both the DARPA STTR front matter and Appendix A state that firms engaging in ITAR, Export Controlled, or CUI work for DARPA must have CMMC Level 2 certification. The comparable ITAR-restricted topic in this release, DV005, carries a projected Level 2 (Self) requirement. Worth clarifying with DARPA; the prudent planning assumption is Level 2 self-assessment.
How long can my technical volume be?
The standard format is 35 pages. The feasibility volume is 10 pages, the technical proposal shall not exceed 20 pages, and the Phase II commercialization strategy shall not exceed 5 pages and should be the last section of the Technical Volume. Appendix A states the commercialization strategy will not count against the proposal page limit, so confirm with DARPA how the three numbers combine if it affects your layout. Font must be at least 10-point.
Can I ask questions through DSIP Topic Q&A?
No. DSIP Topic Q&A is not available for DARPA topics. Technical questions go by email to SBIR_BAA@darpa.mil by October 14, 2026, with the topic number in the subject line and a point of contact name, email, and phone number. DARPA maintains a consolidated FAQ on its Small Business site.
What contract types can DARPA award?
FAR-based firm-fixed-price or cost-plus reimbursement contracts, or Other Transactions for Prototypes agreements under the authority of 10 U.S.C. 4022. Note that the companion DARPA SBIR Release 6 instructions cite 10 U.S.C. 4021 for the same instrument type.
Will DARPA restrict publication?
Possibly, and on a hypersonics topic it is worth planning for. DARPA states it will apply publication or other restrictions if it determines the research presents a high likelihood of disclosing performance characteristics of military systems or manufacturing technologies unique and critical to defense, and that any such award will require DARPA permission before publishing. Raise this with your university partner early.
Are venture capital backed companies eligible?
The DARPA STTR Release 6 instructions contain no provision on majority ownership by venture capital operating companies, hedge funds, or private equity firms, unlike the companion DARPA SBIR Release 6 instructions, which explicitly permit it under three conditions. Do not assume the SBIR provision applies here. Check the DoW STTR Program BAA and the SBA SBIR/STTR Policy Directive, or ask DARPA before October 14.
How much cost documentation do I need?
All proposed costs should be accompanied by documentation substantiating how the cost was derived, such as paystubs or a DCMA rate agreement for labor, contracts or historical invoices for consultants, and quotes, invoices, or market research for materials and equipment. Subcontractor and consultant costs must be detailed at prime-level and substantiated with Subcontractor Pricing Considerations under FAR 15.404-3(b). Subcontractors send unsanitized cost proposals to SBIR_BAA@darpa.mil. For this topic, get a documented rate basis for tunnel run time from the facility operator.
How will my proposal be evaluated?
Against the evaluation criteria in the DoW STTR Program BAA. Proposals are evaluated individually on their own merit rather than against each other. A selectable proposal is one where strengths outweigh weaknesses with no accumulated weaknesses requiring extensive negotiation or resubmission. Non-conforming proposals, and proposals that do not adequately substantiate prior Phase I-equivalent feasibility, are not evaluated at all.
Will I get feedback if not selected?
Yes. DARPA will provide a technical evaluation narrative for each proposal submitted, and an informal feedback session may be requested by email at sbir@darpa.mil, granted at DARPA's sole discretion.
What is the commercial market?
Commercial space launch vehicle design, thermal protection system testing, and academic and commercial wind tunnel diagnostic upgrades. More significantly, the rugged non-optical high-frequency diagnostic systems have immediate application in monitoring turbulent combustion chambers, gas turbine engines, and high-pressure chemical reactors where optical access is similarly restricted, which is a much larger market than wind tunnels.
Who do I contact with questions?
Technical questions go to SBIR_BAA@darpa.mil with the topic number in the subject line, by October 14, 2026. Administrative questions about the DARPA program and these instructions also go to SBIR_BAA@darpa.mil. DSIP technical support is DoDSBIRSupport@reisystems.com with a copy to SBIR_BAA@darpa.mil. Feedback session requests go to sbir@darpa.mil.
Positioning Advice for Companies Considering This Topic
Secure the target facility before you write a word. The Month 12 criterion requires structural and thermal safety and tunnel compatibility certified by the facility operators for high-pressure zones, and the entire option period is a campaign inside their tunnel. This is the one thing in the program you cannot buy with technical excellence. A signed collaboration agreement, a facility letter of intent, or a research institution partner that operates the tunnel is worth more than any diagnostic claim. Name the facility on page one.
Pick one focus area and commit. Nozzle throat, reflected-shock reservoir, or expansion tube driver acoustics are three different physics problems in three different facility classes with three different sensor environments. A proposal that gestures at all three reads as unfocused and, worse, cannot name a single target facility credibly. Choose the one where you have both modeling evidence and a facility relationship.
Bring the sensor and the model together, not separately. The topic frames diagnostics as "informed by multi-fidelity computational flow models," and it explicitly invites hybrid data-assimilation techniques that reconstruct upstream flow states from sparse wall measurements. That combination, rugged sparse wall sensing plus assimilation into a component-scale model, sits exactly where the two technical pillars meet and is the most defensible answer to a problem where you fundamentally cannot instrument the interior.
Demonstrate ruggedness, not just bandwidth. Two attributes are required at the gate and at Month 8: frequency response and physical ruggedness under simulated high-pressure and high-temperature conditions. Most sensor proposals will lead with bandwidth. Showing survival data, thermal drift characterization, and a mounting concept that a facility safety board would accept is the differentiator, and it happens to be exactly what the Month 12 certification will hinge on.
Budget the certification package as real engineering. Structural and thermal safety analysis for a penetration into a high-pressure driver or reservoir is a stress and thermal analysis deliverable reviewed by a third party, not a paperwork item. Staff it, price it, and schedule it against the facility's review cycle rather than against your own calendar.
Design the base year for the option decision. Sixty-two percent of the money is in the option, and DARPA can decline it. The base year has three milestones and two of them, the 100 kHz benchtop demonstration and the operator-certified CDR, are the option decision. Keep the base plan lean and focused on those two things rather than exploring the design space.
Ask the mitigation hardware question. The milestones require preliminary models and then approved final engineering drawings for upstream noise-mitigation hardware, while no task funds its design and no milestone requires proving it reduces noise. Send the question to SBIR_BAA@darpa.mil, scope it as design and analysis in your proposal, and say plainly that this is your reading. Showing that you read the document carefully is itself a signal.
Lead your commercialization strategy with industrial sensing, not wind tunnels. There are a few dozen relevant hypersonic facilities worldwide. There are thousands of combustion chambers, gas turbines, and high-pressure reactors with the same optical access problem, and DARPA named them itself. That inversion turns a niche instrumentation product into a real market, which is what the commercialization strategy is scored on.
Solve the ITAR and university problem in week one. Hypersonics plus a foreign national disclosure requirement plus a university performing at least 30 percent of the work is the most likely reason a strong team here fails to submit. Personnel, status, task assignment, compliance office, Fundamental Research determination, publication restrictions. Do it first.
Cite Kasper and Laufer, and the references for your focus area. Laufer 1964 is the foundation the topic builds on and Kasper is the wall-diagnostics precedent the topic alludes to when it says there have been previous successful attempts. Positioning your approach against those two, then against the specific references for your chosen focus area, demonstrates the awareness of the state of the art that Appendix A explicitly requires you to persuade reviewers of.
Say what "localization" means quantitatively. The topic asks you to trace disturbances from their point of origin to the test section. A proposal that states a spatial resolution or a source discrimination capability, even approximately, is making a testable claim. One that promises to "correlate upstream and downstream measurements" is not. The Month 24 criterion asks for detailed spatial and temporal mapping, so define what detail you expect to achieve.
Do not import the SBIR document's provisions. Three concrete differences matter: the OT authority citation is 4022 here rather than 4021, there is no venture capital ownership provision, and the TABA language omits the statement that TABA sits on top of the cost ceiling. If you are bidding both programs this cycle, read both documents.
DARPA STTR DPA26TZ06-DV005: Fuel-Flexible Spacecraft Electric Propulsion System
Deadline: October 21, 2026
Funding Award Size: $2m
Description: Complete guide to DARPA STTR Direct to Phase II topic DPA26TZ06-DV005, fuel-flexible spacecraft electric propulsion running on air, water, and chemical exhaust. $1M plus $1M option. Closes October 21, 2026.
Quick Answer
DPA26TZ06-DV005 is a DARPA STTR Direct to Phase II topic under the DoW 2026 STTR Broad Agency Announcement, Release 6. DARPA wants one electric propulsion product line that runs on air, water, water and carbon dioxide mixtures, or nitrogen and hydrogen mixtures, at better than 30 percent electrical efficiency, using as much shared componentry as possible. The award is $1,000,000 over 12 months with a $1,000,000 option over 12 months. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The business case is stated in the topic itself and it is unusually explicit for a DARPA description. Thruster development to date has been highly specific to individual fuels, requiring costly development of entirely new systems for new propellants. A flexible fuel electric propulsion system capable of accepting a variety of molecular propellants offers a unique capability for a single product line to serve a significantly larger mission space and total addressable market. DARPA is funding product-line consolidation, not a new thruster physics result.
The feasibility gate is short, hard, and numerical: greater than 30 percent anode thruster efficiency on air or water, greater than 100 hours of cumulative operation on a single test article at that performance level, and vacuum thrust stand data taken in a facility below 3 by 10 to the minus 5 Torr background pressure at full power with in-situ calibrated thrust stands. Three requirements, no partial credit.
This topic carries an ITAR restriction, a projected CMMC Level 2 self-assessment requirement, and an explicit division of labor between the small business and the STTR research institution. All three shape who can realistically bid.
Topic At a Glance
Topic number: DPA26TZ06-DV005
Title: Fuel-Flexible Spacecraft Electric Propulsion System
Agency: Defense Advanced Research Projects Agency (DARPA)
Solicitation: DoW 2026 Small Business Technology Transfer Broad Agency Announcement, Release 6, DARPA Proposal Submission Instructions
Program type: Direct to Phase II (DP2)
Technical volume format: Standard, 35 pages. Feasibility documentation shall not exceed 10 pages, technical proposal shall not exceed 20 pages, and the Phase II commercialization strategy shall not exceed 5 pages
Base award: $1,000,000
Base period of performance: 12 months
Option: $1,000,000 over 12 months
Component Technology Priority Area: Space Technology
Projected CMMC level requirement: Level 2 (Self)
Export control status: ITAR restricted. The technology within this topic is restricted under the International Traffic in Arms Regulation, 22 CFR Parts 120-130, or the Export Administration Regulation, 15 CFR Parts 730-774
Target fuels: air (20 to 50 percent O2, balance N2), water, water and CO2 mixtures derived from combusted hydrocarbons, and nitrogen and hydrogen mixtures derived from decomposed ammonia or hydrazine
Efficiency target: greater than 30 percent electrical efficiency across fuels
Feasibility gate: greater than 30 percent anode thruster efficiency on air or water, greater than 100 hours cumulative operation on a single test article at that level, with vacuum thrust stand data taken below 3 by 10 to the minus 5 Torr at full power using in-situ calibrated thrust stands
Phase II end state: TRL 6 hardware including thruster head, flow systems, and power processing unit
Research institution role: prescribed by DARPA. The research institution leads gas species-specific fundamental research; the small business focuses on productization, system integration, and robustness
Technical and Business Assistance: DARPA will provide up to $25,000 for the Direct to Phase II
Topic Q&A: DSIP Topic Q&A is not available for DARPA topics. Technical questions go to SBIR_BAA@darpa.mil by October 14, 2026
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Submission portal: DSIP at dodsbirsttr.mil
Keywords: spacecraft propulsion, Hall thruster, fuel-flexible, air-breathing, VLEO, in-situ resource utilization, ISRU, oxygenic propellants, multimode propulsion
The Feasibility Bar, Which Is the First Thing to Check
To qualify for this Direct to Phase II topic, applicants must provide concrete evidence of prior Phase I-equivalent capability. Specifically, the applicant must have demonstrated three things.
Performance. Greater than 30 percent anode thruster efficiency on air or water.
Durability. Greater than 100 hours of cumulative operation on a single test article on air or water, demonstrated at the performance level above.
Evidence required. Vacuum thrust stand test data validating thrust, mass flow, power, efficiency, and duration claims. Testing must have been conducted in a facility operating at a background pressure below 3 by 10 to the minus 5 Torr during full-power operation, using thrust stands calibrated in-situ.
Why each clause matters
"Anode thruster efficiency" is a specific figure of merit, not total system efficiency and not thrust efficiency. Report it the way DARPA named it, and show the derivation from measured thrust, mass flow, and discharge power so a reviewer can check the arithmetic.
"On air or water" means you need only one of the two oxygenic propellants demonstrated to qualify. Air is the harder oxidizing environment for cathodes and channel walls; water brings condensation, feed system, and start-up complications. Whichever you have, the Phase II job is extending across the full set, and your proposal should be honest about which transition is the risky one.
"On a single test article" is the sentence that eliminates a lot of otherwise credible data. One hundred cumulative hours accumulated across three rebuilt thrusters does not satisfy it. DARPA is asking about erosion and lifetime on one unit, at the efficiency level claimed, which is a materials question as much as a performance one.
The pressure requirement is a facility gate. Background pressure below 3 by 10 to the minus 5 Torr during full-power operation on an air or water thruster means real pumping capacity against a condensable or oxidizing gas load. Name the facility, state the measured background pressure at full power, and describe the pumping arrangement. Elevated background pressure inflates apparent thrust through entrainment, which is exactly the artifact this requirement exists to exclude.
"Thrust stands calibrated in-situ" means calibration under vacuum in the test configuration, not a bench calibration transferred in. Document the calibration method and the uncertainty.
What the gate screens for
Taken together: an operating electric propulsion company with a real vacuum facility, a thruster that has survived a hundred hours on an aggressive propellant, and the instrumentation discipline to prove it. This is a small population of firms worldwide, and DARPA knows it. If you clear the gate, your competition is other firms that also clear it, which makes the Phase II plan and the research institution partnership the real differentiators.
Two Appendix A rules apply and are worth restating here. Work submitted within the feasibility documentation must have been substantially performed by the proposer or the Principal Investigator. And if the technology is subject to intellectual property, the proposer must own the IP or have obtained license rights prior to proposal submission, with documentation of ownership or license rights included in the Technical Volume.
What DARPA Is Actually Looking For
The objective
Develop a fuel-flexible spacecraft electric propulsion system capable of operating efficiently on air, water, or mixtures of water and carbon dioxide or nitrogen and hydrogen, at greater than 30 percent electrical efficiency.
The problem with the status quo
Spacecraft electric propulsion systems typically rely on noble gas fuels, xenon, krypton, and argon, stored in high pressure tanks. These fuels are unattractive for sustained ambient-air operation in very Low Earth Orbit, or for liquid refueling operations in higher orbits using water or chemical rocket fuels as output propellants.
Thruster development to date has been highly specific to individual fuels, requiring costly development of entirely new systems for new propellants. For example, optimizing for air in a vLEO environment versus water in a possible future higher orbit refueling architecture. These dependencies limit total propulsive capability and overall mission flexibility and responsiveness.
The four fuel classes
This topic seeks a system able to operate on the following, using as much shared componentry as possible.
Air, specified as 20 to 50 percent O2 with the balance N2.
Water, H2O.
Mixtures of water and carbon dioxide derived from combusted hydrocarbons.
Nitrogen and hydrogen mixtures derived from decomposed ammonia or hydrazine.
Read the derivations, not just the molecules. Two of the four fuel classes are exhaust products: combustion exhaust from a hydrocarbon chemical rocket, and decomposition exhaust from a hydrazine or ammonia system. That is the multimode propulsion concept, where a chemical stage's byproducts become electric propulsion propellant, and it is why the milestone schedule asks you to identify standard physical and fluidic interfaces to connect to external hydrocarbon and hydrazine chemical systems.
The mixture ranges also matter. Air at 20 to 50 percent oxygen spans a wide oxidizing severity, and a cathode and channel design that tolerates the top of that range is a different design than one tuned for the bottom.
The commercial framing
Developing a flexible fuel electric propulsion system capable of accepting a variety of molecular propellants offers a unique capability for a single product line to serve a significantly larger mission space and total addressable market.
This STTR topic seeks to develop a complete, integrated fuel-flexible propulsion system that bridges the gap between laboratory demonstration and fieldable spacecraft technology capable of utilizing in-situ resources.
Both sentences point the same direction. This is a productization topic. "Bridges the gap between laboratory demonstration and fieldable spacecraft technology" and the TRL 6 end state tell you the technical risk DARPA is buying down is integration and qualification, not thruster invention.
Phase II Requirements
The Phase II effort will focus on developing a complete, integrated fuel-flexible propulsion system capable of operating across air, water, and nitrogen and hydrogen species.
The product will undergo envelope qualification for vibration, shock, and thermal cycling.
The system components will include a fully integrated system including the thruster, power processing unit, flow control system, and a standardized bus control interface and protocol, for example RS485-based or similar, allowing demonstrated operation via the bus control.
The prescribed division of labor
The STTR research institution will lead the gas species-specific fundamental research to improve and extend efficient operation across all critical gas species.
The STTR small business will focus on productization, system integration, and robustness.
This is the only topic in the DARPA STTR Release 6 release that prescribes the roles, and you should follow it rather than reinventing it. The consequence for your statement of work is concrete: the university tasks are the physics of operating across gas species, meaning ionization, cathode chemistry, wall interactions, and species-dependent performance mapping. The small business tasks are the thruster head, the power processing unit, the flow control system, the bus interface, and the environmental qualification.
That split also maps onto the STTR statutory work-share minimums cleanly. The small business must perform at least 40 percent of the work and the single partnering research institution at least 30 percent. Because DARPA has named the institution's scope as leading the gas species research across all critical species, a 30 percent institution share is credible on its face here in a way it often is not.
Notes on the system requirements
The bus control interface requirement is easy to underweight. "Allowing demonstrated operation via the bus control" means an end-to-end demonstration where the spacecraft bus commands the propulsion system through a standardized protocol, which is a software, electrical, and interface-documentation deliverable, not just a connector.
Envelope qualification for vibration, shock, and thermal cycling is a real cost line. Shaker table and thermal-vacuum time, plus fixture design for a thruster with a flow system attached, should be priced explicitly.
The phrase "across air, water, and nitrogen and hydrogen species" in the Phase II opening lists three, while the objective and the milestone schedule include water and CO2 mixtures as a fourth. Treat all four fuel classes as in scope, since the Month 6 and Month 12 milestones both name CO2 explicitly.
The Milestone Schedule
DARPA structures the effort as two years. The award table specifies a 12 month base and a 12 month option, so Year 1 maps to the base period and Year 2 to the option period.
Year 1: System Design, Interface Definition, and Prototyping
Month 1. Kickoff meeting presenting program schedule plans for Year 1.
Month 3. Finalize system requirements. Review university-led research on performance using air, water, and chemical exhaust mixtures. Identify standard physical and fluidic interfaces to connect to external hydrocarbon and hydrazine chemical systems.
Month 6. Review design for the integrated thruster system. University-led laboratory demonstration and technical report validating efficient performance transfer mechanisms across gas species, air, water, CO2, and N2 plus H2. Present conceptual design for gas generation and conversion mechanisms that will accept and regulate the hydrocarbon combustion and hydrazine decomposition feeds.
Month 9. Engineering model prototype design complete for thruster and supporting systems. Presentation of final test plans.
Month 12. Delivery of technical report on integrated engineering model prototype system design. Delivery of technical report on vacuum test results for prototype hardware across all gas species, including functional demonstration of gas feed systems.
Year 2: Qualification, Life Testing, and Design Iteration
Month 1. Kickoff meeting presenting program schedule plans. Note that the source document says "plans for Year 1" here, which appears to be a copy error from the Year 1 kickoff; the intent is plainly the Year 2 plan.
Month 3. Updated system design based on Year 1 test results, initial environmental testing covering vibration, shock, and thermal-vacuum, and other lessons learned.
Month 6. Written report on updated performance testing across all gas species.
Month 9. Complete 200-hour short duration wear tests on all gas species, or minimal time to evaluate wear rate following thruster burn-in. Identify preferred gas species for the 1000-hour wear test based on initial results and business case.
Final deliverables. TRL 6 hardware including thruster head, flow systems, and power processing unit. Final test report on system design and performance results including thrust, specific impulse, and overall thruster efficiency, and summarizing completed 1,000-hour and 200-hour wear tests. Technology Transition and Spaceflight Integration Plan.
Reading the schedule, including one tight spot
The base period is a design and integration year that ends with an engineering model tested across all gas species. Note that Month 6 of Year 1 requires the university's validated performance transfer mechanisms across all four gas species, which means the institution's work is front-loaded and its subaward has to be executable in the first weeks of the award.
The option year is qualification and life testing, and it contains a schedule tension worth planning for. The 1000-hour wear test species is identified at Year 2 Month 9, and the final deliverables summarize a completed 1,000-hour wear test. A thousand hours is roughly 42 days of continuous operation, which fits inside the last three months only with a facility dedicated to it and no significant interruptions. Two consequences for your proposal. First, you need continuous, uninterrupted access to a vacuum facility for that window, so name it and show the schedule. Second, the sensible approach is to start the 1000-hour test earlier than Month 9 on the species you expect to select, and to say so in your option statement of work, rather than accepting a serialized reading of the milestone that leaves no margin.
Note also the escape clause in the Month 9 wear test language: "or minimal time to evaluate wear rate following thruster burn-in." DARPA has left room for a shorter test that still characterizes the wear rate. If your erosion measurement approach can establish a wear rate in less than 200 hours per species, propose it explicitly and justify the methodology, because the alternative is 200 hours times four species inside a 12 month option.
The "business case" criterion at Year 2 Month 9 is worth noticing too. DARPA asks you to select the 1000-hour test species partly on commercial grounds, which is consistent with the topic's product-line framing. Say which species you expect to select and why, in market terms.
Phase III Dual Use
Phase III efforts will focus on flight qualification, scaling, and transition to acquisition for operational spacecraft.
The transition target is versatile use across three architectures. Very low Earth orbit satellites for air-breathing propulsion. Rideshare-compatible inert water-fueled satellites with no pressurant tanks required. Multimode propulsion systems capable of high specific impulse electric propulsion using combustion exhaust, water plus CO2, from hydrocarbon-based chemical rocket fuels, or decomposition exhaust, N2 plus H2, from hydrazine-based chemical rocket fuels.
Target military transition includes Space Development Agency or DARPA vLEO constellations.
Target commercial transition includes imagery, communication, and satellite servicing vehicles, or deep-space In-Situ Resource Utilization architectures.
The rideshare point deserves emphasis because it is the clearest near-term commercial argument. A water-fueled satellite carries no high-pressure noble gas tank, which removes a significant rideshare range-safety and integration obstacle. That is a procurement-relevant advantage today, independent of the vLEO or ISRU cases, and it is worth quantifying in your commercialization strategy.
The named military customer, Space Development Agency or DARPA vLEO constellations, is specific enough to act on. If you have any engagement with either, or with a prime building toward vLEO, that belongs in your transition strategy and possibly in an optional letter of intent.
Export Control, Which Constrains Your Team
The technology within this topic is restricted under the International Traffic in Arms Regulation, 22 CFR Parts 120-130, which controls the export and import of defense-related material and services including export of sensitive technical data, or the Export Administration Regulation, 15 CFR Parts 730-774, which controls dual use items.
Offerors must disclose any proposed use of foreign nationals, their countries of origin, the type of visa or work permit possessed, and the statement of work tasks intended for accomplishment by the foreign nationals, in accordance with the Announcement.
Offerors are advised foreign nationals proposed to perform on this topic may be restricted due to the technical data under US export control laws.
Why this is a bigger issue on this topic than on most
This is an STTR with a research institution performing at least 30 percent of the work, and university propulsion laboratories are typically staffed substantially by international graduate students and postdocs. An ITAR restriction plus a foreign national disclosure requirement plus a 30 percent university work share is a combination you need to resolve before submission, not during negotiation.
Concretely: identify which institution personnel will perform the work, their citizenship or visa status, and the specific tasks assigned, since Appendix A requires that disclosure in the technical volume. Discuss the export control posture with the institution's research compliance office early, because some universities will not accept ITAR-restricted work under their standard publication and open-research policies. If the university's participation depends on the work qualifying as Fundamental Research, that determination interacts directly with the ITAR restriction and needs to be settled up front. DARPA's own cost instructions require you to either separate Fundamental Research tasks into their own statement of work or identify them within the prime statement of work.
A note on the source document
The ITAR paragraph appears twice in this topic: once as a standalone block before the objective, stating the technology "is restricted," and again appended to the end of the objective sentence, stating it "may be restricted" and citing section 3.5 of the Announcement rather than the Announcement generally. The two versions differ in wording. The conservative reading, and the one to plan against, is the stronger of the two: treat the topic as ITAR restricted. If the distinction matters to your teaming plan, it is a reasonable question for SBIR_BAA@darpa.mil.
Note also that the projected CMMC requirement for this topic is Level 2 with self-assessment, consistent with the ITAR restriction, and higher than the Level 1 projected for the other two topics in this release. Firms engaging in Controlled Unclassified Information, Export Controlled, or ITAR work for DARPA must have CMMC Level 2 self-assessment certification, so confirm your SPRS posture before you submit.
Funding, Cost Structure, and DARPA Mechanics
The award
$1,000,000 over a 12 month base, plus a $1,000,000 option over 12 months, for $2,000,000 across 24 months if the option is exercised.
The even split is unusual and informative. Half the money is behind the option, and the option is where the life testing and qualification live. The Government reserves the right to award all, some, one, or none of the options based on available funding and the performer's technical performance, which means the qualification work is genuinely contingent on your base year results. Structure the base year to make the option decision easy.
The resources made available under each topic will depend on the quality of the proposals received and the availability of funds.
Contract type
Multiple awards are anticipated. DARPA may award FAR-based Government contracts, firm-fixed-price or cost-plus reimbursement, or Other Transactions for Prototypes agreements under the authority of 10 U.S.C. 4022, subject to approval of the Contracting Officer or Agreements Officer respectively.
Note that the companion DARPA SBIR Release 6 instructions cite 10 U.S.C. 4021 for the same instrument type. If you are preparing proposals to both programs this cycle, do not assume identical OT paperwork.
In all cases, the Government Contracting Officer reserves the right to select award instrument type regardless of what was proposed, and to negotiate all terms with selectees. DARPA reserves the right to remove a proposal from award consideration if the parties fail to reach agreement within a reasonable time or if the proposer fails to provide requested additional information within three business days. Complete the DARPA SBIR/STTR Pre-Award Checklist on the DARPA Small Business website before selection.
DARPA also notes it will apply publication or other restrictions if it determines the research presents a high likelihood of disclosing performance characteristics of military systems or manufacturing technologies unique and critical to defense, and that any resulting award will require DARPA permission before publishing. On an ITAR-restricted topic with a university partner, flag this to your institution's compliance office alongside the export control discussion.
Templates are mandatory
Templates for Volume 2 Technical Volume and Volume 3 Cost Volume are provided as attachments to the announcement at dodsbirsttr.mil. Use of these templates is mandatory. The Volume 3 Direct to Phase II Cost Proposal Template is an Excel spreadsheet on the DARPA Small Business site.
Cost substantiation
All proposed costs should be accompanied by documentation to substantiate how the cost was derived: paystubs or a DCMA rate agreement for direct labor, historical invoices or a current contract for consultants, and historical invoices, current quotes, or market research for materials and equipment. You do not have to propose the cheapest supplier, but you should explain the choice.
All subcontractor and consultant costs must be detailed at the same level as prime contractor costs and substantiated with Subcontractor Pricing Considerations under FAR 15.404-3(b), entered in the Explanatory Material section of the cost proposal form. Subcontractors should send unsanitized cost proposals directly to SBIR_BAA@darpa.mil.
For this topic the substantiation burden is concentrated in three places. The university subaward, which is a large fraction of the work. Vacuum facility time, which is the dominant direct cost for a 200-hour and 1000-hour wear test campaign and needs a rate basis. And environmental qualification, meaning shaker, shock, and thermal-vacuum time plus fixtures.
Title to property furnished by the Government or acquired with Government funds vests with DARPA unless transfer of title is determined more cost effective, which is worth reading if your plan includes buying significant test equipment. Cost sharing is permitted but not required and is not an evaluation factor.
Technical and Business Assistance
The Small Business Innovation and Economic Security Act Section 7 mandates agencies to offer TABA. DARPA will provide up to $25,000 for the Direct to Phase II.
Note precisely what this document does and does not say. It states the $25,000 figure, but unlike the companion DARPA SBIR Release 6 instructions it does not state that TABA sits in addition to the cost ceiling and is not subject to profit or fee. Do not assume the SBIR language transfers. TABA requests will be reviewed by the respective contracting office or specialist at time of award to ensure compliance with TABA requirements.
For this topic, export control and ITAR compliance counsel is the standout TABA use, followed by space industry business development given the topic's explicit product-line and total-addressable-market framing.
Questions and the FAQ
DSIP Topic Q&A will not be available for these DARPA topics. Technical questions must be submitted by October 14, 2026, by email to SBIR_BAA@darpa.mil with the topic number in the subject line, including the name, email address, and telephone number of a point of contact. All questions must be in English.
Questions submitted within seven calendar days of the proposal due date may not be answered. DARPA posts a consolidated Frequently Asked Questions document under the topic number summary on its Small Business site, updated on an ongoing basis until one week prior to the proposal due date.
DSIP technical support is available Monday through Friday, 9:00 a.m. to 5:00 p.m. Eastern, at DoDSBIRSupport@reisystems.com with a copy to SBIR_BAA@darpa.mil.
DARPA will not accept any late proposals.
Proposal format details
The Technical Volume must be a single PDF including graphics. Virus check before uploading. Do not lock or encrypt the file. Do not embed active graphics such as videos or moving pictures. Number all pages consecutively. Font no smaller than 10-point on 8.5 by 11 inch paper with one-inch margins. The header on each page should contain your company name, the topic number, and the DSIP-assigned proposal number, and may sit in the one-inch margin.
The Proposal Cover Sheet must include a technical abstract of no more than 3000 characters. Do not include marketing material, which will not be evaluated.
Classification, marking, and registrations
All proposals must be UNCLASSIFIED or CUI. No classified information. No proprietary information on the Proposal Coversheet in Volume 1, which may be released publicly if selected for award. Proprietary or CUI content may go in the Technical Volume, marked with the appropriate CUI Control Block on the first page. The Cost Volume should be marked CUI for PROPIN. Volumes 4 through 7 marked as appropriate.
Titles, abstracts, anticipated benefits, and keywords of selected proposals undergo DARPA Policy and Security Review and may be revised or redacted, with final versions potentially appearing on the DoW SBIR/STTR awards website and sbir.gov/awards.
Maintain an accurate and active SAM.gov entity registration. Confirm CMMC Level 2 self-assessment in SPRS, given both the projected topic requirement and the ITAR restriction. DARPA points to sprs.csd.disa.mil/nistsp.htm and Project Spectrum at projectspectrum.io.
On venture capital ownership
The DARPA STTR Release 6 instructions contain no provision addressing majority ownership by venture capital operating companies, hedge funds, or private equity firms. The companion DARPA SBIR Release 6 instructions do contain such a provision, explicitly permitting it under three conditions.
This is a meaningful gap for this topic in particular, because electric propulsion companies capable of clearing the feasibility gate are frequently venture funded. Do not read the SBIR provision across. Eligibility is governed by the DoW STTR Program BAA and the SBA SBIR/STTR Policy Directive, and if your ownership structure raises the question, resolve it before investing in a proposal.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW STTR Program BAA. Proposals that do not comply with the requirements detailed in this BAA and the research objectives of the corresponding topic are considered non-conforming and will not be evaluated nor considered for award.
Appendix A adds a second trap: proposals that do not adequately substantiate prior Phase I-equivalent feasibility for the components addressed will be deemed non-responsive and will not be evaluated for award.
The Government will evaluate each proposal in its entirety, documenting strengths and weaknesses against each criterion, and determine overall selectability. Proposals are not evaluated against each other but on their own individual merit. A selectable proposal is one where strengths outweigh weaknesses with no accumulated weaknesses requiring extensive negotiations or a resubmitted proposal.
Awards will be made to proposers whose proposals are most advantageous to the Government, consistent with the DoW STTR Program BAA criteria and availability of funding.
Notification of selection or non-selection within 90 calendar days of BAA close, by email to the Corporate Official on the Proposal Cover Sheet. DARPA will provide a technical evaluation narrative for each proposal, and an informal feedback session may be requested at sbir@darpa.mil at DARPA's sole discretion.
Company Commercialization Report information will not be considered during evaluations.
Protests regarding the selection decision go, as prescribed in FAR 33.106(b) and FAR 52.233-3, to DARPA Contracts Management Office, 675 N. Randolph Street, Arlington, VA 22203, by email to CMO_SBIRProtests@darpa.mil and sbir@darpa.mil.
Post-award support
DARPA provides Transition and Commercialization Support Program services to Phase II awardees upon contract execution at no cost. Awardees may also be eligible for the Embedded Entrepreneurship Initiative, invitation-only at DARPA's sole discretion, typically no more than $310,000 per awardee over the duration of the award, supporting a Senior Commercialization Advisor relationship, investor working group connections, and hiring an embedded entrepreneur to execute a Go-to-Market strategy. Your commercialization strategy section is used to assess EEI suitability, and EEI selection happens independently after award selection.
The References
Only three, which is the shortest reference list in either DARPA Release 6 document, and each one marks a distinct piece of the argument.
Rovey, Lyne, Mundahl, Rasmont, Glascock, Wainwright, and Berg, "Review of chemical-electric multimode space propulsion," Progress in Aerospace Sciences, 2020. This is the multimode propulsion foundation, and it is the framework behind the two exhaust-derived fuel classes.
Bendimerad, Savransky, and Petro, "Optimization of refueling strategies for electric propulsion space missions," Journal of Spacecraft and Rockets, 2024. This is the orbital refueling case, which is why water matters.
Andreussi et al., "A review of air-breathing electric propulsion: from mission studies to technology verification," Journal of Electric Propulsion, 2022. This is the vLEO air-breathing case.
The short list is itself a signal. DARPA is not asking you to survey a literature. The three references define three mission architectures, and the topic's premise is that one product should serve all three. A proposal that maps its fuel flexibility onto these three architectures explicitly is speaking the topic's language.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
Technical question deadline: October 14, 2026, to SBIR_BAA@darpa.mil with the topic number in the subject line
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Selection notification: within 90 calendar days of BAA close
Base period: 12 months from award
Option: 12 additional months if exercised
A working backward plan
Before September 23. Verify the three feasibility requirements against your actual records: anode efficiency above 30 percent on air or water, more than 100 cumulative hours on a single test article at that efficiency, and thrust stand data from a facility measured below 3 by 10 to the minus 5 Torr at full power with in-situ calibration. Pull the raw thrust, mass flow, and power data and confirm you can show the efficiency derivation. Confirm the 100 hours are on one article. Resolve IP ownership or licensing, since documentation must be in the Technical Volume. Commit your research institution partner and, critically, work the export control question with their research compliance office now, since an ITAR-restricted topic with foreign national graduate students is the single most likely thing to derail this teaming arrangement. Identify institution personnel, citizenship or visa status, and assigned tasks for the required disclosure. Settle the Fundamental Research determination. Confirm CMMC Level 2 self-assessment is current in SPRS. Reserve vacuum facility windows for the wear test campaign, including the 1000-hour run. Download the mandatory Volume 2 and Volume 3 templates. Read the FAQ and keep rechecking it. Decide your contract type. Confirm SAM registration.
September 23 through October 5. Draft the 10 page feasibility documentation first, since inadequate substantiation makes the proposal non-responsive. Include the reference list on its last page, counting toward the limit, and the one-page commercialization potential summary. Then draft the 20 page technical proposal against Appendix A's required sections, with the statement of work as a substantial portion, structured around the two-year milestone set. Address the fuel transition risk honestly: which species you have, which is hardest, and how the university's species-specific research de-risks it. Include the foreign citizens disclosure section. Send questions to SBIR_BAA@darpa.mil early.
October 6 through October 14. Build the cost volume in the mandatory Excel template across the 12 month base and 12 month option. Get the university's budget at prime-level detail with Subcontractor Pricing Considerations, and remember its Month 6 deliverable means its work starts immediately. Price vacuum facility time for testing across four gas species, the 200-hour wear tests, and the 1000-hour wear test. Price environmental qualification: vibration, shock, thermal-vacuum, and fixtures. Price the power processing unit and the bus interface development. Draft the 5 page transition and commercialization strategy against Appendix A's nine elements, using the rideshare and SDA vLEO angles.
October 15 through October 18. Assemble Volume 5 with data rights assertions, IP documentation, CVs, subcontractor pricing considerations, and any optional letters of intent from spacecraft integrators or vLEO programs that substantiate specific claims. Complete Volume 6 training and the Volume 7 foreign affiliations webform, which the Corporate Official must submit before certification is possible. Run compliance: single unlocked PDF, no embedded video, 10-point minimum font, consecutive page numbers, correct header, 3000 character abstract, CUI marking, mandatory Excel cost template, no marketing material.
October 19 through October 20. Submit and certify in DSIP, and confirm the mandatory supporting documents actually uploaded, since a completed submission in DSIP does not indicate that they did.
Frequently Asked Questions
What is DARPA STTR topic DPA26TZ06-DV005?
DPA26TZ06-DV005 is a DARPA STTR Direct to Phase II topic titled "Fuel-Flexible Spacecraft Electric Propulsion System," released under the DoW 2026 STTR Broad Agency Announcement, Release 6. The objective is to develop a fuel-flexible spacecraft electric propulsion system capable of operating efficiently on air, water, or mixtures of water and carbon dioxide or nitrogen and hydrogen, at greater than 30 percent electrical efficiency.
How much funding is available?
The base award is $1,000,000 over 12 months, with a $1,000,000 option over 12 months, for a maximum of $2,000,000 across 24 months if the option is exercised. DARPA will also provide up to $25,000 in Technical and Business Assistance for the Direct to Phase II.
When is the proposal deadline?
The topic opens September 23, 2026 and proposals are due October 21, 2026 through the Defense SBIR/STTR Innovation Portal at dodsbirsttr.mil. DARPA will not accept late proposals.
What must my prior work already demonstrate?
Three things. Greater than 30 percent anode thruster efficiency on air or water. Greater than 100 hours of cumulative operation on a single test article on air or water at that performance level. And vacuum thrust stand test data validating thrust, mass flow, power, efficiency, and duration claims, taken in a facility operating below 3 by 10 to the minus 5 Torr background pressure during full-power operation, using thrust stands calibrated in-situ.
Do I need to have demonstrated on both air and water?
No. The feasibility requirement says air or water. Extending across the full set of fuels is the Phase II job.
Can I combine hours across multiple test articles?
No. The requirement specifies greater than 100 hours of cumulative operation on a single test article, which makes it an erosion and lifetime demonstration on one unit rather than an aggregate operating total.
Why does the background pressure requirement matter?
Elevated background pressure in a vacuum facility inflates apparent thrust through gas entrainment. Requiring below 3 by 10 to the minus 5 Torr at full power, with in-situ calibrated thrust stands, is how DARPA excludes that artifact. Name your facility, state the measured background pressure at full power, and document the calibration method and uncertainty.
What fuels must the Phase II system handle?
Air specified as 20 to 50 percent O2 with the balance N2; water; mixtures of water and carbon dioxide derived from combusted hydrocarbons; and nitrogen and hydrogen mixtures derived from decomposed ammonia or hydrazine. The system should use as much shared componentry as possible across them.
What is the Phase II end state?
TRL 6 hardware including the thruster head, flow systems, and power processing unit, plus a final test report covering thrust, specific impulse, and overall thruster efficiency and summarizing completed 1,000-hour and 200-hour wear tests, plus a Technology Transition and Spaceflight Integration Plan. The system also undergoes envelope qualification for vibration, shock, and thermal cycling, and must demonstrate operation via a standardized bus control interface such as RS485 or similar.
How are the research institution and small business roles divided?
DARPA prescribes them for this topic. The STTR research institution will lead the gas species-specific fundamental research to improve and extend efficient operation across all critical gas species. The STTR small business will focus on productization, system integration, and robustness. This is the only topic in this STTR release that specifies the split.
What are the STTR work-share requirements?
STTR awards require a formal partnership with a single partnering research institution, with the small business performing at least 40 percent of the work and the research institution at least 30 percent. The DARPA instructions direct proposers to the DoW STTR Program BAA for these general requirements.
Is this topic ITAR restricted?
Yes. The topic states that the technology is restricted under ITAR, 22 CFR Parts 120-130, or EAR, 15 CFR Parts 730-774. Offerors must disclose any proposed use of foreign nationals, their countries of origin, visa or work permit type, and the statement of work tasks they will perform. Foreign nationals proposed to perform on this topic may be restricted.
How does ITAR interact with the university partnership?
This is the practical issue to resolve before submitting. University propulsion laboratories are often staffed substantially by international students and postdocs, and the institution must perform at least 30 percent of the work. Identify institution personnel and their citizenship or visa status, assign tasks accordingly, and work the question with the institution's research compliance office early, since some universities will not accept ITAR-restricted work under standard open-research policies. The Fundamental Research determination interacts directly with this.
What CMMC level applies?
The projected requirement for this topic is CMMC Level 2 with self-assessment, higher than the Level 1 projected for the other two topics in this STTR release. Firms engaging in Controlled Unclassified Information, Export Controlled, or ITAR work for DARPA must have CMMC Level 2 self-assessment certification.
How long can my technical volume be?
The standard format is 35 pages. Feasibility documentation shall not exceed 10 pages, the technical proposal shall not exceed 20 pages, and the Phase II commercialization strategy shall not exceed 5 pages and should be the last section of the Technical Volume. Appendix A states the commercialization strategy will not count against the proposal page limit, so confirm with DARPA how the three numbers combine if it affects your layout. Font must be at least 10-point.
Is the 1000-hour wear test feasible inside the option year?
It is tight. The species for the 1000-hour test is identified at Year 2 Month 9, and a thousand hours is roughly 42 days of continuous operation. The practical approach is to begin the test earlier on the species you expect to select, secure uninterrupted facility access, and say so explicitly in your option statement of work. Note also that DARPA allows "minimal time to evaluate wear rate following thruster burn-in" as an alternative to the full 200-hour tests, so a justified shorter-duration wear rate methodology is worth proposing.
Can I ask questions through DSIP Topic Q&A?
No. DSIP Topic Q&A is not available for DARPA topics. Technical questions go by email to SBIR_BAA@darpa.mil by October 14, 2026, with the topic number in the subject line and a point of contact name, email, and phone number. DARPA maintains a consolidated FAQ on its Small Business site.
What contract types can DARPA award?
FAR-based firm-fixed-price or cost-plus reimbursement contracts, or Other Transactions for Prototypes agreements under the authority of 10 U.S.C. 4022. Note that the companion DARPA SBIR Release 6 instructions cite 10 U.S.C. 4021 for the same instrument type.
Are the templates mandatory?
Yes. Templates for Volume 2 Technical Volume and Volume 3 Cost Volume are attachments to the announcement, and use of these templates is mandatory. The Volume 3 template is an Excel spreadsheet.
Are venture capital backed companies eligible?
The DARPA STTR Release 6 instructions contain no provision on majority ownership by venture capital operating companies, hedge funds, or private equity firms, unlike the companion DARPA SBIR Release 6 instructions, which explicitly permit it under three conditions. Do not assume the SBIR provision applies here. Check the DoW STTR Program BAA and the SBA SBIR/STTR Policy Directive, or ask DARPA before October 14.
Will DARPA restrict publication?
Possibly. DARPA states it will apply publication or other restrictions if it determines the research presents a high likelihood of disclosing performance characteristics of military systems or manufacturing technologies unique and critical to defense, and that any such award will require DARPA permission before publishing. Raise this with your university partner alongside the export control discussion.
How will my proposal be evaluated?
Against the evaluation criteria in the DoW STTR Program BAA. Proposals are evaluated individually on their own merit rather than against each other. A selectable proposal is one where strengths outweigh weaknesses with no accumulated weaknesses requiring extensive negotiation or resubmission. Non-conforming proposals, and proposals that do not adequately substantiate prior Phase I-equivalent feasibility, are not evaluated at all.
Will I get feedback if not selected?
Yes. DARPA will provide a technical evaluation narrative for each proposal submitted, and an informal feedback session may be requested by email at sbir@darpa.mil, granted at DARPA's sole discretion.
What is the commercial market?
Rideshare-compatible water-fueled satellites with no pressurant tanks required, imagery and communication satellites, satellite servicing vehicles, and deep-space in-situ resource utilization architectures. The named military transition targets are Space Development Agency or DARPA vLEO constellations.
Who do I contact with questions?
Technical questions go to SBIR_BAA@darpa.mil with the topic number in the subject line, by October 14, 2026. Administrative questions about the DARPA program and these instructions also go to SBIR_BAA@darpa.mil. DSIP technical support is DoDSBIRSupport@reisystems.com with a copy to SBIR_BAA@darpa.mil. Feedback session requests go to sbir@darpa.mil.
Positioning Advice for Companies Considering This Topic
Prove the three feasibility numbers before you write anything else. Anode efficiency above 30 percent, more than 100 hours on one article, and thrust stand data from a facility below 3 by 10 to the minus 5 Torr with in-situ calibration. Pull the raw data and rebuild the efficiency calculation. If the 100 hours were accumulated across rebuilt units, or the background pressure was higher than specified, you do not clear the gate and the proposal is non-responsive rather than merely weak.
Solve the ITAR and university problem first. This is the most likely reason a strong team fails to submit. An ITAR-restricted topic requires a foreign national disclosure, the institution must perform at least 30 percent of the work, and propulsion labs are internationally staffed. Talk to the institution's research compliance office in the first week, name the personnel and their status, assign tasks accordingly, and settle the Fundamental Research determination. Do not discover this in November.
Follow DARPA's prescribed role split rather than inventing your own. The institution leads gas species-specific fundamental research; you lead productization, integration, and robustness. That mapping makes a 30 percent institution share credible and gives you a clean statement of work structure. Deviating from it invites a reviewer to ask why.
Be honest about which fuel transition is hard. You qualified on air or water. Getting to all four classes is the technical risk, and the two exhaust-derived classes, water plus CO2 from hydrocarbon combustion and N2 plus H2 from hydrazine decomposition, bring feed system and gas conditioning problems that are different in kind from the thruster problem. The Month 6 milestone asks for a conceptual design for the gas generation and conversion mechanisms that accept and regulate those feeds. Do not leave that as an integration detail.
Treat "as much shared componentry as possible" as a scored design principle. The entire commercial premise is one product line serving multiple missions. Show a component-level table of what is common across all four fuels and what must change, because that table is the argument for the topic's own thesis.
Design the base year to make the option decision easy. Half the money is behind the option, and DARPA reserves the right to exercise all, some, or none based on technical performance. The base year ends with an engineering model tested across all gas species. Make the Month 12 deliverables unambiguous evidence that qualification is worth funding.
Get ahead of the 1000-hour test schedule. The species is nominally selected at Year 2 Month 9, and a thousand hours is about six weeks of continuous operation. Propose starting earlier on your expected species, name the facility, show the dedicated window, and use DARPA's own "minimal time to evaluate wear rate" language to justify a defensible shorter methodology for the other species if you have one.
Do not skip the bus control interface. "Demonstrated operation via the bus control" through a standardized protocol is a real deliverable with software, electrical, and documentation content. It is also the requirement most relevant to a spacecraft integrator reading your transition plan, so it is worth more proposal space than its one sentence in the topic suggests.
Lead your commercialization strategy with rideshare. A water-fueled satellite with no high-pressure noble gas tank removes a genuine range-safety and integration obstacle for rideshare payloads today. That is a nearer-term, more concrete market argument than vLEO air-breathing or deep-space ISRU, and it is the one a commercial buyer will recognize immediately. Then layer the SDA and DARPA vLEO constellation targets and the multimode architectures on top.
Map your proposal onto the three references. Rovey 2020 for multimode, Bendimerad 2024 for refueling, Andreussi 2022 for air-breathing. Three references define three mission architectures, and the topic's thesis is that one product serves all three. Saying that explicitly, architecture by architecture, is speaking the topic's language.
Price the facility time properly. Testing four gas species, running 200-hour wear tests, running a 1000-hour wear test, and doing vibration, shock, and thermal-vacuum qualification is a facility-hours problem more than a labor problem. All costs must be substantiated, so get rate documentation for the facility and the environmental test house early.
Note the document differences if you are also bidding DARPA SBIR this cycle. The OT authority citation differs, there is no venture capital ownership provision here, and the TABA language does not include the SBIR document's statement that TABA sits on top of the cost ceiling. Read both documents rather than assuming they match.
DARPA STTR DPA26TZ06-DV004: SHIELDER, Scalable Hard-Mask Materials with Improved Etch Resistance for Extreme-Aspect-Ratio Fabrication
Deadline: October 21, 2026
Funding Award Size: $2m
Description: Complete guide to DARPA STTR Direct to Phase II topic DPA26TZ06-DV004, SHIELDER scalable hard-mask materials for extreme-aspect-ratio nanofabrication. $1.5M plus $450K option. Closes October 21, 2026.
Quick Answer
DPA26TZ06-DV004 is a DARPA STTR Direct to Phase II topic under the DoW 2026 STTR Broad Agency Announcement, Release 6. DARPA wants a fundamentally new class of nanofabrication hard mask, one that survives aggressive plasma etching well enough to enable 100:1 aspect ratio structures with sidewall roughness under 2 nanometers. The award is $1,500,000 over 24 months with a $450,000 option over 12 months. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The problem statement is a manufacturing bottleneck, not a science project. Current hard masks are thick CVD inorganic films such as silicon dioxide and silicon nitride, or sputtered metals. To get enough etch resistance they must be deposited thick, which induces high stress, structural instability, pattern distortion, and reduced feature fidelity during the etch. Conventional metal masks are polycrystalline, and their grain boundaries erode unevenly under ion bombardment, propagating severe line-edge and sidewall roughness into the finished device.
DARPA is explicit about the kind of answer it wants. Proposed approaches should bypass those degradation mechanisms by using low-dimensional, continuously ordered, inherently grain-free, or self-regenerating structures. That phrase is the design brief. If your material is polycrystalline and thick, you are proposing an incremental improvement to the thing DARPA said is broken.
Because this is an STTR, you must team with a research institution, and the small business and the institution each have statutory minimum shares of the work. That partnership is not optional packaging, it is an eligibility requirement.
Topic At a Glance
Topic number: DPA26TZ06-DV004
Title: Scalable Hard-mask materials with Improved Etch resistance and Low Degradation for Extreme-aspect-Ratio fabrication (SHIELDER)
Agency: Defense Advanced Research Projects Agency (DARPA)
Solicitation: DoW 2026 Small Business Technology Transfer Broad Agency Announcement, Release 6, DARPA Proposal Submission Instructions
Program type: Direct to Phase II (DP2). This topic is soliciting Direct to Phase II proposals only
Technical volume format: Standard, 35 pages. Feasibility documentation shall not exceed 10 pages, technical proposal shall not exceed 20 pages, and the Phase II commercialization strategy shall not exceed 5 pages
Base award: $1,500,000
Base period of performance: 24 months
Option: $450,000 over 12 months
Component Technology Priority Areas: Advanced Materials, Microelectronics
Projected CMMC level requirement: Level 1
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic, which distinguishes it from the other two topics in this STTR release
Feasibility gate: measured etch selectivity exceeding 50:1, demonstrated aspect ratio greater than 25:1, and sidewall roughness under 5 nm RMS, all with metrology data
Phase II end state: etch selectivity greater than 150:1, aspect ratio 100:1, line-edge and sidewall roughness under 2 nm RMS
Research institution partner: required, as with all STTR awards
Technical and Business Assistance: DARPA will provide up to $25,000 for the Direct to Phase II
Topic Q&A: DSIP Topic Q&A is not available for DARPA topics. Technical questions go to SBIR_BAA@darpa.mil by October 14, 2026
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Submission portal: DSIP at dodsbirsttr.mil
Keywords: nanofabrication, dry etching, hard mask, etching selectivity, high-aspect-ratio, pattern transfer
The Feasibility Bar, Which Is the First Thing to Check
This topic is soliciting Direct to Phase II proposals only. Proposals will be considered for DP2 funding for teams that provide compelling evidence of the feasibility of their novel hard-mask materials at a laboratory or R&D fabrication facility. Proposers must provide data showing that Phase I feasibility has been achieved through prior work. Four categories of documentation are required.
Etch selectivity data
Experimental data demonstrating the successful deposition, synthesis, or transfer of low-stress mask films, alongside measured etch selectivity exceeding 50:1 against target substrates such as silicon or complex oxides, under aggressive dry plasma conditions.
Note the two halves. You need the film, made by a real process, and you need a measured selectivity number above 50:1 in aggressive conditions. "Low-stress" is in the requirement, which means film stress is a reported quantity, not an afterthought.
High-aspect-ratio pattern transfer
Proof-of-concept fabrication data showing deep pattern transfer into a substrate, achieving an aspect ratio greater than 25:1 without significant mask degradation, faceting, or critical dimension loss.
The three named failure modes are the vocabulary DARPA will use to read your data. Show cross sections that let a reviewer confirm the absence of mask erosion, top-corner faceting, and CD loss, and label them in those terms.
Edge fidelity and roughness verification
High-resolution metrology data, for example cross-sectional SEM, TEM, or AFM, verifying pristine pattern transfer, specifically demonstrating etched structures with nanometer-scale sidewall roughness under 5 nm RMS.
RMS is specified. Report the measurement method, the sampling length, and the instrument, because roughness numbers are meaningless without them and a reviewer in this field knows it.
Scalability and integration pathway
Initial feasibility data or a substantiated extrapolation demonstrating a viable pathway to scale the masking technology from lab-scale prototypes to wafer-level dimensions, for example via CVD, ALD, or scalable continuous transfer methods.
Proposers must document how their proposed materials and process flows align with standard commercial fabrication or DoW constraints, specifically addressing CMOS compatibility, thermal budget, vendor and foundry transferability, and any additional constraints imposed by the application of interest.
This fourth item is the one most likely to be underwritten by a strong materials team, and it is the one where "a substantiated extrapolation" is explicitly allowed in place of data. Take the invitation, but substantiate it. Four named constraints are listed, and each deserves an explicit answer: is the material CMOS-compatible, what is the thermal budget, can a foundry or vendor actually run it, and what does your target application add.
What the gate screens for
Read the four together and the intended team becomes clear: a small business with a working deposition or transfer process and real etch data, partnered with a research institution that understands the erosion physics. Selectivity above 50:1 with sub-5 nm sidewalls at better than 25:1 aspect ratio is not a paper result. It is a measured result from a cleanroom.
Note also the standard Appendix A rule that applies here: work submitted within the feasibility documentation must have been substantially performed by the proposer or the Principal Investigator. Data from a collaborator you have not yet teamed with does not satisfy the gate. And if the technology in your feasibility documentation is subject to intellectual property, you must either own that IP or have obtained license rights prior to proposal submission, with documentation of ownership or license rights included in the Technical Volume.
What DARPA Is Actually Looking For
The objective
Develop and demonstrate novel nanofabrication hard-mask materials that exhibit substantially improved plasma etch resistance, enabling the implementation of extreme high-aspect-ratio structures with excellent pattern-transfer fidelity, minimal sidewall roughness, and precise dimensional control.
Proposed solutions should overcome the fundamental limitations of conventional lithography masks in a scalable fabrication environment by creating next-generation technologies capable of delivering high-resolution nanoscale features encountered across semiconductor, photonic, MEMS, and quantum device platforms.
Why DARPA cares
Next-generation defense-relevant microsystem technologies, such as 3D integrated circuits, high-density memory arrays, MEMS inertial sensors and RF filters, and integrated photonic devices, rely on the precise manufacturing of extreme high-aspect-ratio features within a chip.
Emerging architectures for multiferroic memory and logic components, such as those being pioneered under DARPA's Fast and Curious program, stand to benefit immensely from novel process flows capable of delivering deep trenches with ultra-low line-edge roughness.
As critical dimensions of these devices continue to shrink to improve performance or reduce size and cost, the ability to accurately transfer lithographic patterns into underlying substrates such as silicon via aggressive plasma etching has become a primary manufacturing bottleneck.
The named reference to DARPA's Fast and Curious program is a signal worth following. If your mask enables deep, ultra-smooth trenches in multiferroic stacks, saying so in the program's own terms gives a reviewer an immediate transition story.
What is wrong with the state of the art
Current state-of-the-art hard masks predominantly utilize thick chemical vapor deposition inorganic films, for example silicon dioxide and silicon nitride, or sputtered metals. These conventional technologies face fundamental physical limitations at extreme scales.
To achieve the necessary etch resistance, standard masks must be deposited with significant thickness, which induces high stress, structural instability, pattern distortion, and reduced feature fidelity during the etch process.
Conventional metal masks exhibit polycrystalline structures. Their inherent grain boundaries erode unevenly under ion bombardment, propagating severe line-edge and sidewall roughness that degrades the electrical and structural integrity of the final device.
The design brief
This STTR topic seeks highly innovative masking materials that disrupt the current paradigm. The goal is to identify and develop solutions that deliver ultra-high etch selectivity while addressing the weaknesses of current state-of-the-art solutions.
Proposed approaches should inherently bypass the degradation mechanisms of traditional masks by utilizing low-dimensional, continuously ordered, inherently grain-free, or self-regenerating structures to ensure pristine pattern transferring with low line-edge roughness even when subjected to aggressive, high-density plasma.
Four structural strategies are named. Low-dimensional points at two-dimensional crystalline materials, and DARPA cites a 2026 Nature Materials paper on two-dimensional crystalline hard masks for high-aspect-ratio nanofabrication. Continuously ordered and inherently grain-free both point away from polycrystalline films. Self-regenerating is the most unusual of the four and the most open: a mask that replenishes itself during the etch is a different failure model entirely. Pick your mechanism, name it in DARPA's vocabulary, and explain the physics.
Substrates and tools
Solutions do not have to be limited to silicon processing and may target compound semiconductors, wide-bandgap materials, piezoelectric substrates, ceramics, heterogeneous material stacks, or any other microsystems platform relevant to the Department of War.
Compatibility with existing or minimally-modified reactive ion etching, inductively coupled plasma, Bosch Deep RIE, or related plasma fabrication tools is strongly preferred.
That preference matters commercially. A mask that requires a new etch tool has a much harder transition path than one that drops into an existing DRIE or ICP chamber, and DARPA has told you which it prefers.
Metrics of interest
Proposers should identify the underlying physical mechanisms responsible for enhanced pattern-transfer performance and demonstrate a clear path toward scalable manufacturing.
Metrics of interest include improvements in etch selectivity, maximum achievable aspect ratio, critical-dimension control, sidewall roughness, mask thickness reduction, process throughput, and compatibility with wafer-scale manufacturing.
Seven metrics. Selectivity and aspect ratio get the headline numbers, but mask thickness reduction and process throughput are the two that a fab actually cares about and that most proposals will neglect. A thinner mask that etches deeper is the whole point of the topic.
Phase II Requirements
Direct to Phase II: develop, integrate, and demonstrate the following five capabilities.
Scalable deposition, synthesis, or transfer methodologies for hard-mask materials. Candidate masks must demonstrate an etch selectivity exceeding that of state-of-the-art inorganic films or sputtered metals, for example greater than 150:1 mask-to-substrate under aggressive plasma conditions.
High-fidelity pattern transfer capabilities demonstrating extreme aspect ratios of 100:1 without mask failure, critical dimension loss, or top-edge faceting.
Sub-nanometer dimensional control, with candidate masks demonstrating etched structures with sidewall and line-edge roughness under 2 nm root-mean-square.
Compatibility with standard lithographic workflows. The hard mask itself must be readily patternable, for example via secondary chemical plasma or resist, while maintaining extreme resistance to the primary deep-etch plasma.
Mitigation of unwanted metallic or particulate contamination to levels acceptable for standard semiconductor, MEMS, or photonic device fabrication lines.
The patternability paradox, which deserves a section of your proposal
The fourth requirement is the hardest intellectual problem in the topic and the easiest to gloss over. The mask must be easy to pattern and nearly impossible to etch. Those are the same physical property pointed in opposite directions.
Every credible answer resolves this with a selectivity mechanism that is chemistry-specific rather than energy-specific: a secondary plasma chemistry or resist process that attacks the mask readily, and a primary deep-etch chemistry to which it is nearly inert. Say which two chemistries you use, what the selectivity ratio is in each direction, and how you keep the patterning step from damaging the mask's key structural property. A proposal that states a 150:1 selectivity number and never explains how the mask itself gets patterned has left out half the process.
The demonstration environment
Hard mask demonstration must be conducted in a standard cleanroom environment where the process can be scaled to production and which supports the processing of standard-size wafers using commercial Deep Reactive Ion Etching or other Inductively Coupled Plasma tools.
Patterning performance must be compared between the novel hard mask and a standard baseline mask material such as silicon dioxide, silicon nitride, or a sputtered metal, for example TiN or Cr.
The goal is to demonstrate that the novel mask successfully maintains structural integrity in aggressive etch conditions where standard masks fail, for example the novel mask maintains CD fidelity and sub-nanometer sidewall smoothness at etch depths that would completely erode or severely facet the standard mask.
Compared to standard masks, the proposed solution must deliver a statistically significant reduction in defect propagation and edge roughness.
Two words to take literally. "Standard cleanroom environment where the process can be scaled to production" means a university research cleanroom running quarter-wafer coupons may not satisfy the requirement on its own, so name your facility and its wafer capability. And "statistically significant" means a designed comparison with enough samples to support a statistical claim, not a pair of representative micrographs. Budget the wafer count.
Commercialization and transition obligations inside the technical work
This DP2 will also require a commercialization and transition plan along with technology development.
Throughout the phase, the proposers must collaborate with commercial and military end-users to refine operational requirements and deployment scenarios of their developed solution.
Manufacturing and scaling plans for production must also be developed before the end of the program.
The final report must also include technology transfer documents outlining planned opportunities for commercial and military applications.
Note "throughout the phase." End-user collaboration is a continuous requirement rather than a final deliverable, which in practice means named foundry, fab, or program-office contacts in your proposal. This is separate from and additional to the 5 page commercialization strategy required by Appendix A.
Deliverables
The deliverables for DP2 will include the final prototype mask formulation and process recipe, a comprehensive testing and validation report outlining the final demonstration, the technology transfer plan with commercialization plans, and the manufacturing and scaling strategy.
The Base Milestone Schedule
DP2 base milestones for this program should include the following.
Month 3. Establish and report baseline parameters for the candidate mask material and develop a comprehensive chemical vapor deposition, or equivalent scalable deposition, process map.
Month 6. Demonstrate initial solution-based, or alternative scalable, coating and dry etching. Achieve stable precursor formulation, continuous film formation on small-scale substrates, and an initial mask-to-substrate etch selectivity of at least 50:1.
Month 9. Screen multiple mask candidates and execute the first patterned etch using the down-selected scalable coating method. Demonstrate initial high-aspect-ratio patterning, for example greater than 75:1, with a line-edge roughness of 5 nm or less using a process flow that is scalable to production levels.
Month 12. Demonstrate process scalability with a target mask-to-substrate etch selectivity of at least 100:1.
Month 15. Expand process integration for the mask deposition. Achieve strict film thickness uniformity, for example less than 10 percent variation, across intermediate-scale substrates and demonstrate patterning resolution with minimum feature sizes of 100 nanometers or less.
Month 18. Validate mask performance on alternative, non-standard semiconductor or dielectric substrates. Demonstrate a minimum selectivity of 50:1 on at least two alternative materials and scale the coating process to larger wafer sizes.
Month 21. Execute integrated, wafer-level processing demonstrations. Show successful transfer or direct integration of the vapor-deposited films and validate extreme etching capabilities on application-specific substrates, for example semiconductor devices, MEMS components, or photonic structures.
Month 24. Final Phase II demonstration. Achieve extreme etch selectivity greater than 150:1, ultra-high aspect ratios greater than 100:1 on primary semiconductor substrates, and strict dimensional control with a line-edge roughness of 2 nanometers or less.
Reading the schedule
The performance ramp is legible and aggressive: selectivity goes 50:1 at Month 6, 100:1 at Month 12, 150:1 at Month 24, while aspect ratio goes 75:1 at Month 9 to 100:1 at Month 24 and roughness goes 5 nm at Month 9 to 2 nm at Month 24.
Notice that Month 6 asks for a selectivity you already had to demonstrate to win the award. That is intentional. The early base period is about reproducing your result inside a scalable coating process, not about beating it. Months 15 through 21 are the scale-up spine: uniformity across intermediate substrates, then two alternative material systems, then wafer-level integration.
Note also two schedule facts worth planning around. The milestones name specific deposition language, "chemical vapor deposition or equivalent scalable deposition" at Month 3 and "solution-based or alternative scalable coating" at Month 6, which suggests DARPA anticipated a solution-processed route while leaving the door open. If your process is neither, say clearly at Month 3 what your equivalent is.
One gap in the source document worth a question
The topic specifies base milestones through Month 24 and labels them "DP2 Base milestones." The award structure table lists a $450,000 option over 12 months. The topic does not describe what the option period covers or provide option milestones.
Appendix A requires a Phase II Option Statement of Work "if applicable, specified in the corresponding topic," which leaves it ambiguous whether an option SOW is required here and, if so, against what scope. This is a good candidate for a question to SBIR_BAA@darpa.mil before October 14. In the meantime, the defensible approach is to propose an option scope that follows naturally from the Month 24 end state, most plausibly foundry qualification, extended substrate coverage, or pilot production of the mask material, and to say plainly that you are proposing it in the absence of stated option milestones.
Phase III Dual Use
Phase III efforts should focus on transitioning the developed hard-mask technology to DoW and commercial semiconductor fabrication facilities.
DoW applications include the domestic manufacturing of secure high-density memory, advanced electro-optical components, and high-performance MEMS and RF components for electronic warfare and GPS-denied navigation.
Commercially, this technology addresses fundamental physical scaling bottlenecks in the global semiconductor industry. Direct dual-use applications include the production of next-generation 3D semiconductor transistors, multiferroic memory and logic components, and high-coherence solid-state quantum devices.
The commercial case here is unusually strong for a defense topic, because high-aspect-ratio etch is a rate limiter in 3D NAND, DRAM capacitor formation, through-silicon vias, and MEMS timing and inertial devices, all of which are large existing markets with identifiable buyers. The GPS-denied navigation mention is the specific defense pull: high-performance MEMS inertial sensors depend on deep, smooth, high-aspect-ratio trenches, and that is a stated DoW need with named programs behind it.
The STTR Partnership Requirement
This is an STTR, not an SBIR, and the difference is structural rather than administrative. STTR awards require a formal partnership between the small business concern and a single partnering research institution, with statutory minimum shares of the work performed by each. The small business must perform at least 40 percent of the work and the single partnering research institution must perform at least 30 percent, with the balance allocated between them or to other subcontractors.
Two notes specific to this document. First, the DARPA STTR Release 6 instructions do not restate the work-split percentages. They direct proposers to follow all general instructions provided in the DoW STTR Program BAA, which is where the split, the research institution eligibility rules, the required allocation-of-rights agreement, and the intellectual property provisions live. Read that document, not just this one.
Second, unlike the two other topics in this release, DPA26TZ06-DV004 does not assign specific roles to the small business and the research institution. Topic DV005 in this same release does, stating that the research institution will lead fundamental research while the small business focuses on productization and integration. The absence of that language here means you define the division of labor yourself, and a reviewer will read it as a proxy for whether the partnership is real.
For a topic like this one, the natural split writes itself. The research institution owns the erosion physics, the mechanism identification DARPA explicitly asked for, and the advanced metrology, which is where TEM and AFM capability usually lives. The small business owns the scalable deposition or transfer process, the cleanroom demonstration on standard wafers, the contamination control, and the foundry transfer path. Say which institution, which faculty PI, which instruments, and which tasks.
Funding, Cost Structure, and DARPA Mechanics
The award
$1,500,000 over a 24 month base, plus a $450,000 option over 12 months, for $1,950,000 across 36 months if the option is exercised.
This is the largest base award and the longest total period of the three topics in this STTR release, and the only one where the option is small relative to the base. The resources made available under each topic will depend on the quality of the proposals received and the availability of funds. The Government reserves the right to award all, some, one, or none of the options based on available funding and the performer's technical performance.
Contract type
Multiple awards are anticipated. DARPA may award FAR-based Government contracts, firm-fixed-price or cost-plus reimbursement, or Other Transactions for Prototypes agreements under the authority of 10 U.S.C. 4022, subject to approval of the Contracting Officer or Agreements Officer respectively.
Note the authority citation. The DARPA SBIR Release 6 instructions cite 10 U.S.C. 4021 for the same instrument, while this STTR document cites 10 U.S.C. 4022. If you are preparing both an SBIR and an STTR proposal in this cycle, do not assume the OT paperwork is identical.
In all cases, the Government Contracting Officer reserves the right to select award instrument type, regardless of the instrument type proposed, and to negotiate all instrument terms and conditions with selectees.
DARPA points proposers to the DARPA SBIR/STTR Pre-Award Checklist on its Small Business website and asks that it be completed prior to being selected for award. It also reserves the right to remove a proposal from award consideration if the parties fail to reach agreement on terms within a reasonable time, or if the proposer fails to provide requested additional information within three business days.
Templates are mandatory
Templates for Volume 2 Technical Volume and Volume 3 Cost Volume are provided as attachments to the announcement posted at dodsbirsttr.mil. Use of these templates is mandatory. The Direct to Phase II Volume 3 Cost Proposal Template is an Excel spreadsheet available on the DARPA Small Business site.
Cost substantiation, which is stricter than most proposers expect
All proposed costs should be accompanied by documentation to substantiate how the cost was derived. DARPA gives examples: paystubs or a DCMA rate agreement for direct labor, historical invoices or a current contract for consultants, and historical invoices, current quotes, or market research for materials and equipment.
Proposers do not necessarily have to propose the cheapest item or supplier, but should explain the decision to choose one over another. Failure to include documentation with your proposal will delay contract negotiation.
All subcontractor and consultant costs must be detailed at the same level as prime contractor costs for labor, travel, and equipment, and must be substantiated with Subcontractor Pricing Considerations under FAR 15.404-3(b), entered in the Explanatory Material section of the online cost proposal form. Subcontractors should send unsanitized cost proposals directly to SBIR_BAA@darpa.mil.
This matters a great deal on an STTR, where the research institution is a subcontractor performing at least 30 percent of the work. Its budget has to be built out at prime-level detail and price-analyzed by you.
If subcontractors will be performing Fundamental Research, you must either provide a separate statement of work outlining the work that qualifies as Fundamental Research, or identify within the prime statement of work which tasks are fundamental research. On a university-partnered STTR this is a live question, and getting it right early affects publication rights and contract terms.
Cost sharing is permitted but is not required and will not be an evaluation factor.
Technical and Business Assistance
The Small Business Innovation and Economic Security Act Section 7 mandates agencies to offer TABA. DARPA will provide up to $25,000 for the Direct to Phase II.
Worth noting precisely: this STTR document states the $25,000 figure but, unlike the companion DARPA SBIR Release 6 instructions, does not include language stating that TABA is in addition to the cost ceiling and not subject to profit or fee. Treat the SBIR wording as not automatically transferring, and if the distinction affects your budget, ask DARPA before October 14. TABA funding requests will be reviewed by the respective contracting office or specialist at time of award to ensure compliance with TABA requirements.
For this topic, the highest-value TABA uses are intellectual property counsel, given that the feasibility documentation itself requires documented IP ownership or license rights, and manufacturing or foundry transition consulting, since foundry transferability is one of the four named integration constraints.
Questions and the FAQ
DSIP Topic Q&A will not be available for these DARPA topics. Technical questions related to improving the understanding of a topic's requirements must be submitted by October 14, 2026, by email to SBIR_BAA@darpa.mil with the topic number in the subject line, including the name, email address, and telephone number of a point of contact. All questions must be in English.
Questions submitted within seven calendar days of the proposal due date may not be answered. DARPA posts a consolidated Frequently Asked Questions document under the topic number summary on its Small Business site, updated on an ongoing basis until one week prior to the proposal due date.
DSIP technical support is available Monday through Friday, 9:00 a.m. to 5:00 p.m. Eastern, by email to DoDSBIRSupport@reisystems.com with a copy to SBIR_BAA@darpa.mil.
DARPA will not accept any late proposals.
Proposal format details
The Technical Volume must be a single PDF file including graphics. Perform a virus check before uploading, since a detected virus may cause rejection of the proposal. Do not lock or encrypt the file. Do not include or embed active graphics such as videos or moving pictures.
Number all pages consecutively. Font size should not be smaller than 10-point on standard 8.5 by 11 inch paper with one-inch margins. The header on each page of the Technical Volume should contain your company name, the topic number, and the proposal number assigned by DSIP when the Cover Sheet was created, and may be placed in the one-inch margin.
The Proposal Cover Sheet must include a brief technical abstract of no more than 3000 characters describing the proposed R&D project with a discussion of anticipated benefits and potential commercial applications.
Do not include marketing material. Marketing material will not be evaluated.
Classification, marking, and registrations
All proposals are required to be UNCLASSIFIED or CUI. Do not include any classified information in your proposal submission. Do not include any proprietary information on the Proposal Coversheet in Volume 1, as it may be released publicly if selected for award.
Proprietary or other CUI information may be included in the Technical Volume as needed, marked appropriately as CUI with the appropriate CUI Control Block on the first page of Volume 2. The Cost Volume should be marked CUI for PROPIN. Volumes 4 through 7 should be marked as appropriate based on content.
Proposal titles, abstracts, anticipated benefits, and keywords of proposals selected for contract award will undergo a DARPA Policy and Security Review and are subject to revision or redaction. Final approved versions may appear on the DoW SBIR/STTR awards website and the SBA's award website at sbir.gov/awards.
Proposers should ensure they have an accurate and active entity registration on SAM.gov. Firms engaging in Controlled Unclassified Information, Export Controlled, or ITAR work for DARPA must have CMMC Level 2 self-assessment certification. The projected requirement for this topic is Level 1, and this topic carries no ITAR or EAR restriction paragraph. DARPA points to sprs.csd.disa.mil/nistsp.htm and notes Project Spectrum at projectspectrum.io as an assistance resource.
On venture capital ownership
The DARPA STTR Release 6 instructions contain no provision addressing majority ownership by venture capital operating companies, hedge funds, or private equity firms. This is a real difference from the companion DARPA SBIR Release 6 instructions, which include an explicit section permitting such ownership under three conditions.
Do not read the SBIR provision across to this document. Eligibility for STTR awards is governed by the DoW STTR Program BAA and the SBA SBIR/STTR Policy Directive, and if your firm's ownership structure raises the question, get an answer from DARPA or from the DoW STTR Program BAA before you invest in a proposal rather than after.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW STTR Program BAA. DARPA will conduct an evaluation of each conforming proposal. Proposals that do not comply with the requirements detailed in this BAA and the research objectives of the corresponding topic are considered non-conforming and will not be evaluated nor considered for award.
There is a second, topic-specific non-responsiveness trap in Appendix A: proposals that do not adequately substantiate prior Phase I-equivalent feasibility for the components addressed will be deemed non-responsive and will not be evaluated for award.
Using the evaluation criteria, the Government will evaluate each proposal in its entirety, documenting the strengths and weaknesses relative to each criterion, and will determine the proposal's overall selectability for funding. Proposals will not be evaluated against each other but on their own individual merit.
A selectable proposal is one where the strengths of the overall proposal outweigh its weaknesses, with no accumulated weaknesses that would require extensive negotiations or a resubmitted proposal. A non-selectable proposal is one where the strengths do not outweigh its weaknesses.
Awards will be made to proposers whose proposals are determined to be the most advantageous to the Government, consistent with instructions and evaluation criteria specified in the DoW STTR Program BAA and availability of funding.
Proposing firms will be notified of selection or non-selection status within 90 calendar days of the closing date of the BAA. The Corporate Official indicated on the Proposal Cover Sheet will be notified by email. DARPA will provide a technical evaluation narrative to the proposer for each proposal submitted, and an informal feedback session may be requested by email at sbir@darpa.mil, provided at the sole discretion of DARPA.
Company Commercialization Report information will not be considered by DARPA during proposal evaluations.
Protests regarding the selection decision should be submitted, as prescribed in FAR 33.106(b) and FAR 52.233-3, to DARPA Contracts Management Office, 675 N. Randolph Street, Arlington, VA 22203, by email to CMO_SBIRProtests@darpa.mil and sbir@darpa.mil.
Post-award support
DARPA provides Transition and Commercialization Support Program services to Phase II awardees upon contract execution at no cost to awardees. Awardees may also be eligible for the Embedded Entrepreneurship Initiative, an invitation-only program at DARPA's sole discretion, typically no more than $310,000 per awardee over the duration of the award, supporting a Senior Commercialization Advisor relationship, investor working group connections, and hiring an embedded entrepreneur to execute a Go-to-Market strategy. Information in your commercialization strategy section is used to determine suitability for EEI participation, and selection for EEI is made independently after selection for award.
The References
Seven, and they map cleanly onto three arguments.
The problem is real and long-standing: Wu, Kumar, and Pamarthy, "High aspect ratio silicon etch: A review," Journal of Applied Physics, 2010. Huff, "Recent advances in reactive ion etching and applications of high-aspect-ratio microfabrication," Micromachines, 2021. Tang, Sandoughsaz, and Najafi, "Ultra high aspect-ratio and thick deep silicon etching (UDRIE)," IEEE MEMS, 2017.
The DoW pull: DARPA's Fast and Curious program, at darpa.mil.
The solution space DARPA has in mind: Esmeraldo Paiva et al., "High Aspect Ratio Nanoscale Pores through BCP-Based Metal Oxide Masks and Advanced Dry Etching," ACS Applied Materials and Interfaces, 2023. Bernet et al., "Highly selective anisotropic dry etching of smooth SiO2 nanostructures using SF6 plasma and Cr hard mask: Toward sustainable plasma etching," Journal of Vacuum Science and Technology B, 2026. Venkatram et al., "Two-dimensional crystalline hard masks for high-aspect-ratio nanofabrication," Nature Materials, 2026.
The last of these is the most consequential. A 2026 Nature Materials paper on two-dimensional crystalline hard masks is almost certainly the proximate inspiration for the "low-dimensional, continuously ordered, inherently grain-free" language in the design brief. Read it, and position your approach relative to it explicitly, whether you are extending it, competing with it, or doing something else entirely. The block-copolymer metal oxide mask reference points to a second recognized route. If your approach is neither, the burden is on you to explain why it beats both.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
Technical question deadline: October 14, 2026, to SBIR_BAA@darpa.mil with the topic number in the subject line
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Selection notification: within 90 calendar days of BAA close
Base period: 24 months from award
Option: 12 additional months if exercised
A working backward plan
Before September 23. Test yourself against the four feasibility categories with real data: measured selectivity above 50:1, aspect ratio above 25:1 with no degradation or faceting or CD loss, sidewall roughness under 5 nm RMS with named metrology, and a substantiated scale-up pathway addressing CMOS compatibility, thermal budget, foundry transferability, and application constraints. Confirm the feasibility work was substantially performed by you or your PI. Resolve IP ownership or licensing now, because documentation of ownership or license rights must be in the Technical Volume. Identify and commit your research institution partner, with a named faculty PI, specific instruments, and a task split that satisfies the STTR work-share minimums, and start the institution's subaward paperwork immediately, since university contracting offices are slow in October. Confirm cleanroom access that supports standard-size wafers and commercial DRIE or ICP tools. Download the mandatory Volume 2 and Volume 3 templates. Read the FAQ and keep rechecking it. Read the Venkatram 2026 Nature Materials paper and the two other solution-space references. Decide your contract type. Confirm SAM registration.
September 23 through October 5. Draft the 10 page feasibility documentation first, since it is the gate and since inadequate substantiation makes the proposal non-responsive. Include the reference list on the last page of the feasibility documentation, counting toward its page limit, and the one-page commercialization potential summary. Then draft the 20 page technical proposal against Appendix A's required sections, with the statement of work as a substantial portion. Address the patternability paradox explicitly. Send your questions to SBIR_BAA@darpa.mil early enough to matter.
October 6 through October 14. Build the cost volume in the mandatory Excel template across 24 months plus the 12 month option, and get your research institution's budget in at prime-level detail with Subcontractor Pricing Considerations. Substantiate every cost with paystubs, rate agreements, quotes, or invoices. Decide and document the Fundamental Research treatment for the university tasks. Price wafer counts for the statistically significant baseline comparison, alternative substrate materials, and metrology time. Draft the 5 page transition and commercialization strategy against Appendix A's nine required elements.
October 15 through October 18. Assemble Volume 5 with data rights assertions, IP documentation, CVs, subcontractor pricing considerations, and any optional advocacy or letters of intent that substantiate specific commercialization claims. Complete Volume 6 training and the Volume 7 foreign affiliations webform, remembering that the Corporate Official cannot certify until Volume 7 is submitted. Run compliance: single unlocked PDF, no embedded video, 10-point minimum font, consecutive page numbers, correct header on every page, 3000 character abstract, CUI marking on Volume 2 and Volume 3, mandatory Excel cost template, no marketing material.
October 19 through October 20. Submit and certify in DSIP. Confirm that mandatory supporting documents actually uploaded, since a completed proposal submission in DSIP does not indicate that they did.
Frequently Asked Questions
What is DARPA STTR topic DPA26TZ06-DV004?
DPA26TZ06-DV004, called SHIELDER, is a DARPA STTR Direct to Phase II topic titled "Scalable Hard-mask materials with Improved Etch resistance and Low Degradation for Extreme-aspect-Ratio fabrication," released under the DoW 2026 STTR Broad Agency Announcement, Release 6. The objective is to develop and demonstrate novel nanofabrication hard-mask materials with substantially improved plasma etch resistance, enabling extreme high-aspect-ratio structures with excellent pattern-transfer fidelity, minimal sidewall roughness, and precise dimensional control.
How much funding is available?
The base award is $1,500,000 over 24 months, with a $450,000 option over 12 months, for a maximum of $1,950,000 across 36 months if the option is exercised. DARPA will also provide up to $25,000 in Technical and Business Assistance for the Direct to Phase II.
When is the proposal deadline?
The topic opens September 23, 2026 and proposals are due October 21, 2026 through the Defense SBIR/STTR Innovation Portal at dodsbirsttr.mil. DARPA will not accept late proposals.
Can I submit a Phase I proposal?
No. This topic is soliciting Direct to Phase II proposals only.
What must my prior work already show?
Four things, documented with data. Measured etch selectivity exceeding 50:1 against target substrates under aggressive dry plasma, along with successful deposition, synthesis, or transfer of low-stress mask films. Proof-of-concept pattern transfer at an aspect ratio greater than 25:1 without significant mask degradation, faceting, or critical dimension loss. High-resolution metrology such as cross-sectional SEM, TEM, or AFM verifying sidewall roughness under 5 nm RMS. And initial feasibility data or a substantiated extrapolation showing a viable pathway to wafer-level scale, addressing CMOS compatibility, thermal budget, vendor and foundry transferability, and application-specific constraints.
Who has to have performed the feasibility work?
Work submitted within the feasibility documentation must have been substantially performed by the proposer or the Principal Investigator. Additionally, if the technology is subject to intellectual property, you must own the IP or have obtained license rights prior to proposal submission, with documentation included in the Technical Volume.
What are the Phase II performance targets?
Etch selectivity greater than 150:1 mask-to-substrate under aggressive plasma conditions, extreme aspect ratios of 100:1 without mask failure or CD loss or top-edge faceting, and sidewall and line-edge roughness under 2 nm RMS. The mask must also be readily patternable, and metallic or particulate contamination must be mitigated to levels acceptable for standard semiconductor, MEMS, or photonic fabrication lines.
What kind of material is DARPA looking for?
Approaches that inherently bypass the degradation mechanisms of traditional masks by using low-dimensional, continuously ordered, inherently grain-free, or self-regenerating structures. DARPA specifically identifies the problems with the state of the art as excessive required thickness, which induces stress and pattern distortion, and polycrystalline grain boundaries, which erode unevenly under ion bombardment and propagate roughness.
Does the solution have to work on silicon?
No. Solutions may target compound semiconductors, wide-bandgap materials, piezoelectric substrates, ceramics, heterogeneous material stacks, or any other microsystems platform relevant to the Department of War. Compatibility with existing or minimally-modified RIE, ICP, Bosch DRIE, or related plasma tools is strongly preferred.
Where must the Phase II demonstration take place?
In a standard cleanroom environment where the process can be scaled to production, supporting the processing of standard-size wafers using commercial Deep Reactive Ion Etching or other Inductively Coupled Plasma tools.
Do I have to compare against a baseline mask?
Yes. Patterning performance must be compared between the novel hard mask and a standard baseline such as silicon dioxide, silicon nitride, or a sputtered metal like TiN or Cr, and the proposed solution must deliver a statistically significant reduction in defect propagation and edge roughness.
Do I need a research institution partner?
Yes. STTR awards require a formal partnership with a single partnering research institution, with statutory minimum work shares for each party: at least 40 percent by the small business and at least 30 percent by the research institution. The DARPA instructions direct proposers to the DoW STTR Program BAA for those general requirements. Unlike topic DV005 in this same release, DV004 does not prescribe how the roles divide, so you define the split.
How long can my technical volume be?
The standard format is 35 pages. Feasibility documentation shall not exceed 10 pages, the technical proposal shall not exceed 20 pages, and the Phase II commercialization strategy shall not exceed 5 pages and should be the last section of the Technical Volume. Appendix A states that the commercialization strategy will not count against the proposal page limit, so confirm with DARPA how the three numbers combine if it affects your layout. Font must be at least 10-point on 8.5 by 11 inch paper with one-inch margins.
What CMMC level applies?
The projected requirement for this topic is CMMC Level 1. Firms engaging in Controlled Unclassified Information, Export Controlled, or ITAR work for DARPA more broadly must have CMMC Level 2 self-assessment certification.
Is this topic ITAR restricted?
No topic-level ITAR or EAR restriction paragraph appears on DPA26TZ06-DV004. The other two topics in this STTR release, DV005 and DV006, both carry one.
Can I ask questions through DSIP Topic Q&A?
No. DSIP Topic Q&A is not available for DARPA topics. Technical questions go by email to SBIR_BAA@darpa.mil by October 14, 2026, with the topic number in the subject line, and must include a point of contact name, email, and phone number. DARPA maintains a consolidated FAQ on its Small Business site.
What contract types can DARPA award?
FAR-based firm-fixed-price or cost-plus reimbursement contracts, or Other Transactions for Prototypes agreements under the authority of 10 U.S.C. 4022. Note that the companion DARPA SBIR Release 6 instructions cite 10 U.S.C. 4021 for the same instrument type.
Are the templates mandatory?
Yes. Templates for Volume 2 Technical Volume and Volume 3 Cost Volume are provided as attachments to the announcement and use of these templates is mandatory. The Volume 3 template is an Excel spreadsheet.
How much cost documentation do I need?
All proposed costs should be accompanied by documentation substantiating how the cost was derived, such as paystubs or a DCMA rate agreement for labor, contracts or historical invoices for consultants, and quotes, invoices, or market research for materials and equipment. Subcontractor and consultant costs must be detailed at the same level as prime costs and substantiated with Subcontractor Pricing Considerations under FAR 15.404-3(b). Subcontractors send unsanitized cost proposals to SBIR_BAA@darpa.mil.
Is cost sharing required?
No. Cost sharing is permitted but is not required and will not be an evaluation factor.
Are venture capital backed companies eligible?
The DARPA STTR Release 6 instructions contain no provision on majority ownership by venture capital operating companies, hedge funds, or private equity firms, unlike the companion DARPA SBIR Release 6 instructions, which explicitly permit it under three conditions. Do not assume the SBIR provision applies here. Check the DoW STTR Program BAA and the SBA SBIR/STTR Policy Directive, or ask DARPA before October 14.
What happens if my feasibility documentation is thin?
Appendix A states that proposals which do not adequately substantiate prior Phase I-equivalent feasibility for the components addressed will be deemed non-responsive and will not be evaluated for award. That is a separate and stricter trap than being scored poorly.
How will my proposal be evaluated?
Against the evaluation criteria in the DoW STTR Program BAA. Proposals are evaluated individually on their own merit rather than against each other. A selectable proposal is one where strengths outweigh weaknesses with no accumulated weaknesses requiring extensive negotiation or resubmission. Proposals that do not comply with the BAA requirements or the research objectives of the topic are non-conforming and are not evaluated.
Will I get feedback if not selected?
Yes. DARPA will provide a technical evaluation narrative for each proposal submitted, and an informal feedback session may be requested by email at sbir@darpa.mil, granted at DARPA's sole discretion.
Can I include advocacy letters?
Yes, optionally, and they do not count against the page limit. They should only be submitted to substantiate transition or commercialization claims actually made in your commercialization strategy, and DARPA asks you not to submit them merely for the sake of including them. Letters from Government personnel will not be considered. Faxed or separately emailed letters will not be accepted.
What is the commercial market?
Domestic manufacturing of secure high-density memory, advanced electro-optical components, and high-performance MEMS and RF components for electronic warfare and GPS-denied navigation on the defense side. Commercially, next-generation 3D semiconductor transistors, multiferroic memory and logic components, and high-coherence solid-state quantum devices.
Who do I contact with questions?
Technical questions go to SBIR_BAA@darpa.mil with the topic number in the subject line, by October 14, 2026. Administrative questions about the DARPA program and these proposal instructions also go to SBIR_BAA@darpa.mil. DSIP technical support is DoDSBIRSupport@reisystems.com with a copy to SBIR_BAA@darpa.mil. Feedback session requests go to sbir@darpa.mil.
Positioning Advice for Companies Considering This Topic
Lead with the mechanism, not the number. DARPA asked proposers to identify the underlying physical mechanisms responsible for enhanced pattern-transfer performance, and it named four structural strategies it has in mind. A proposal that opens with "our mask achieves 60:1 selectivity" is weaker than one that opens with "our mask is grain-free, which removes the uneven erosion pathway that limits sputtered metals, and here is the selectivity that follows."
Answer the patternability paradox in its own section. The mask must be readily patternable and extremely etch resistant. Name the two chemistries, give the selectivity in both directions, and show that patterning does not damage the property that makes the mask work. This is the requirement most likely to separate serious proposals from optimistic ones.
Report roughness like a metrologist. RMS values with no stated method, scan length, or instrument are not evidence to a reviewer who does this for a living. Give the AFM scan parameters or the SEM and TEM conditions, and be clear about the distinction between line-edge roughness and sidewall roughness, since the topic uses both terms and sets different targets at different milestones.
Take mask thickness reduction seriously as a selling point. It is in the metrics of interest and it is the mechanism behind the whole topic: thick masks cause stress, distortion, and instability. If your material delivers the same etch depth at a fraction of the thickness, that is the headline, and it is more persuasive to a fab than a selectivity ratio.
Name your cleanroom and your wafer size. The Phase II demonstration requires a standard cleanroom where the process can scale to production, running standard-size wafers on commercial DRIE or ICP tools. If your work to date is on coupons in a university lab, explain how you get to wafers, and budget it.
Budget the statistical comparison. "Statistically significant reduction in defect propagation and edge roughness" against a baseline mask means wafer counts, replicates, and a designed experiment. Most proposals will show two micrographs. Showing a plan for a real comparison is cheap differentiation.
Build the research institution partnership as a technical argument, not a compliance step. DV004 does not prescribe the role split, which means a reviewer reads yours as evidence of whether the partnership is genuine. The natural division is that the institution owns erosion physics, mechanism identification, and advanced metrology, while the small business owns scalable deposition, cleanroom demonstration, contamination control, and the foundry path. Name the institution, the PI, the instruments, and the tasks, and start the subaward paperwork immediately.
Resolve the Fundamental Research question early. With a university performing at least 30 percent of the work, whether those tasks are Fundamental Research affects publication rights and contract terms, and DARPA requires you to either separate that work into its own statement of work or identify it within the prime statement of work. Decide before you write the SOW, not after.
Address all four integration constraints by name. CMOS compatibility, thermal budget, vendor and foundry transferability, and application-specific constraints. Each is a sentence or a paragraph, and skipping any of them in the feasibility documentation is an easy weakness for an evaluator to write down.
Position against the 2026 Nature Materials two-dimensional crystalline hard mask paper. It is cited in the topic and it almost certainly shaped the design brief's language. Whether you are extending that work, competing with it, or pursuing a self-regenerating route instead, saying so explicitly shows you know the state of the art, which Appendix A requires you to persuade reviewers of.
Ask about the option period. The topic gives base milestones through Month 24 and no option milestones, while the award table lists $450,000 over 12 months. Send that question to SBIR_BAA@darpa.mil, and in the meantime propose an option scope that follows from the Month 24 end state, such as foundry qualification or pilot production, stating plainly that you are doing so absent stated milestones.
Do not import the SBIR document's provisions. Three concrete differences between this STTR document and the companion DARPA SBIR Release 6 instructions matter to a proposer: the OT authority citation, the absence of any venture capital ownership provision, and the absence of the language stating that TABA sits on top of the cost ceiling. If you are submitting to both programs this cycle, read both documents.
DARPA SBIR DPA26BZ06-DV029: ICU-in-a-Box, Autonomous Extracorporeal Multiple-Organ Support Therapies
Deadline: October 21, 2026
Funding Award Size: $2m
Description: Complete guide to DARPA SBIR Direct to Phase II topic DPA26BZ06-DV029, ICU-in-a-Box autonomous extracorporeal multiple-organ support. $1.5M plus $500K option. Closes October 21, 2026.
Quick Answer
DPA26BZ06-DV029 is a DARPA SBIR Direct to Phase II topic under the DoW 2026 SBIR Broad Agency Announcement, Release 6. DARPA wants a single, portable, battery-operated extracorporeal life support device that runs itself: one cannula into a central vein, and from that single access point the machine provides lung support, cardiovascular support, fluid resuscitation, and optionally electrolyte management and medication delivery, autonomously, for days, in a place where no critical care physician exists. The award is $1,500,000 over 24 months with a $500,000 option over 6 months. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The clinical logic is stated plainly. Severe hemorrhage is the number one cause of preventable combat deaths, followed by airway compromise. Following initial trauma, sepsis and multiple-organ failure predominate. Point of injury care has gotten much better, so more casualties survive the first hour, but in large-scale combat operations and in remote Special Operations settings, access to definitive medical care may not be available for days. More survivors plus fewer evacuation opportunities equals a population of casualties who need ICU-level organ support in a place that has none.
There is one scoping sentence that decides whether your idea fits. Approaches that only enable a user or users to interact with and control multiple different systems, meaning system of systems solutions, are not in scope. The goal is a single, portable device delivering all the desired clinical interventions through a single intravenous cannula. If your concept is an integrated cart of best-in-class modules with a unifying controller, this topic is not for you.
Topic At a Glance
Topic number: DPA26BZ06-DV029
Title: ICU-in-a-Box: Autonomous Extracorporeal Multiple-Organ Support Therapies
Agency: Defense Advanced Research Projects Agency (DARPA)
Solicitation: DoW 2026 Small Business Innovation Research Broad Agency Announcement, Release 6, DARPA Proposal Submission Instructions
Program type: Direct to Phase II (DP2). This topic solicits Direct to Phase II proposals only
Technical volume format: White Paper and Slide Deck. The white paper shall not exceed 20 pages and the slide deck shall not exceed 15 slides
Base award: $1,500,000
Base period of performance: 24 months
Option: $500,000 over 6 months, for a pre-clinical pilot study
Component Technology Priority Area: Biotechnology
Projected CMMC level requirement: Level 1
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic
Central architectural constraint: a single portable device, one patient-inserted cannula, preferably no larger than 15Fr (5mm), into the internal jugular or femoral vein
Explicitly out of scope: system of systems solutions, and autonomous cannulation or decision support for cannulation
Testing basis: large animal model testing, with IACUC and ACURO regulatory timelines built into the milestone schedule
Technical and Business Assistance: up to $25,000 per Phase II project, in addition to the cost ceiling
Topic Q&A: DSIP Topic Q&A is not available for DARPA topics. Technical questions go to SBIR_BAA@darpa.mil by October 14, 2026
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Submission portal: DSIP at dodsbirsttr.mil
Keywords: autonomous systems, life support systems, resuscitation, trauma, shock, casualties, ARDS, multiple organ failure
The Feasibility Bar, Which Is the First Thing to Check
This is a Direct to Phase II topic and DARPA is unusually numeric about what you must already have. Proposers must provide data demonstrating that the following has been achieved to be considered for award: a prototype, portable, battery operated extracorporeal system that can do four things.
One. Achieve at least 2 L/min of blood flow and lung support, meaning both oxygenation and ventilation.
Two. Deliver at least 75 mL/min oxygen.
Three. Exchange at least 40 mL/min carbon dioxide.
Four. At least one of the following three closed-loop algorithms.
An algorithm that maintains oxygenation and ventilation within specified ranges for SpO2 and EtCO2 by adjusting combinations of blood flow, sweep gas, and FiO2.
An algorithm that maintains mean arterial pressure within a specified range using at least one vasopressor and, as needed, fluid resuscitation, preferably blood and alternatively crystalloid.
An algorithm that controls blood flow, ultrafiltration, and monitoring of an extracorporeal renal replacement or blood purification system for removing potassium.
Read the fourth requirement carefully
The first three are hardware performance numbers. The fourth is the one that eliminates most applicants, because it requires that you already have a working closed-loop control algorithm, not a plan for one. Notice that you need only one of the three, and that they are ordered from most to least conventional in this space. The oxygenation and ventilation control algorithm is the most natural for an ECMO or ECCO2R company. The mean arterial pressure algorithm is the most natural for an autonomous resuscitation company, and DARPA cites the Pinsky 2024 porcine work as evidence that class of algorithm exists. The potassium removal algorithm is the most natural for a portable dialysis or blood purification company.
That structure tells you something about the intended applicant pool. DARPA expects one of three different kinds of company to walk in with one of three different kinds of demonstrated autonomy, and then to build the rest during Phase II. You are not expected to have all three. You are expected to have one, plus the hardware numbers, plus a credible plan to integrate the others.
The additional documentation asks
Beyond the four-part gate, DARPA lists specific things your feasibility documentation must address.
You must describe current O2 transfer and carbon dioxide removal rates that can be achieved by the proposed system. Give real measured numbers, not design targets.
The proposed system must directly interface with an oxygen source, such as a canister or an oxygen concentrator, to provide oxygenation and carbon dioxide removal to mitigate acidosis and lactate buildup. This is a hard interface requirement. A system that assumes wall gas is not a prolonged field care system.
Proposals should provide evidence for the maximum duration of support with their proposed system, ideally greater than 72 hours. Note "ideally." Duration is scored but not gated, and 72 hours is the aspiration that reflects the days-long evacuation delay DARPA is designing against.
Documentation should include, reference, or summarize all relevant information including but not limited to technical reports, test data, prototype designs and models, and performance goals and results.
Proposers must also describe the potential commercial applications as part of the feasibility substantiation.
What DARPA Is Actually Looking For
The objective
Develop, with large animal model testing, an autonomous, portable extracorporeal life support platform for prolonged field care that integrates resuscitation, cardiovascular and pulmonary support, and optionally electrolyte and medication delivery capabilities.
The clinical and operational problem
Severe hemorrhage is the number one cause of preventable combat deaths, followed by airway compromise, with increased artillery, thermobaric and incendiary munitions use potentially increasing lung injury prevalence in future conflicts. Following initial trauma, sepsis and multiple-organ failure predominate.
Improvements in point of injury care, tourniquets and hemostatic dressings and similar interventions, have significantly increased survival in early trauma phases. Such care is ideally followed by pre-hospital care from trained medical providers during casualty evacuation. However, as reflected by the large-scale combat operations in the Russia-Ukraine war and with Special Operations missions in remote, austere locations, access to definitive medical care may not be available for days, especially in conflict with near-peer adversaries.
More people surviving early phases of trauma combined with decreased evacuation opportunities highlights a pressing need for solutions that can provide advanced and sustained resuscitative care for the organ failure sequelae of severe trauma, such as hypotension, respiratory failure, and acute renal failure.
Why DARPA thinks this is now possible
The topic description contains a compressed technology readiness argument, and it is worth understanding because your proposal should either agree with it or explain where it is wrong.
A growing body of technical capabilities and supporting literature suggests that by focusing on the key drivers of mortality, advances in autonomous systems across several areas of critical care could be combined into a single ICU-in-a-Box.
Extracorporeal membrane oxygenation is an alternative to using intubation and mechanical ventilation, under deep sedation, for acute respiratory distress syndrome. Because an ECMO system is essentially just a pump and a membrane lung, it offers a strong opportunity to automate oxygenation support, along with other key combat casualty care needs.
Current ECMO systems require highly trained operators, largely because of the need to insert large cannulas and the associated thrombotic risks, but recent advances can mitigate these limitations. Assistive insertion devices and ECMO systems with smaller cannulas reduce the difficulty of insertion. Antithrombotic coatings, portable systems, and alternatives to conventional membrane oxygenators reduce thrombotic risk, at least over shorter periods. AI and machine learning advances in general, as well as ECMO management computer models, could support automation.
Once venous access is established for automated ECMO, it enables additional interventions. Automated algorithms have stabilized mean arterial pressure and heart rate for short periods following severe hemorrhage in porcine models. Renal replacement therapy could mitigate life-threatening electrolyte abnormalities. Extracorporeal blood purification techniques could remove inflammatory mediators, infectious agents, or toxins.
This SBIR seeks to combine such capabilities into a single, portable, integrated intervention and delivery platform for resuscitation, cardiovascular and lung support. There is additional potential to automate sedation, analgesia, and renal replacement therapies. Although these efforts remain relatively immature, their addition to the integrated system could further extend ICU-level, autonomous extracorporeal life support in the austere and operational setting.
The argument in one line: the venous access is the expensive part, and once you have paid for it you should get everything through it.
The scoping constraint that decides eligibility of the concept
DARPA is interested in novel approaches to develop a fully integrated, portable system that can provide key components of resuscitation during prolonged casualty care.
Approaches that only enable a user or users to interact with and control multiple different systems, meaning system of systems solutions, are not in scope. The goal is to create a single, portable device that is capable of providing all the desired clinical interventions through a single intravenous cannula inserted into a central vein, internal jugular or femoral.
Given the ultimate goal to be able to deploy the system in conditions where experienced medical care is not available, a primary goal of this SBIR is to develop and integrate algorithms that enable the system to be autonomous.
This is the sentence to read twice. A great many credible teams would naturally propose an integrated rack: an ECMO console, an infusion manifold, a hemofilter, and a supervisory controller. That is exactly what DARPA has ruled out. One device. One cannula. All interventions through it.
The Operating Concept, Which Defines Your Autonomy Requirements
DARPA lists assumptions and expectations that together describe who is standing next to this machine and what they can and cannot do. This is the most useful part of the topic for scoping your control system.
Remote monitoring and control are essential to engage remote expert guidance and intervention when the system experiences edge or unanticipated conditions.
A casualty is cannulated. Autonomous cannulation, or decision support for cannulation, is out of scope.
Local users will have basic medical training, for example a field medic, and can set up, initiate, and attach the automated system if guided. But once attached, the system should monitor the casualty itself and provide autonomous interventions to sustain mean arterial pressure, oxygenation, and ventilation. It should address alerts and alarms, meaning system functioning, autonomously, or engage a remote expert automatically without local operator intervention.
In addition to initiation of the device, local caregivers can be assumed to change IV bags after direction, retrieve and attach blood products after direction, and retrieve and attach medication after direction.
Sufficient displays to provide status information about the patient and the current and past treatments deployed by the system should be included.
It is also permissible to include a clinical decision support mode that a local operator could respond to in the event a remote expert is unavailable.
What that operating concept actually implies
The medic is a pair of hands, not a clinician. Every action the local user performs is prompted by the machine, described in the topic as "after direction." Your interface design problem is therefore instruction generation under stress, not clinical consultation.
Alarms must be handled without the local operator. This is a strong statement. Conventional ECMO alarm response is the entire justification for a bedside specialist. DARPA is asking you to either resolve the alarm autonomously or escalate to a remote expert automatically, with the medic in neither loop.
The remote expert is part of the architecture, not a fallback. Remote monitoring and control are called essential, which means bandwidth, latency, degraded-link behavior, and cyber security are design requirements rather than integration details. Note that network connection and cyber security appear explicitly in the SWaP-C3 requirement below.
The clinical decision support mode is permitted, not required. Offer it as a degraded operating mode for a denied-communications scenario and you address the obvious objection that a comms-dependent life support system is a fragile life support system.
Required Capabilities
DARPA lists these as requirements. Treat each as a section heading in your white paper.
Minimal patient contact and minimal sensing
The system must be readily deployable with minimal contact points with the patients.
Sensor packages should be minimized and be easily applied to monitor patient state. Proposals that do not require invasive arterial monitoring are preferred. Proposers must specify what sensors are being recommended, how they are intended to be used in austere and operational military conditions, and how they are providing the system sufficient physiological information to support the autonomy.
That last clause is the hard one. You are being asked to run closed-loop perfusion pressure control without an arterial line, which is the sensor a modern ICU would use for exactly this purpose. Your proposal has to close that observability gap explicitly: which noninvasive or intracircuit measurements substitute, what the error characteristics are, and how the controller stays safe when the estimate degrades.
One cannula
The system may have one patient-inserted cannula, preferably no larger than 15Fr (5mm). Proposers must specify which vessels they intend to use, jugular or femoral.
Fifteen French for a single-access circuit supporting 2 L/min or more of blood flow is a demanding hydraulic and hemolysis problem, and it is the reason the topic cites the 13Fr versus 15Fr cannula comparison literature. Address recirculation fraction, shear, and hemolysis directly.
Circuit design, medication delivery, and safety parameters
The intended circuit design must be provided and include how sufficient blood, fluids, and medications will be effectively delivered.
Proposals must specifically describe how the proposed solution manages continuous medication delivery, for example sedation and analgesia and vasopressors, and bolus fluids, for example blood or crystalloid, within the integrated circuit.
Proposers must also specify safety parameters that will be implemented in the integrated system.
Note "within the integrated circuit." Vasopressors, sedation, and blood products all enter through the same single access. That means your circuit has to handle continuous low-rate infusion and rapid bolus volume in the same plumbing, with the mixing, dosing accuracy, and air management problems that implies.
SWaP-C3 and the backpack
As DARPA ultimately envisions a system intended for use in prolonged field care scenarios, proposers must describe the planned size, weight, and power, plus network connection, cyber security, and cost considerations, together abbreviated SWaP-C3, for the system.
It is anticipated that further system development for miniaturization may occur beyond the scope of this SBIR, but the potential to integrate all the provided components in a portable form factor, for example a M9 backpack, must be demonstrated.
Read the concession and the requirement together. You do not have to deliver a backpack in 24 months. You do have to demonstrate the potential to get there, which in practice means a component-level mass, volume, and power budget that closes against a stated backpack envelope, plus a bill of materials, which is a named Month 24 deliverable.
Lung support
For lung support, the system will be able to provide ECMO functionality for direct oxygen transfer to the blood.
Preferably, full oxygen support, approximately 250 mL/min, would be achievable. In cases where partial oxygen support is proposed, at least 125 mL/min, proposers can assume that the patient is already intubated for respiratory support. Proposals that integrate with mechanical ventilation when present are preferred.
In addition to oxygen support, the system should also provide at least 80 mL/min CO2 removal.
There is a real strategic choice here. Full support at roughly 250 mL/min oxygen is the ambitious path and removes the ventilator from the field equipment set. Partial support at 125 mL/min is permitted, but you then inherit an intubated patient, which reimports the sedation, airway management, and ventilator logistics that the topic's own opening argument was trying to avoid. If you choose partial support, say why, and take the stated preference for ventilator integration seriously.
Cardiovascular support and the hemorrhage model
For cardiovascular support, the system must also be able to provide fluid resuscitation as needed.
It must be capable of maintaining perfusion pressure, not just blood pressure, throughout the validation tests, for example at least 24 hours, and be able to support significant hemorrhagic trauma models, for example a model emulating an uncontrolled bleeding scenario of at least 50 percent of the estimated blood volume over six hours.
This support can include blood, artificial blood products, freeze dried blood products, and crystalloid.
When fluid is ineffective at maintaining perfusion pressure, the system should titrate vasopressors to maintain specified mean arterial pressure.
All interventions must be autonomously controlled by the system.
The parenthetical "not just blood pressure" is a deliberate clinical distinction and you should honor it. A controller that holds a MAP number while tissue perfusion fails is the classic failure mode of pressure-targeted resuscitation. Explain what your system uses as a perfusion surrogate, whether lactate, mixed venous saturation, urine output, capillary refill, or an intracircuit measure, and how the controller arbitrates between fluid and vasopressor.
Algorithm development and validation
Proposers must describe the planned development and validation of the closed-loop auto-titration algorithms for control of the oxygen saturation level.
Likewise, they must describe the planned development and validation of the fluid resuscitation algorithms, as well as integration of the oxygenation and resuscitation algorithms into the overall system.
The word "integration" is doing work. Two independently correct controllers, one moving blood flow and sweep gas for gas exchange and one moving volume and vasopressor for pressure, are coupled through the same circuit and the same patient. Increasing blood flow to improve oxygenation changes preload and pressure. Describe the arbitration or supervisory layer, not just the two loops.
Animal model design
Proposers must clearly describe the design of all proposed systems and large animal model tests, as well as their relevance to clinically relevant scenarios, such as similarity to human intravascular blood flow, changes in blood oxygenation levels, trauma biomarkers such as lactate, potassium, and pH, thrombotic and hemolysis risk checks, and histological assessments.
Proposers may choose to utilize existing benchtop hardware and approaches for validation of the system functionality, or, alternatively, respondents may propose the development of iterative or new validation approaches as part of their proposal.
Optional Capabilities That Strengthen a Proposal
Additional capabilities of interest include renal replacement therapy and fluid management interventions to treat electrolyte derangements such as hyperkalemia.
For these, the system would provide basic kidney support via the same circuit that supports oxygenation and fluid resuscitation. This would include providing electrolyte removal and fluid resuscitation to compensate for acute metabolic derangements. This could include filtering the blood to remove small molecules, or other interventions such as absorption, to compensate for cytokines, endotoxins, and similar mediators that can impact sepsis or other inflammatory derangements.
More specifically the system should be demonstrated to be able to maintain adequate lactate levels, temperature, and pH, approximately within 20 percent of baseline values, for at least a 24-hour period.
Additional functionality that is not required but is of interest, and would boost the strength of a proposed solution, includes automated sedation and analgesia, and delivery of scheduled medications such as antibiotics or other as-needed medications such as bolus pain control.
Two notes. First, "via the same circuit" again. The optional capabilities are not bolt-ons, they must live inside the single-access architecture. Second, the temperature target inside the renal and metabolic bullet is easy to miss, and it implies thermal management of an extracorporeal circuit in an austere environment, which is a power budget item.
The Milestone Schedule, Which Is Effectively a Work Plan
DARPA specifies interim and end-phase goals for a 24-month base effort followed by a 6-month option, and says responders are strongly encouraged to propose additional interim assessments to further demonstrate progress toward the stated goals.
Month 1. Report on initial circuit architectures, sensors, algorithms, animal models, development strategy, and timeline.
Month 3. Interim report on integration of extracorporeal support subsystems, meaning sensor packages, lung and cardiovascular systems, plus any planned optional capabilities such as electrolyte support. Interim report on development of automation algorithms. Report on development of initial ex vivo and computational simulation testing environments.
Month 6. Report on development strategy for large animal models. Animal models may be used for testing each subsystem separately, at 6 hours for hemorrhage models and 12 hours for other interventions, including regulatory IACUC and ACURO timelines.
Month 9. Interim report describing performance, ex vivo and simulation, of the prototype system and automation algorithms, and refinements to ex vivo testing environments. Report on regulatory status and approval of large animal models.
Month 12. Interim report describing performance, ex vivo and simulation, of the prototype system and automation algorithms. Interim report describing development status of large animal models.
Month 15. Interim report describing in vivo performance of the prototype system in separate animal models. Report on clinical regulatory approval strategy, for example FDA engagements.
Month 18. Interim report describing in vivo performance of the prototype system in separate animal models. Report on strategy for developing a large animal model that combines assessments for all subsystems, for example a polytrauma model, in order to validate system performance during simultaneous deployment of all automation algorithms, maintaining hemostasis for 24 hours, including regulatory IACUC and ACURO timelines.
Month 21. Report on regulatory status and approval of the large animal polytrauma model.
Month 24. Final report summarizing approach, prototype architecture and algorithms, and separate large animal model testing results including thrombotic, hemolysis, and histological assessments as well as efficacy of algorithms in maintaining goal biomarkers such as SpO2, EtCO2, and MAP. Bill of materials for the prototype, plus SWaP-C3 feasibility for future fully portable designs, and use of consumables such as oxygen, fluids, and vasopressors. Report on regulatory development strategy and any FDA engagements. Report on development of the polytrauma model and recommendations for its use in a preclinical study for in vivo assessments of the prototype device, including updated recommendations for a regulatory strategy.
Option Period 1
Proposals that respond to Option Period 1 may also include a sequential 6-month option period for a pre-clinical pilot study to demonstrate and validate the functionality of their integrated breadboard system in a large animal model.
Month 25. Report on regulatory approval status for the 24-hour large animal model that combines all interventions, for example the polytrauma model. Report on experimental testing strategy and timeline.
Month 28. Interim report describing integration of all prototype subsystems and automation algorithms. Interim report describing in vivo performance of the prototype system in a 24-hour large animal polytrauma model.
Month 30. Final report documenting final prototype architecture and algorithms, methods, all results, and proposed strategy for a portable form factor compatible with far forward deployment in austere conditions, meaning a SWaP-C3 assessment, and final regulatory status and FDA engagements.
What the schedule tells you about risk
The schedule is built around animal regulatory approval, and it appears twice: subsystem models approved by Month 9, and a combined polytrauma model approved by Month 21. That is the schedule's critical path, not the engineering. IACUC plus ACURO review is a multi-month process that you do not control. A proposal that treats these as line items rather than as the dominant schedule risk is not being honest about the plan, and DARPA has told you where to look by writing regulatory timelines into two separate milestones.
Note also the sequencing logic. Subsystems are validated in separate animal models during the base period. The combined polytrauma model, with all algorithms running simultaneously for 24 hours, is what Option Period 1 buys. So the base period ends with a system that has been proven in pieces, and the option proves it whole. Your Month 24 deliverable set is written accordingly: a bill of materials, a SWaP-C3 feasibility assessment, and a recommendation for the preclinical study rather than the study itself.
The word "breadboard" in the Option Period 1 description is a deliberate expectation-setter. DARPA is not expecting a packaged product at Month 30.
Phase III Dual Use
This SBIR has potential applicability across DoW and commercial entities.
Commercial applications include integration of this autonomous extracorporeal support platform into civilian critical care transport, meaning rotor-wing, fixed-wing, and ground ambulance services; rural and community hospital use as a bridge-to-transfer where on-site critical care expertise is unavailable; and mass-casualty or disaster-response settings where trained critical care staff are scarce or evacuation times are long.
Military and DoW applications include incorporation into Prolonged Casualty Care and equipment sets for Special Operations Forces, forward surgical teams, and casualty evacuation platforms across air, ground, and maritime domains, including uncrewed evacuation systems. This enables extended en-route and prolonged field care for casualties with severe injuries, battle or disease non-battle, in contested or resource-limited operational environments. In these settings, an autonomous, portable extracorporeal life support system will provide options to sustain the force that do not currently exist.
The civilian case here is stronger than in most DARPA medical topics, because the same staffing shortage exists in the commercial market. ECMO capability is concentrated in a small number of high-volume centers, and the limiting resource is trained perfusion and intensive care staff rather than hardware. A device that reduces operator skill requirements addresses a documented civilian access problem, which is a real Phase III story rather than a courtesy paragraph. The rotor-wing and fixed-wing transport market in particular already pays for portable, battery-operated, oxygen-canister-fed equipment.
Funding, Cost Structure, and DARPA Mechanics
The award
$1,500,000 over a 24 month base, plus a $500,000 option over 6 months, for $2,000,000 across 30 months if the option is exercised.
The resources made available for each topic will depend on the quality of the proposals received and the availability of funds. The Government reserves the right to award all, some, one, or none of the options based on available funding and the performer's technical performance.
Be realistic about what $1.5 million buys against this scope. Large animal studies alone, across multiple separate subsystem models with IACUC and ACURO overhead, consume a substantial share. This is a topic where existing hardware, an existing animal facility relationship, and an existing regulatory pathway are worth more than a large engineering team.
Contract type, which you must elect
DARPA may award FAR-based contracts, firm-fixed-price or cost-plus reimbursement, or Other Transactions for Prototype under the authority of 10 U.S.C. 4021, subject to approval of the Contracting Officer or Agreements Officer respectively. Proposers must state their requested contract type in their proposal.
Cost-plus reimbursement requires including your Defense Contract Management Agency Final Determination Letter showing approval of your accounting system. An Other Transaction for Prototype requires including a completed OT using the Model OT for Prototype from the DARPA Small Business site, plus completed OT Certifications, both loaded in Volume 5, with at minimum the color-coded areas completed and redlines with explanations for any article you wish to negotiate. Firm-fixed-price requires no additional action.
For a program with animal regulatory schedule risk outside your control, the choice of contract type is worth real thought rather than defaulting to firm-fixed-price.
Templates are mandatory
Templates for Volume 2 Technical Volume and Volume 3 Cost Volume are provided as attachments on the DARPA Small Business website. Use of the DARPA Cost Proposal template is mandatory.
Technical and Business Assistance
Phase II awardees may request up to $25,000 per Phase II project. TABA funding is in addition to the cost ceilings and is not subject to profit or fee. Requests will be reviewed by the respective contracting office or specialist at time of award.
For this topic, regulatory consulting is the highest-value use of TABA. An autonomous, closed-loop, drug-delivering extracorporeal life support device is a genuinely novel regulatory object, and DARPA has already put FDA engagement strategy into the Month 15 and Month 24 deliverables. Getting experienced device regulatory counsel involved early is directly aligned with a contract requirement.
Questions and the FAQ
DSIP Topic Q&A will not be available for these DARPA topics. Technical questions must be submitted by October 14, 2026, by email to SBIR_BAA@darpa.mil with the topic number in the subject line, including the name, email address, and telephone number of a point of contact. Questions submitted within seven calendar days of the proposal due date may not be answered. DARPA posts a consolidated Frequently Asked Questions document under the topic number summary on its Small Business site, updated on an ongoing basis until one week prior to the proposal due date.
DARPA will not accept late proposals.
Classification, marking, and registrations
All proposals are required to be UNCLASSIFIED or CUI. Do not include any classified information in your proposal submission. Do not include any proprietary information on the Proposal Coversheet in Volume 1.
Proposal titles, abstracts, anticipated benefits, and keywords of proposals selected for contract award will undergo a DARPA Policy and Security Review and are subject to revision or redaction by DARPA. Final approved versions may appear on the DoW SBIR/STTR awards website and the SBA's award website at sbir.gov/awards.
Proposers should ensure they have an accurate and active entity registration on SAM.gov. Those engaging in ITAR or CUI work for DARPA must have CMMC Level 2 certification, though the projected requirement for this topic is Level 1. DARPA points to sprs.csd.disa.mil/nistsp.htm and notes Project Spectrum at projectspectrum.io as an assistance resource.
Venture capital, hedge fund, and private equity ownership
Proposers that are more than 50 percent owned by multiple venture capital operating companies, hedge funds, private equity firms, or any combination of these as set forth in 13 CFR 121.702 are eligible to submit proposals in response to DARPA topics advertised within this BAA. Three conditions apply: register with the SBA Company Registry Database before submitting; submit the Majority-Owned VCOC, HF, and PEF Certification, with the SBIR VC Certification available on the DARPA Small Business site, in Supporting Documents Volume 5; and immediately notify the Contracting Officer, register in the appropriate SBA database, and submit the required certification if you enter that ownership class after submitting but before receiving a funding agreement.
This matters more here than on most topics. Medical device companies with working extracorporeal hardware are almost always venture funded, and DARPA's permissive posture keeps that pool eligible.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW SBIR Program BAA. DARPA will conduct an evaluation of each conforming proposal. Proposals that do not comply with the requirements detailed in this BAA and the research objectives of the corresponding topic are considered non-conforming and are therefore not evaluated nor considered for award.
Using the evaluation criteria, the Government will evaluate each proposal in its entirety, documenting the strengths and weaknesses relative to each evaluation criteria, and based on those will determine the proposal's overall selectability for funding. Proposals will not be evaluated against each other but on their own individual merit.
A selectable proposal is one where the strengths of the overall proposal outweigh its weaknesses, with no accumulated weaknesses that would require extensive negotiations or a resubmitted proposal. A non-selectable proposal is one where the strengths do not outweigh its weaknesses.
Proposing firms will be notified of selection or non-selection status within 90 calendar days of the closing date of the BAA. The Corporate Official indicated on the Proposal Cover Sheet will be notified by email. In accordance with the SBA SBIR/STTR Policy Directive, Appendix I, paragraph 4, subparagraph (d), DARPA will provide a technical evaluation narrative to the proposer for each proposal submitted in response to a topic. An informal feedback session may additionally be requested via email at sbir@darpa.mil, provided at the sole discretion of DARPA.
Company Commercialization Report information will not be considered by DARPA during proposal evaluations.
Protests regarding the selection decision should be submitted, as prescribed in FAR 33.106(b) and FAR 52.233-3, to DARPA Contracts Management Office, 675 N. Randolph Street, Arlington, VA 22203, by email to CMO_SBIRProtests@darpa.mil and sbir@darpa.mil.
Note the "non-conforming" language against the scoping constraint. A system of systems proposal is arguably non-conforming with the research objectives of the topic rather than merely weak, which means it may not be evaluated at all.
Post-award support
DARPA provides Transition and Commercialization Support Program services to Phase II and DP2 awardees upon contract execution at no cost to awardees. Awardees may also be eligible for the Embedded Entrepreneurship Initiative, an invitation-only program at DARPA's sole discretion, typically no more than $310,000 per awardee over the duration of the award, supporting a Senior Commercialization Advisor relationship, investor working group connections, and hiring an embedded entrepreneur to execute a Go-to-Market strategy.
The References
Twenty of them, the longest reference list in this DARPA release, and they map the proposal you are expected to write.
The combat casualty care case: Eastridge 2012 on death on the battlefield 2001 to 2011; Remondelli 2023 on casualty care implications of large-scale combat operations; Neff 2013 on extracorporeal organ support following trauma; Chovanes 2012 on the evolution of damage control surgery; Cohen 2012 on hemostatic resuscitation; Kotwal 2016 on the golden hour policy; Keenan and Riesberg 2017 on prolonged field care beyond the golden hour; Epstein 2023 on lessons from the war in Ukraine.
The ECMO technology case: Geetha 2024, a comprehensive review of extracorporeal membrane oxygenation; Szentgyorgyi 2025 on the evolution of ECMO and advanced developments; Conrad on the clinical trials of the intravascular oxygenator; Osmani 2026 comparing 13Fr and 15Fr arterial return cannulas for V-A ECMO in ECPR; Combes 2024, the European expert consensus on ECCO2R for acute hypoxemic respiratory failure, which is the source of the 80 mL/min CO2 removal figure.
The automation case: Pladet 2023 on clinical decision support for ECMO, subtitled "Will we fly by wire?"; Pinsky 2024 on autonomous precision resuscitation during ground and air transport of an animal hemorrhagic shock model; Brattain 2021 on an AI-enabled, ultrasound-guided handheld robotic device for femoral vascular access; Daga 2026 on closed-loop anesthesia; Nagata 2023 on automated control of propofol, remifentanil, and rocuronium versus anesthesiologist management; Aissou 2012 on pupillary reflex measurement for objective analgesia assessment; Wieringa 2025 on portable, wearable, and implantable artificial kidneys.
Three of these deserve particular attention. Pinsky 2024 is the existence proof for the mean arterial pressure algorithm option in the feasibility gate, and it is an animal transport study, which is close to the operational scenario. Combes 2024 is where the CO2 removal number comes from, so your gas exchange claims should be framed in its terms. Pladet 2023 is the closest thing in the literature to what DARPA is asking for on the control side, and its title question is the one your proposal has to answer.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
Technical question deadline: October 14, 2026, to SBIR_BAA@darpa.mil with the topic number in the subject line
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Selection notification: within 90 calendar days of BAA close
Base period: 24 months from award
Option Period 1: 6 additional months if exercised
A working backward plan
Before September 23. Test yourself against the four-part feasibility gate with measured data, not projections: 2 L/min blood flow with oxygenation and ventilation, 75 mL/min oxygen delivery, 40 mL/min CO2 exchange, and at least one working closed-loop algorithm from the three listed. Assemble the technical reports, test data, prototype designs, and performance results that substantiate it. Measure and document your actual maximum duration of support, and note how it compares to the greater-than-72-hour aspiration. Confirm your system interfaces directly with a canister or concentrator oxygen source. Decide jugular versus femoral and fix your cannula size. Line up your large animal facility and start the IACUC conversation now, because approval timelines are the schedule's critical path and appear in two separate milestones. Download the mandatory DARPA Volume 2 and Volume 3 templates. Read the FAQ and keep rechecking it. Decide your contract type and prepare the corresponding documents. Confirm SAM registration. If venture-backed, register with the SBA Company Registry and obtain the SBIR VC Certification.
September 23 through October 5. Draft the 20 page white paper and 15 slide deck. Lead with the feasibility evidence, then the single-device architecture, then the autonomy and remote expert concept of operations, then the required capabilities in DARPA's own order, then the animal model and regulatory plan. State plainly and early that your solution is a single device through a single cannula, because that is the screening question.
October 6 through October 14. Build the cost volume in the mandatory template across 24 months plus the 6 month option. Price large animal studies realistically, including per-study facility cost, veterinary support, IACUC and ACURO administrative burden, and the difference between 6-hour hemorrhage models and 12-hour intervention models. Price disposables, oxygenator and filter development, consumables, and regulatory consulting. Consider whether TABA covers the regulatory work.
October 15 through October 18. Assemble Volume 5 with contract-type documents and certifications, complete Volume 7 and the Volume 4 CCR, and run compliance: 20 page white paper, 15 slide deck, unclassified or CUI only, no proprietary information on the coversheet, mandatory cost template.
October 19 through October 20. Submit and certify in DSIP.
Frequently Asked Questions
What is DARPA SBIR topic DPA26BZ06-DV029?
DPA26BZ06-DV029 is a DARPA SBIR Direct to Phase II topic titled "ICU-in-a-Box: Autonomous Extracorporeal Multiple-Organ Support Therapies," released under the DoW 2026 SBIR Broad Agency Announcement, Release 6. The objective is to develop, with large animal model testing, an autonomous, portable extracorporeal life support platform for prolonged field care that integrates resuscitation, cardiovascular and pulmonary support, and optionally electrolyte and medication delivery capabilities.
How much funding is available?
The base award is $1,500,000 over 24 months, with a $500,000 option over 6 months for a pre-clinical pilot study, for a maximum of $2,000,000 across 30 months if the option is exercised. Phase II awardees may also request up to $25,000 in Technical and Business Assistance, which is in addition to the cost ceiling.
When is the proposal deadline?
The topic opens September 23, 2026 and proposals are due October 21, 2026 through the Defense SBIR/STTR Innovation Portal at dodsbirsttr.mil. DARPA will not accept late proposals.
Can I submit a Phase I proposal for this topic?
No. This topic solicits Direct to Phase II proposals only.
What must my existing prototype already do?
You must provide data showing a prototype, portable, battery operated extracorporeal system that achieves at least 2 L/min of blood flow with lung support, delivers at least 75 mL/min oxygen, exchanges at least 40 mL/min carbon dioxide, and includes at least one of three closed-loop algorithms: oxygenation and ventilation control within SpO2 and EtCO2 ranges via blood flow, sweep gas, and FiO2; mean arterial pressure maintenance using at least one vasopressor plus fluid as needed; or control of blood flow, ultrafiltration, and monitoring for potassium removal in an extracorporeal renal replacement or blood purification system.
Do I need all three algorithms?
No. You need at least one demonstrated. The three options correspond to three different kinds of company, and the expectation is that you bring one demonstrated autonomy capability and build the others during Phase II.
Is a system of systems approach acceptable?
No. DARPA states that approaches which only enable a user or users to interact with and control multiple different systems, meaning system of systems solutions, are not in scope. The goal is a single, portable device providing all the desired clinical interventions through a single intravenous cannula. Because non-conforming proposals are not evaluated, this is a screening issue rather than a scoring issue.
How many cannulas can my system use?
One. The system may have one patient-inserted cannula, preferably no larger than 15Fr (5mm), and proposers must specify whether they intend to use the internal jugular or the femoral vein.
Is autonomous cannulation part of this topic?
No. DARPA states that a casualty is assumed to be cannulated, and that autonomous cannulation or decision support for cannulation is out of scope.
What oxygen and CO2 performance does Phase II target?
Preferably full oxygen support of approximately 250 mL/min. Partial oxygen support of at least 125 mL/min is acceptable, but then you may assume the patient is already intubated for respiratory support, and proposals that integrate with mechanical ventilation when present are preferred. The system should also provide at least 80 mL/min CO2 removal.
What hemorrhage scenario must the system handle?
The system must be able to support significant hemorrhagic trauma models, for example a model emulating an uncontrolled bleeding scenario of at least 50 percent of the estimated blood volume over six hours, and must maintain perfusion pressure, not just blood pressure, throughout validation tests of at least 24 hours.
Can I use an arterial line for monitoring?
You can, but proposals that do not require invasive arterial monitoring are preferred. Sensor packages should be minimized and easily applied, and you must specify which sensors you recommend, how they work in austere and operational military conditions, and how they provide the system sufficient physiological information to support the autonomy.
Who operates the system in the field?
A local user with basic medical training, such as a field medic, who can set up, initiate, and attach the system if guided. After attachment the system monitors the casualty itself and provides autonomous interventions for mean arterial pressure, oxygenation, and ventilation. It must handle alerts and alarms autonomously or engage a remote expert automatically without local operator intervention. Local caregivers can be assumed to change IV bags, retrieve and attach blood products, and retrieve and attach medication, all after direction from the system.
Is remote connectivity required?
Yes. DARPA states that remote monitoring and control are essential to engage remote expert guidance and intervention when the system experiences edge or unanticipated conditions. Network connection and cyber security are also explicit elements of the SWaP-C3 description you must provide. A clinical decision support mode for use when a remote expert is unavailable is permitted.
Do I have to deliver a backpack-sized system?
Not within this SBIR. DARPA anticipates that further miniaturization may occur beyond the scope of this effort, but you must demonstrate the potential to integrate all provided components in a portable form factor, for example a M9 backpack, and a SWaP-C3 feasibility assessment plus a bill of materials are Month 24 deliverables.
What are the optional capabilities?
Renal replacement therapy and fluid management to treat electrolyte derangements such as hyperkalemia, provided through the same circuit, with demonstrated maintenance of lactate, temperature, and pH within approximately 20 percent of baseline for at least 24 hours. Also of interest, though not required, are automated sedation and analgesia and delivery of scheduled or as-needed medications.
What animal testing is expected?
Large animal model testing throughout. During the base period, separate models validate individual subsystems, at 6 hours for hemorrhage models and 12 hours for other interventions. A combined polytrauma model that runs all automation algorithms simultaneously for 24 hours is developed during the base period and executed in Option Period 1. IACUC and ACURO regulatory timelines appear explicitly in the Month 6 and Month 18 milestones.
What is the biggest schedule risk?
Animal regulatory approval. DARPA writes IACUC and ACURO timelines into two separate milestones and requires reports on regulatory approval status at Months 9 and 21. Those reviews are not under your control and gate the in vivo work that everything else depends on.
How long can my technical volume be?
The white paper shall not exceed 20 pages and the slide deck shall not exceed 15 slides.
Can I ask questions through DSIP Topic Q&A?
No. DSIP Topic Q&A is not available for DARPA topics. Technical questions go by email to SBIR_BAA@darpa.mil by October 14, 2026, with the topic number in the subject line. DARPA maintains a consolidated FAQ document on its Small Business site under the topic number summary.
Do I have to choose a contract type?
Yes. Proposers must state their requested contract type: FAR-based firm-fixed-price, FAR-based cost-plus reimbursement, or an Other Transaction for Prototype under 10 U.S.C. 4021. Cost-plus requires your DCMA Final Determination Letter, and an Other Transaction requires the completed Model OT and OT Certifications in Volume 5.
Is the cost template mandatory?
Yes. Use of the DARPA Cost Proposal template is mandatory, and templates for both Volume 2 and Volume 3 are attachments on the DARPA Small Business website.
What CMMC level applies?
The projected requirement for this topic is CMMC Level 1. Firms engaging in ITAR or CUI work for DARPA more broadly must have CMMC Level 2 certification.
Are venture capital backed companies eligible?
Yes. Firms more than 50 percent owned by multiple venture capital operating companies, hedge funds, private equity firms, or any combination as set forth in 13 CFR 121.702 may propose, provided they register with the SBA Company Registry Database before submitting, include the Majority-Owned VCOC, HF, and PEF Certification in Volume 5, and notify the Contracting Officer if they enter that ownership class after submission but before award.
How will my proposal be evaluated?
Against the evaluation criteria in the DoW SBIR Program BAA. Proposals are evaluated individually on their own merit rather than against each other. A selectable proposal is one where strengths outweigh weaknesses with no accumulated weaknesses requiring extensive negotiation or resubmission. Proposals that do not comply with the BAA requirements or the research objectives of the topic are non-conforming and are not evaluated.
Will I get feedback if not selected?
Yes. DARPA will provide a technical evaluation narrative for each proposal submitted, and an informal feedback session may be requested by email at sbir@darpa.mil, granted at DARPA's sole discretion.
What is the commercial market?
Civilian critical care transport by rotor-wing, fixed-wing, and ground ambulance; rural and community hospitals as a bridge-to-transfer where on-site critical care expertise is unavailable; and mass-casualty or disaster-response settings where trained critical care staff are scarce or evacuation times are long.
Who do I contact with questions?
Technical questions go to SBIR_BAA@darpa.mil with the topic number in the subject line, by October 14, 2026. General DARPA SBIR inquiries and feedback session requests go to sbir@darpa.mil.
Positioning Advice for Companies Considering This Topic
Answer the single-device question in your first paragraph. The scoping sentence ruling out system of systems solutions is the most consequential line in the topic, and a non-conforming proposal is not evaluated at all. Do not bury the architecture claim on page eight. State on page one that your solution is one portable device delivering all interventions through one cannula, and then prove it with a circuit diagram.
Pick your feasibility algorithm and lead with the data. You need one of the three closed-loop algorithms demonstrated, and which one you have tells DARPA what kind of company you are. Show the measured performance, the animal or bench conditions, the control bounds, and the failure behavior. Then explain how you extend to the other functions, because that extension is what the $1.5 million buys.
Solve the observability problem out loud. DARPA prefers no invasive arterial monitoring while requiring closed-loop perfusion pressure control. That tension is the technical heart of the proposal. Name your sensors, state their accuracy under motion, temperature, and hypoperfusion, explain how you estimate perfusion rather than just pressure, and describe controller behavior when the estimate degrades.
Take "perfusion pressure, not just blood pressure" literally. It is a clinically meaningful distinction and a reviewer with a critical care background will look for it. Explain what surrogate you use, how fluid and vasopressor are arbitrated, and how you avoid the well-known failure of hitting a MAP target while tissue perfusion collapses.
Describe the supervisory layer, not just the loops. Gas exchange control and hemodynamic control are coupled through one circuit and one patient. Increasing blood flow for oxygenation changes preload. Someone has to arbitrate. DARPA asked specifically about integration of the oxygenation and resuscitation algorithms, and that arbitration logic is the answer.
Design for the medic who is not a clinician. Every local action in the topic is "after direction." Your interface generates instructions, not consultations. Show what the display says, how the machine asks for a blood product, and what happens when the medic is doing something else.
Treat the remote expert link as a design requirement and then defeat the obvious objection. Remote monitoring and control are called essential, which invites the question of what happens in a denied-communications environment. The permitted clinical decision support mode is your answer. Build it, describe the degraded operating modes, and specify how long the system can run fully autonomously with no link.
Put the animal regulatory pathway at the front of your schedule risk discussion. IACUC and ACURO timelines appear in two milestones and approval status reports at Months 9 and 21. If you already have an approved protocol at a facility with a hemorrhagic shock or polytrauma model, that is your single strongest schedule credential. Say so early and name the facility.
Close the SWaP-C3 budget on paper. You are not required to build a backpack, but you must demonstrate the potential to reach one. A component-level mass, volume, and power table against a stated M9 backpack envelope, plus the bill of materials DARPA wants at Month 24, converts an aspiration into an engineering argument. Do not forget thermal management, since the optional metabolic capability includes maintaining temperature.
Be honest about hemolysis and thrombosis at 15Fr. Single-access, 2 L/min or more, days of runtime, and a 5 mm cannula is a hard combination, and DARPA cited the cannula comparison and antithrombotic coating literature because it knows. Recirculation fraction, shear, coating strategy, anticoagulation approach in a setting with no laboratory, and the thrombotic and hemolysis risk checks named in the Month 24 deliverable all belong in the white paper.
Choose full versus partial lung support deliberately and defend it. Partial support at 125 mL/min is allowed but assumes an intubated patient, which reintroduces the ventilator, the sedation, and the airway management that the topic's opening argument was trying to eliminate. If you go partial, explain the tradeoff and show how you integrate with mechanical ventilation, which is the stated preference.
Add interim assessments. DARPA strongly encourages responders to propose additional interim assessments beyond the listed milestones. Doing so costs you nothing in page count if done well and signals program management maturity on a schedule that has real regulatory uncertainty in it.
Use the slide deck for the circuit. Fifteen slides is a generous visual allowance next to a 20 page white paper. A single-access circuit that carries blood flow, gas exchange, medication infusion, bolus fluid, and optional filtration is far easier to understand as a diagram than as prose. Spend slides on the circuit schematic, the control architecture, the sensor placement, the SWaP-C3 packaging concept, and the animal test configuration.
DARPA SBIR DPA26BZ06-DV028: TRIAGE-X, Autonomous Casualty Triage and Treatment in Chemically Contaminated Mass Casualty Events
Deadline: October 21, 2026
Funding Award Size: $2m
Description: Complete guide to DARPA SBIR Direct to Phase II topic DPA26BZ06-DV028, TRIAGE-X autonomous chemical casualty triage and treatment. $1.5M plus $500K option. Closes October 21, 2026.
Quick Answer
DPA26BZ06-DV028 is a DARPA SBIR Direct to Phase II topic under the DoW 2026 SBIR Broad Agency Announcement, Release 6. DARPA wants autonomous medical triage systems that recognize chemical toxidromes and physically deliver the corresponding antidote or life-saving intervention. Not recommend it. Deliver it. The award is $1,500,000 over 24 months with a $500,000 option over 6 months. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The operational logic is the reason this exists. When a mass casualty incident occurs inside a chemically contaminated area, responders cannot enter until the threat is assessed and characterized. To be safe they must don protective equipment first, which delays initial assessment, triage, and treatment. That equipment then imposes a physiological and functional penalty: slower movement, higher energy expenditure, heat stress, impaired respiration, and reduced manual dexterity, all of which further impede the ability to perform life-saving interventions. Meanwhile effective stabilization occurs in the cold zone, after casualties have been decontaminated, compounding the delay to evacuation and definitive care. An autonomous system that performs initial triage and treatment in the hot zone means responders either avoid entry entirely or gain time to don protective equipment.
The feasibility bar is high and specific. DARPA states that Phase I work is expected to have been completed before award, and lists concrete capabilities your existing prototype must already demonstrate, including autonomously finding and localizing casualties to within 2 meters, standoff vital signs assessment, and operating at scale across at least 20 casualties in a single scenario, all supported by a written technical description or published report with data and video.
Topic At a Glance
Topic number: DPA26BZ06-DV028
Title: TRIAGE-X: Autonomous Casualty Triage and Treatment in Chemically Contaminated Mass Casualty Events
Agency: Defense Advanced Research Projects Agency (DARPA)
Solicitation: DoW 2026 Small Business Innovation Research Broad Agency Announcement, Release 6, DARPA Proposal Submission Instructions
Program type: Direct to Phase II (DP2)
Technical volume format: White Paper and Slide Deck. The white paper shall not exceed 20 pages and the slide deck shall not exceed 15 slides
Base award: $1,500,000
Base period of performance: 24 months
Option: $500,000 over 6 months
Component Technology Priority Areas: Advanced Computing and Software; Biotechnology; Human-Machine Interfaces; Integrated Sensing and Cyber; Trusted AI and Autonomy
Projected CMMC level requirement: Level 2 (Self)
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic
Minimum deliverable: a demonstrated prototype that detects at least two representative chemical toxidromes and applies appropriate initial treatments following human-in-the-loop treatment approval
Out of scope: identification of the specific chemical agent, and chemical-specific sensors
Technical and Business Assistance: up to $25,000 per Phase II project, in addition to the cost ceiling
Topic Q&A: DSIP Topic Q&A is not available for DARPA topics. Technical questions go to SBIR_BAA@darpa.mil by October 14, 2026
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Submission portal: DSIP at dodsbirsttr.mil
Keywords: medical triage, mass casualty incidents, autonomous systems, robotics, medical countermeasures, chemical warfare injuries
The Feasibility Bar, Which Is the First Thing to Check
DARPA states that Phase I work is expected to have been completed before award. Proposers must submit evidence that their existing prototype can do six things.
Autonomously find and localize casualties and report their location to within 2 meters.
Assess trauma patterns and assign a trauma triage category.
Perform accurate standoff vital-signs assessment, including at minimum heart rate and respiratory rate.
Operate at scale, defined as at least 20 casualties in a single scenario, performing assessment in under 90 seconds for each casualty, and surveying a 30,000 square foot area without a battery change.
Operate across diverse, realistic environments: indoor and outdoor, daytime and nighttime, ideally but not required in adverse weather conditions such as rain and smoke.
Produce a human-interpretable output through a graphical interface that conveys casualty locations and priorities of care.
Plus a system requirement: the system must be portable, with all components easily transported by a vehicle or drone.
How the evidence must be presented
Capability demonstration: a written technical description or published report supported by data and videos demonstrating the prototype performing the capabilities above in the stated environments are required, provided via link.
Note "provided via link." Your videos are delivered as links rather than embedded files, which means hosting them somewhere durable and accessible to Government reviewers. Check the links work from outside your network and will stay live through the evaluation period, which runs up to 90 days past October 21.
What this bar actually screens for
Read the list together and the intended applicant becomes clear: a team that already fields an autonomous casualty search and triage system. The 2 meter localization, the sub-90-second per-casualty assessment, the 20-casualty scale, and the 30,000 square foot single-battery survey are all performance numbers from a working platform, not projections.
DARPA says so directly elsewhere in the topic. The DARPA Triage Challenge has demonstrated that autonomous air and ground systems, using standoff sensors, can locate casualties, characterize injury patterns, and support triage prioritization in realistic mass casualty incident conditions. Using DARPA Triage Challenge triage support capabilities as a starting point, this solicitation aims to extend capabilities to operating in chemically hazardous environments where injury patterns have distinct signatures or toxidromes.
If you competed in or built toward the DARPA Triage Challenge, this topic is written for you. If you have not, the honest question is whether you can document all six capabilities with data and video before October 21.
What DARPA Is Actually Looking For
The objective
Develop and demonstrate capability with autonomous medical triage systems to recognize chemical toxidromes and physically deliver the corresponding antidote or life-saving intervention.
The operational problem
DARPA seeks innovative approaches to autonomous chemical, biological, radioactive, and nuclear triage that measurably improve casualty survival and increase the effective capacity of a medical response during CBRN disasters.
Early situational awareness and timely treatment is a matter of life and death in battlefield and disaster medical triage. Responder vulnerability in hazardous environments caused by CBRN attacks adds significant difficulty to the medical response.
When a mass casualty incident occurs within a chemically contaminated area, the medical response is hindered by uncertainty while the threat is assessed and characterized. To ensure safety of the responders, they must don protective equipment before entry, delaying initial assessment, triage, and treatment with life-saving interventions.
Personal protective equipment also imposes a physiological and functional penalty on responders resulting in slower movement, higher energy expenditure, heat stress, impaired respiration, and reduced manual dexterity. These penalties further impede responders' ability to perform life-saving interventions.
Effective stabilization occurs in the cold zone after casualties have been decontaminated, compounding the delay to evacuation and definitive care.
Success here extends proven autonomous triage into one of the most challenging settings where human entry is dangerous and delayed, and where early antidote delivery most directly determines casualty survival.
What the system must do
Proposed systems should be able to locate and assess casualties for traumatic injuries and chemical toxidromes, generate triage recommendations, and deliver appropriate initial treatments to reverse, stabilize, or prevent the toxic effects of chemicals.
Solutions should be designed for realistic operational use in difficult terrain under challenging environmental conditions.
What DARPA deliberately does not prescribe
DARPA does not prescribe a specific platform, sensor suite, or level of autonomy. Offerors are encouraged to propose the human-machine teaming arrangement that best improves decision quality and casualty outcomes, and to define how their system fails safely when conditions exceed its competence.
That second clause is unusual and important. Defining how your system fails safely when conditions exceed its competence is a stated requirement. Autonomous systems administering medication in a contaminated environment need explicit competence boundaries and safe behavior at those boundaries. Address it directly rather than treating it as a risk paragraph.
Approaches should be realistic in scope, repurposed within the period of performance, and grounded in how CBRN mass casualty response actually unfolds.
At minimum, the effort shall deliver a demonstrated prototype that detects at least two representative chemical toxidromes and applies appropriate initial treatments for the assessed toxidrome following human-in-the-loop treatment approval.
Two toxidromes minimum. Treatment decisions always made by humans. Those two constraints bound the program.
Agent identification is explicitly out of scope
Performers shall develop the capability to assess and treat at least two chemical agent categories from the topic's toxidrome table. Agent detection is out of scope for this effort.
The requirements table repeats it: identification of the specific chemical agent is out of scope, and chemical-specific sensors are also out of scope.
This is a significant scoping decision and it should shape your architecture. You are not building a chemical detector. You are building a system that reads the casualty's physical presentation, the toxidrome, and infers the likely agent category from clinical signs. That is a medical inference problem from observable physiology and behavior, not a chemistry problem. Proposals that lean on chemical sensing have misread the topic.
The Toxidromes and Treatments
The topic provides three tables. Table 1 defines the target toxidromes, representative agents, physical presentation, and anticipated treatment to demonstrate. Table 2 defines how those toxidromes will be simulated on mannikins for demonstration. Table 3 defines the required technical capabilities.
Table 1, the five toxidrome categories
Cholinergic, or nerve-agent, syndrome. Representative agents: sarin, VX, Novichok, organophosphates. Physical presentation: DUMBBELS, meaning diarrhea, urination, miosis or constricted pupils, muscle weakness, bronchospasm, bronchorrhea, bradycardia or decreased heart rate, emesis, lacrimation, salivation and sweating. Anticipated treatment: intramuscular pharmaceutical delivery of simulated antidote, atropine and pralidoxime, plus airway management.
Blood-agent, or cyanide, syndrome. Representative agents: hydrogen cyanide, cyanogen chloride. Physical presentation: rapid collapse, gasping, tachypnea or fast respiratory rate progressing to apnea or no breathing, altered mental status, seizures, cardiovascular instability with low blood pressure and fast heart rate, skin flushing, central cyanosis, dilated pupils. Anticipated treatment: IV or IO pharmaceutical delivery, such as a cyanokit, plus airway management.
Pulmonary or choking-agent syndrome. Representative agents: chlorine, phosgene. Physical presentation: cough, dyspnea or difficulty breathing, wheezing, stridor, eye and airway irritation, respiratory distress, delayed pulmonary edema. Anticipated treatment: delivery of bronchodilators plus airway management.
Vesicant or blister-agent syndrome. Representative agent: sulfur mustard. Physical presentation: blistering skin burns, airway irritation, pain. Anticipated treatment: skin decontamination with removal of substance from exposed skin and clothing, and washing skin or using decontamination cloths, plus airway management.
Opioid or respiratory-depressant syndrome. Representative agent: fentanyl-type incapacitants. Physical presentation: pinpoint pupils, respiratory depression progressing to apnea, decreased consciousness. Anticipated treatment: intranasal or intramuscular naloxone plus airway management.
Two footnotes to Table 1 matter. Airway management includes providing supplemental oxygen by mask AND escalating to either non-invasive, meaning bag-valve mask or mechanical, or invasive, meaning endotracheal intubation or surgical airway, ventilation. And the cyanide treatment note specifies that IV or IO delivery includes autonomous placement of an intravenous or intraosseous catheter.
That last item is worth pausing on. Autonomous placement of an IV or IO catheter is a substantially harder robotic task than an intramuscular injection. If you select the blood-agent category as one of your two, you are committing to autonomous vascular or intraosseous access. Choosing cholinergic and opioid syndromes, both of which use intramuscular or intranasal delivery, is a materially easier engineering path. Choose deliberately and justify the choice.
Table 2, simulated presentations on mannikins
Performers will demonstrate their platform's ability to recognize simulated toxidromes in silico, meaning computer-based simulation, and on high-fidelity mannikins with the simulated toxidrome presentations outlined in Table 2.
The simulated cues are concrete and tell you exactly what your perception system must detect. For cholinergic syndrome: damp or wet clothing around the groin area and on the ground, constricted pupils, coughing sounds, rapid breathing at a rate greater than 30, slow palpable pulses at the wrist and groin at a rate under 40, emesis from the mouth with head turned and vomit on the side of the face and on the ground, and damp "skin" sprayed on with liquid.
For blood-agent syndrome: coughing sounds, rapid breathing greater than 30, altered mental status or confusion shown as mumbling words and not responding verbally to instructions, flushed red-appearing skin, dilated pupils.
For pulmonary syndrome: coughing sounds, rapid breathing greater than 30, red or irritated eyes with scleral injection, wheezing on auscultation of lung sounds, cyanosis around the lips.
For vesicant syndrome: gelatin on skin, skin blisters and burns, coughing sounds, rapid breathing greater than 20.
For opioid syndrome: constricted pupils, slow breathing at a rate under 10, unconscious and non-responsive.
Read that list as your sensing requirements specification. Pupillometry, respiratory rate, pulse palpation or standoff equivalent, auscultation, skin color and moisture assessment, blister detection, verbal responsiveness testing, and vomit detection. Several of these require close approach or contact, not just standoff sensing, which interacts with the mobility and manipulation design.
Table 3, required technical capabilities
Casualty Assessment. Locate casualty and report clinical signs including hemorrhage, respiratory distress, mental status, pain, emesis, injuries and injury location. Evaluate casualties for chemical toxidromes and assign a probable target agent category from Table 1. Assign triage category, urgent or non-urgent. Identification of the specific chemical agent is out of scope, and chemical-specific sensors are also out of scope.
Intervention Delivery. The platform must physically administer treatment appropriate to the chosen toxidrome according to Table 1 on a mannikin, and this is marked as required on a mannikin. Final demonstration requires the system payload and automation to administer treatment to at least 5 casualties in succession without battery changes or re-supply, also marked as required on a mannikin.
That five-casualty-in-succession requirement without battery change or resupply is a hard payload and endurance constraint. It means carrying five doses and the power to deliver them, and it interacts with the 30,000 square foot survey requirement from the feasibility bar.
Platform, Environment, and Autonomy. Platform-agnostic: autonomous UAV and UGV, and the performer may choose the sensor suite excluding chemical-specific sensors. System must demonstrate safe physical interaction with casualties represented by high-fidelity mannikins. A single robust platform must operate across and transition between realistic environments, meaning difficult terrain, inside and outside of buildings, and potentially subterranean settings, and this is marked as required over real terrain. Dedicated single-environment platforms are not sought.
That last sentence rules out a design optimized for one setting. One platform, multiple environments, including transitions between them.
End of Program Demonstration. Detection and assessment may be validated computationally through assessment of a test set of images and videos showing injuries and clinical signs such as respiratory distress, pupillary changes, salivation, sweating, skin changes and blisters. Intervention delivery on multiple casualties must be validated by physical demonstration, marked as required on a mannikin and over real terrain. The essential program output is a validated prototype.
Note the split: perception can be validated computationally against a test set, but intervention delivery must be physically demonstrated. That is a sensible and demanding division, and it means your budget needs real robotic manipulation development and mannikin testing, not just a vision pipeline.
Phase II
Building on the feasibility demonstrated in the documentation above, the Phase II effort will extend their qualified baseline system to assess and treat chemical toxidromes.
In this effort, performers will develop, integrate, and validate a prototype autonomous chemical casualty triage system and demonstrate it in a relevant environment.
The system shall identify casualties, assess them for life threatening and other traumatic injuries, assess casualties for chemical toxidromes, and provide relevant physical treatments.
Requirements: performers shall develop the capability to assess and treat at least two chemical agent categories from Table 1. Agent detection is out of scope for this effort. Treatment recommendations and options for each casualty shall be presented through a graphical interface to responders. Treatment decisions will always be made by humans.
The human-in-the-loop requirement is stated three separate times in the topic. Whatever autonomy you propose for search, assessment, and navigation, the treatment decision itself is a human decision presented through an interface. Design and describe that interface carefully, because it is the point where an autonomous system hands a life-or-death decision to a person who may be outside the hot zone.
Phase III Dual Use
Military: autonomous triage and toxidrome-directed treatment in chemically contaminated, battlefield environments protects medics from entry and prolongs casualty survival until human care is possible.
Civilian: similarly, autonomous systems capable of assessing and delivering life sustaining treatments during disaster responses, industrial chemical accidents, hazmat operations, and chemical terror attacks in which fire and EMS response will be delayed provides additional safety for responders and a survival advantage for casualties during public-health emergencies. The autonomous system will perform initial triage and treatment in the hot zone so that responders either avoid entry entirely or gain time to don protective equipment.
The civilian market here is concrete and reachable. Industrial chemical facilities, hazmat response teams, fire departments in petrochemical corridors, and emergency management agencies all face the same hot-zone entry delay. Opioid and fentanyl-related mass exposure incidents are a further civilian use case that maps directly onto one of the five toxidrome categories.
Funding, Cost Structure, and DARPA Mechanics
The award
$1,500,000 over a 24 month base, plus a $500,000 option over 6 months, for $2,000,000 across 30 months if the option is exercised.
The resources made available for each topic will depend on the quality of the proposals received and the availability of funds. The Government reserves the right to award all, some, one, or none of the options based on available funding and the performer's technical performance.
Contract type, which you must elect
DARPA may award FAR-based contracts, firm-fixed-price or cost-plus reimbursement, or Other Transactions for Prototype under the authority of 10 U.S.C. 4021, subject to approval of the Contracting Officer or Agreements Officer respectively. Proposers must state their requested contract type in their proposal.
Cost-plus reimbursement requires including your Defense Contract Management Agency Final Determination Letter showing approval of your accounting system. An Other Transaction for Prototype requires including a completed OT using the Model OT for Prototype from the DARPA Small Business site, plus completed OT Certifications, both loaded in Volume 5, with at minimum the color-coded areas completed and redlines with explanations for any article you wish to negotiate. Firm-fixed-price requires no additional action.
Templates are mandatory
Templates for Volume 2 Technical Volume and Volume 3 Cost Volume are provided as attachments on the DARPA Small Business website. Use of the DARPA Cost Proposal template is mandatory.
Technical and Business Assistance
Phase II awardees may request up to $25,000 per Phase II project. TABA funding is in addition to the cost ceilings and is not subject to profit or fee. Requests will be reviewed by the respective contracting office or specialist at time of award.
For this topic, regulatory strategy is worth considering, since a system that autonomously administers medication will eventually face a device and drug-delivery regulatory pathway, and intellectual property protections given the robotics and algorithm content.
Questions and the FAQ
DSIP Topic Q&A will not be available for these DARPA topics. Technical questions must be submitted by October 14, 2026, by email to SBIR_BAA@darpa.mil with the topic number in the subject line, including the name, email address, and telephone number of a point of contact. Questions submitted within seven calendar days of the proposal due date may not be answered. DARPA posts a consolidated Frequently Asked Questions document under the topic number summary on its Small Business site, updated on an ongoing basis until one week prior to the proposal due date.
DARPA will not accept late proposals.
Classification, marking, and registrations
All proposals are required to be UNCLASSIFIED or CUI. Do not include any classified information in your proposal submission. Do not include any proprietary information on the Proposal Coversheet in Volume 1.
Proposal titles, abstracts, anticipated benefits, and keywords of proposals selected for contract award will undergo a DARPA Policy and Security Review and are subject to revision or redaction by DARPA. Final approved versions may appear on the DoW SBIR/STTR awards website and the SBA's award website at sbir.gov/awards.
Proposers should ensure they have an accurate and active entity registration on SAM.gov. Those engaging in ITAR or CUI work for DARPA must have CMMC Level 2 certification, and the projected requirement for this topic is Level 2 with self-assessment. DARPA points to sprs.csd.disa.mil/nistsp.htm and notes Project Spectrum at projectspectrum.io as an assistance resource.
Venture capital, hedge fund, and private equity ownership
Proposers that are more than 50 percent owned by multiple venture capital operating companies, hedge funds, private equity firms, or any combination of these as set forth in 13 CFR 121.702 are eligible to submit proposals in response to DARPA topics advertised within this BAA. Three conditions apply: register with the SBA Company Registry Database before submitting; submit the Majority-Owned VCOC, HF, and PEF Certification, with the SBIR VC Certification available on the DARPA Small Business site, in Supporting Documents Volume 5; and immediately notify the Contracting Officer, register in the appropriate SBA database, and submit the required certification if you enter that ownership class after submitting but before receiving a funding agreement.
DARPA's permissive posture matters here, since autonomous robotics companies capable of clearing this feasibility bar are commonly venture-funded.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW SBIR Program BAA. DARPA will conduct an evaluation of each conforming proposal. Proposals that do not comply with the requirements detailed in this BAA and the research objectives of the corresponding topic are considered non-conforming and are therefore not evaluated nor considered for award.
Using the evaluation criteria, the Government will evaluate each proposal in its entirety, documenting the strengths and weaknesses relative to each evaluation criteria, and based on those will determine the proposal's overall selectability for funding. Proposals will not be evaluated against each other but on their own individual merit.
A selectable proposal is one where the strengths of the overall proposal outweigh its weaknesses, with no accumulated weaknesses that would require extensive negotiations or a resubmitted proposal. A non-selectable proposal is one where the strengths do not outweigh its weaknesses.
Proposing firms will be notified of selection or non-selection status within 90 calendar days of the closing date of the BAA. The Corporate Official indicated on the Proposal Cover Sheet will be notified by email. In accordance with the SBA SBIR/STTR Policy Directive, Appendix I, paragraph 4, subparagraph (d), DARPA will provide a technical evaluation narrative to the proposer for each proposal submitted in response to a topic. An informal feedback session may additionally be requested via email at sbir@darpa.mil, provided at the sole discretion of DARPA.
Company Commercialization Report information will not be considered by DARPA during proposal evaluations.
Protests regarding the selection decision should be submitted, as prescribed in FAR 33.106(b) and FAR 52.233-3, to DARPA Contracts Management Office, 675 N. Randolph Street, Arlington, VA 22203, by email to CMO_SBIRProtests@darpa.mil and sbir@darpa.mil.
Post-award support
DARPA provides Transition and Commercialization Support Program services to Phase II and DP2 awardees upon contract execution at no cost to awardees. Awardees may also be eligible for the Embedded Entrepreneurship Initiative, an invitation-only program at DARPA's sole discretion, typically no more than $310,000 per awardee over the duration of the award, supporting a Senior Commercialization Advisor relationship, investor working group connections, and hiring an embedded entrepreneur to execute a Go-to-Market strategy.
The References
Six, and the mix tells you what literacy DARPA expects: chemical casualty medicine plus its own prior program.
Dembek ZF, editor. Medical aspects of biological warfare. Government Printing Office; 2008.
Ramesh AC, Kumar S. Triage, monitoring, and treatment of mass casualty events involving chemical, biological, radiological, or nuclear agents. Pharmacy and Bioallied Sciences.
The DARPA Triage Challenge program summary, at darpa.mil, attachment dtc-challenge-program-summary.pdf dated 2026-05.
Keating B, Eide KL, Vaag JR, Lund-Kordahl I. Tactical Triage: Adapting Care and Decision-Making for High-Threat Environments.
U.S. Department of the Air Force (2021). Air force tactics, techniques, and procedures 3-42.32: Installation medical all hazards response, IMAHR.
Marrs TC. Toxicology of organophosphate nerve agents. Chemical warfare agents: toxicology and treatment. 2007.
The DARPA Triage Challenge program summary is the most operationally useful of these. It describes the baseline capability DARPA says this topic extends, and it will tell you what the agency already considers demonstrated. Read it before writing, and position your system relative to it explicitly.
The Marrs organophosphate toxicology chapter and the Dembek volume are the clinical grounding for toxidrome recognition and treatment. The AFTTP 3-42.32 Installation Medical All Hazards Response publication is the doctrine for how the Air Force actually organizes a chemical mass casualty response, which is where your system has to fit.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
Technical question deadline: October 14, 2026, to SBIR_BAA@darpa.mil with the topic number in the subject line
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Selection notification: within 90 calendar days of BAA close
Base period: 24 months from award
Option: 6 additional months if exercised
A working backward plan
Before September 23. Assess yourself against the six feasibility capabilities honestly, since Phase I work is expected to have been completed before award. Assemble the written technical description or published report with supporting data and videos, and host the videos at durable links accessible from outside your network. Download the mandatory DARPA Volume 2 and Volume 3 templates. Read the FAQ and keep rechecking it. Read the DARPA Triage Challenge program summary and position relative to it. Choose your two toxidrome categories, weighing the intramuscular and intranasal routes against the IV and IO route which requires autonomous catheter placement. Secure access to high-fidelity mannikins and a test environment covering difficult terrain, indoor and outdoor, and ideally subterranean transitions. Decide your contract type and prepare the corresponding documents. Confirm SAM registration and your CMMC Level 2 self-assessment in SPRS. If venture-backed, register with the SBA Company Registry and obtain the SBIR VC Certification.
September 23 through October 5. Draft the 20 page white paper and 15 slide deck. Structure around the toxidrome selection and the clinical inference from Table 2 cues, the intervention delivery mechanism, the platform and environment transitions, the human-in-the-loop treatment interface, and the fail-safe behavior when conditions exceed system competence. Use the slide deck for platform imagery, sensing geometry, and demonstration evidence, which is what slides do better than prose.
October 6 through October 14. Build the cost volume in the mandatory template across 24 months plus the 6 month option. Price mannikin acquisition or access, drug-delivery payload development, test range or facility access covering multiple environments, and the five-casualty-in-succession endurance testing.
October 15 through October 18. Assemble Volume 5 with contract-type documents and certifications, complete Volume 7 and the Volume 4 CCR, and run compliance: 20 page white paper, 15 slide deck, unclassified or CUI only, no proprietary information on the coversheet, mandatory cost template, working video links.
October 19 through October 20. Submit and certify in DSIP.
Frequently Asked Questions
What is DARPA SBIR topic DPA26BZ06-DV028?
DPA26BZ06-DV028 is a DARPA SBIR Direct to Phase II topic titled "TRIAGE-X: Autonomous Casualty Triage and Treatment in Chemically Contaminated Mass Casualty Events," released under the DoW 2026 SBIR Broad Agency Announcement, Release 6. The objective is to develop and demonstrate capability with autonomous medical triage systems to recognize chemical toxidromes and physically deliver the corresponding antidote or life-saving intervention.
How much funding is available?
$1,500,000 for a 24 month base period, plus a $500,000 option over 6 months, for $2,000,000 across 30 months if the option is exercised. Up to $25,000 in Technical and Business Assistance may be requested in addition to the cost ceiling.
When is the proposal deadline?
October 21, 2026. DARPA will not accept late proposals. Note the separate technical question deadline of October 14, 2026.
What must my existing prototype already do?
Six things. Autonomously find and localize casualties and report their location to within 2 meters. Assess trauma patterns and assign a trauma triage category. Perform accurate standoff vital-signs assessment including at minimum heart rate and respiratory rate. Operate at scale, meaning at least 20 casualties in a single scenario, assessment in under 90 seconds per casualty, and surveying a 30,000 square foot area without a battery change. Operate across diverse realistic environments, indoor and outdoor, daytime and nighttime, ideally but not required in rain and smoke. And produce a human-interpretable output through a graphical interface conveying casualty locations and priorities of care. The system must also be portable, with all components easily transported by a vehicle or drone.
How do I document that?
A written technical description or published report supported by data and videos demonstrating the prototype performing those capabilities in the stated environments, provided via link. Host the videos somewhere durable and accessible to Government reviewers from outside your network, and make sure the links stay live through the evaluation period, which can run 90 days past the close date.
Is this topic written for DARPA Triage Challenge participants?
Effectively yes. The topic states that the DARPA Triage Challenge has demonstrated that autonomous air and ground systems using standoff sensors can locate casualties, characterize injury patterns, and support triage prioritization in realistic mass casualty incident conditions, and that this solicitation uses DTC triage support capabilities as a starting point to extend into chemically hazardous environments. The DTC program summary is one of the six cited references.
Do I need to detect the chemical agent?
No, and this is a significant scoping decision. Agent detection is out of scope for this effort. The requirements table states that identification of the specific chemical agent is out of scope and that chemical-specific sensors are also out of scope. You are inferring a probable agent category from the casualty's physical presentation, the toxidrome, not measuring the chemical.
How many toxidromes must I address?
At least two chemical agent categories from Table 1. The minimum deliverable is a demonstrated prototype that detects at least two representative chemical toxidromes and applies appropriate initial treatments for the assessed toxidrome following human-in-the-loop treatment approval.
Which two toxidromes should I choose?
The choice has large engineering consequences. Cholinergic syndrome uses intramuscular delivery of atropine and pralidoxime. Opioid syndrome uses intranasal or intramuscular naloxone. Pulmonary syndrome uses bronchodilator delivery. Vesicant syndrome uses skin decontamination. But blood-agent, cyanide, syndrome requires IV or IO pharmaceutical delivery, and the table footnote specifies that this includes autonomous placement of an intravenous or intraosseous catheter, which is a substantially harder robotic task. Choosing routes that avoid autonomous vascular access is a materially easier path, and you should justify whichever pair you select.
What does airway management require?
Per the Table 1 footnote, airway management includes providing supplemental oxygen by mask AND escalating to either non-invasive, meaning bag-valve mask or mechanical, or invasive, meaning endotracheal intubation or surgical airway, ventilation. Airway management appears as an anticipated treatment for all five toxidrome categories.
How will toxidrome recognition be demonstrated?
Performers will demonstrate their platform's ability to recognize simulated toxidromes in silico, meaning computer-based simulation, and on high-fidelity mannikins with the simulated presentations outlined in Table 2. Those simulated cues include damp clothing and ground, constricted or dilated pupils, coughing sounds, specific respiratory rates, slow palpable pulses at wrist and groin, emesis positioned on the face and ground, sprayed liquid simulating damp skin, gelatin simulating blisters, scleral injection, wheezing on auscultation, cyanosis around the lips, and verbal non-responsiveness.
What sensing does that imply?
Read Table 2 as your sensing specification. It requires pupillometry, respiratory rate measurement, pulse assessment at the wrist and groin, auscultation of lung sounds, skin color and moisture assessment, blister detection, verbal responsiveness testing, and emesis detection. Several of these require close approach or contact rather than pure standoff sensing, which interacts with your mobility and manipulation design.
What is the intervention delivery requirement?
The platform must physically administer treatment appropriate to the chosen toxidrome according to Table 1 on a mannikin. The final demonstration requires the system payload and automation to administer treatment to at least 5 casualties in succession without battery changes or re-supply. Both are marked as required on a mannikin.
What platform should I propose?
DARPA is platform-agnostic: autonomous UAV and UGV are both acceptable, and the performer may choose the sensor suite excluding chemical-specific sensors. However, a single robust platform must operate across and transition between realistic environments, meaning difficult terrain, inside and outside of buildings, and potentially subterranean settings, and dedicated single-environment platforms are not sought.
Who makes the treatment decision?
A human, always. Treatment recommendations and options for each casualty shall be presented through a graphical interface to responders, and treatment decisions will always be made by humans. The topic states this three separate times. Whatever autonomy you propose for search, assessment, and navigation, the treatment decision is presented to a person.
What does DARPA say about failure modes?
Offerors are encouraged to propose the human-machine teaming arrangement that best improves decision quality and casualty outcomes, and to define how their system fails safely when conditions exceed its competence. That second element is a stated expectation, not a risk-section afterthought, and for an autonomous system administering medication it is central.
How is the end of program demonstration structured?
Detection and assessment may be validated computationally through assessment of a test set of images and videos showing injuries and clinical signs such as respiratory distress, pupillary changes, salivation, sweating, skin changes and blisters. Intervention delivery on multiple casualties must be validated by physical demonstration, required on a mannikin and over real terrain. The essential program output is a validated prototype.
How long can my technical volume be?
This topic uses the White Paper and Slide Deck format. The white paper shall not exceed 20 pages and the slide deck shall not exceed 15 slides. Refer to Appendix B, DARPA Direct to Phase II Instructions, for the content of each element and the commercialization strategy.
Can I ask questions through DSIP Topic Q&A?
No. DARPA states DSIP Topic Q&A will not be available for these topics. Technical questions go to SBIR_BAA@darpa.mil with the topic number in the subject line by October 14, 2026. Questions submitted within seven calendar days of the due date may not be answered. DARPA posts a consolidated FAQ, updated until one week before the due date.
Do I have to choose a contract type?
Yes. Proposers must state their requested contract type. DARPA may award FAR-based firm-fixed-price or cost-plus reimbursement contracts, or Other Transactions for Prototype under 10 U.S.C. 4021. Cost-plus requires your DCMA Final Determination Letter showing accounting system approval. An OT requires a completed Model OT plus OT Certifications in Volume 5. Firm-fixed-price requires no additional action.
Is the cost template mandatory?
Yes. Templates for Volume 2 and Volume 3 are on the DARPA Small Business website, and use of the DARPA Cost Proposal template is mandatory.
Are venture capital backed companies eligible?
Yes. Proposers more than 50 percent owned by multiple venture capital operating companies, hedge funds, private equity firms, or any combination as set forth in 13 CFR 121.702 are eligible, subject to registering with the SBA Company Registry Database before submitting, submitting the Majority-Owned VCOC, HF, and PEF Certification in Volume 5, and notifying the Contracting Officer if you enter that class after submitting but before award.
How will my proposal be evaluated?
Against the evaluation criteria in the DoW SBIR Program BAA. DARPA evaluates each conforming proposal in its entirety, documenting strengths and weaknesses relative to each criterion, then determines overall selectability. Proposals are not evaluated against each other. A selectable proposal is one where strengths outweigh weaknesses with no accumulated weaknesses requiring extensive negotiations or resubmission.
Will I get feedback if not selected?
Yes. DARPA will provide a technical evaluation narrative to the proposer for each proposal submitted in response to a topic, per the SBA SBIR/STTR Policy Directive. An informal feedback session may additionally be requested via sbir@darpa.mil, at DARPA's sole discretion.
What is the commercial market?
Autonomous systems capable of assessing and delivering life sustaining treatments during disaster responses, industrial chemical accidents, hazmat operations, and chemical terror attacks in which fire and EMS response will be delayed. Industrial chemical facilities, hazmat teams, fire departments in petrochemical corridors, and emergency management agencies all face the same hot-zone entry delay. Opioid and fentanyl mass exposure incidents map directly onto one of the five toxidrome categories and are a further civilian use case.
Who do I contact with questions?
The DARPA Small Business Programs Office at SBIR_BAA@darpa.mil for both program administration and topic technical questions, with the topic number in the subject line. DSIP technical support at DoDSBIRSupport@reisystems.com with a copy to SBIR_BAA@darpa.mil, Monday through Friday 9:00 a.m. to 5:00 p.m. ET. DARPA also offers free resources through DARPAConnect at DARPAConnect.us.
Positioning Advice for Companies Considering This Topic
Check the six feasibility capabilities before anything else. Phase I work is expected to have been completed before award, and the numbers are specific: 2 meter localization, sub-90-second assessment, 20 casualties, 30,000 square feet on one battery. These are measurements from a working system. If you cannot produce data and video for all six, the effort is better spent elsewhere.
Choose your two toxidromes on engineering grounds, then justify clinically. Autonomous IV or IO catheter placement, required for the cyanide category, is a different order of robotic difficulty from an intramuscular injection or intranasal spray. Cholinergic plus opioid gives you two clinically important categories with tractable delivery routes. If you choose the harder route, explain why your manipulation capability supports it.
Build the perception case from Table 2, not from generalities. DARPA told you exactly what cues the mannikins will present: pupil size, respiratory rate thresholds, palpable pulse rates, wheezing on auscultation, scleral injection, lip cyanosis, gelatin blisters, sprayed liquid, positioned emesis, verbal non-responsiveness. A proposal that walks those cues and states which sensor and algorithm addresses each is far more persuasive than one describing a general-purpose vision stack.
Solve the close-approach problem. Several Table 2 cues, notably auscultation and palpable pulse, are not standoff measurements. Your system has to get close enough to contact a casualty in a contaminated environment and do so safely, since safe physical interaction with high-fidelity mannikins is a stated requirement. That is a manipulation and safety problem distinct from search and navigation.
Take the human-in-the-loop interface seriously as a deliverable. It is mentioned three times, and it is the point where an autonomous system in a hot zone hands a life-or-death decision to a person who may be outside it. Show the interface: what the responder sees, how confidence is expressed, how they approve or decline, and what latency the loop introduces.
Define competence boundaries and fail-safe behavior explicitly. DARPA asked for it. An autonomous system that misreads a toxidrome and delivers the wrong drug is worse than one that declines and flags for human assessment. Saying where your system stops, and what it does when it stops, is a strength rather than an admission.
Design one platform for all environments. Difficult terrain, inside and outside buildings, potentially subterranean, with transitions between them, and dedicated single-environment platforms are explicitly not sought. If your strength is one domain, address the transition problem head on rather than hoping it goes unnoticed.
Budget the five-in-succession endurance requirement. Five treatments delivered without battery change or resupply, on top of a 30,000 square foot survey, is a real payload, power, and consumables engineering problem. Show the budget.
Read the DARPA Triage Challenge program summary and position against it. It is cited as the baseline this topic extends. Stating clearly what your system already does that DTC demonstrated, and what the chemical extension adds, frames your proposal in the agency's own terms.
Ground the medicine properly. Use DUMBBELS correctly, get the atropine and pralidoxime pairing right, understand why naloxone is intranasal or intramuscular, and know why vesicant treatment is decontamination rather than an antidote. The cited Marrs organophosphate chapter and Dembek volume are there for a reason, and clinical fluency separates a robotics company that understands the mission from one that has read a summary.
Use the slide deck for what slides do well. Fifteen slides alongside a 20 page white paper is a generous visual allowance. Platform photographs, sensing geometry diagrams, demonstration stills, and toxidrome cue mappings all communicate faster as images. Do not duplicate the white paper in bullet form.
DARPA SBIR DPA26BZ06-DV027: Casualty Operations and Resource Prediction Software (CORPS)
Deadline: October 21, 2026
Funding Award Size: $750k
Description: Complete guide to DARPA SBIR Direct to Phase II topic DPA26BZ06-DV027, CORPS casualty operations and resource prediction software. $750,000 award. Closes October 21, 2026.
Quick Answer
DPA26BZ06-DV027 is a DARPA SBIR Direct to Phase II topic under the DoW 2026 SBIR Broad Agency Announcement, Release 6. DARPA wants software that predicts casualty and patient streams following large scale medical emergencies in austere environments, and then projects the lifesaving potential of different response options, with limited or no access to real-time data or communications. The award per the topic index is $750,000 over 12 months with no options. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The problem DARPA describes is a decision-making problem under bad information, not a modeling problem in the abstract. Strategic and operational medical planning happens with time and full situational awareness. Crisis action decisions do not: they must be made in much shorter time frames, using sparse, low-quality situational data that cannot be verified because communications are denied, disrupted, intermittent, and limited in bandwidth. And medical response competes with tactical and logistical demands, so the medical benefit of a response option has to be weighed against what that option costs the mission. Without quantitative metrics for the medical benefit of each option, those lifesaving decisions get made subjectively.
Important scheduling note before you plan anything. The topic index states $750,000 over a 12 month period of performance with no Option 1 or Option 2. The Phase II narrative and milestone tables in the same document describe an 18 month base period plus a 6 month option period, with milestones running through Month 24. Those two statements cannot both be right, and this is the first question to send to DARPA.
Topic At a Glance
Topic number: DPA26BZ06-DV027
Title: Casualty Operations and Resource Prediction Software (CORPS)
Agency: Defense Advanced Research Projects Agency (DARPA)
Solicitation: DoW 2026 Small Business Innovation Research Broad Agency Announcement, Release 6, DARPA Proposal Submission Instructions
Program type: Direct to Phase II (DP2). This topic is soliciting DP2 proposals only
Technical volume format: White Paper and Slide Deck. The white paper shall not exceed 20 pages and the slide deck shall not exceed 15 slides
Award per the topic index: $750,000
Period of performance per the topic index: 12 months, with no Option 1 and no Option 2
Award structure per the topic narrative: an 18 month base period plus a 6 month option period, with milestones through Month 24. This conflicts with the index and needs clarification
Component Technology Priority Areas: Advanced Computing and Software; Integrated Network Systems-of-Systems; Trusted AI and Autonomy
Projected CMMC level requirement: Level 2 (Self)
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic
Minimum technology maturity: offerors should only propose products at or above Technology Readiness Level 2
Prototype delivery requirement: no less than 4 prototypes supporting different military user evaluations
Technical and Business Assistance: up to $25,000 per Phase II project, in addition to the cost ceiling
Topic Q&A: DSIP Topic Q&A is not available for DARPA topics. Technical questions go to SBIR_BAA@darpa.mil by October 14, 2026
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Submission portal: DSIP at dodsbirsttr.mil
Keywords: medical response, austere environments, casualty stream predictions, contested logistics, decision aid, automation
The Schedule Discrepancy You Must Resolve
This is the single most important administrative item on this topic, and it affects your cost volume, your milestone plan, and your staffing.
The Current Release Award Structure by Topic table for the White Paper and Slide Deck format lists DPA26BZ06-DV027 with an Award Amount of $750,000, a Period of Performance of 12 months, an Option Amount of N/A, and an Option Period of Performance of N/A.
The Phase II narrative for the same topic states that the 18-month base period should focus on developing, integrating, and demonstrating the core capabilities of the prototype through measurable milestones and evaluations, and that the 6-month option period should focus on refining the prototype based on user feedback, validating performance in operationally relevant scenarios, and enabling transition. The Phase II Milestones table then runs M1 through M6 across Months 1 to 18, and a Phase II Option Milestones table runs O1 through O3 across Months 19 to 24. A closing paragraph restates the 18 month base and 6 month option split, describing the base as "Develop, Integrate, and Demonstrate" and the option as "Refine, Validate, and Enable Transition."
Twelve months with no options and eighteen months plus a six month option are not reconcilable. Possible readings include that the index is correct and the milestone tables were carried over from a longer version of the topic, or that the index row is in error and the narrative governs. We cannot tell from the document.
What to do. Submit this question to SBIR_BAA@darpa.mil with the topic number in the subject line, well before the October 14 deadline, and watch the consolidated FAQ. In the meantime, note that DARPA's instructions state that proposals which do not comply with the requirements detailed in the BAA and the research objectives of the corresponding topic are considered non-conforming and are therefore not evaluated. That makes proposing the wrong period of performance a real risk rather than a cosmetic one. If you cannot get an answer, the safer construction is to propose to the index figures, since the index is the controlling award structure table, while showing in the technical volume how your milestone plan maps onto the six-milestone base structure DARPA described, compressed to the funded period.
What DARPA Is Actually Looking For
The objective
A software tool to predict casualty and patient streams and needs following large scale medical emergencies in austere environments, and project the lifesaving potential of various response options with limited or no access to real-time data or communications.
Why crisis medical planning is different
The challenges underlying crisis medical response in austere and contested environments are significantly different than those underlying long-range planning or response in established theaters.
Strategic and operational planning decisions are made in ample time with full situational awareness regarding the scope and nature of medical needs, availability and positioning of resources, and the immediate and emerging state of the theater.
Crisis action decisions must be made in much shorter time frames, using sparse, low-quality situational data that cannot be verified due to lack of secure communications and denied, disrupted, intermittent, and limited-bandwidth, or DDIL, states.
Furthermore, as resource allocation for medical response must be constrained by tactical and logistical requirements, the medical benefits of early and best medical response must be weighed against the tactical demands of the mission objectives.
Thus, in scenarios where theater realities drive the timing, level, and means of medical response, decision makers must rapidly devise multiple optional courses of action and optimize their selection to accommodate tactical and logistical limitations. Without quantitative metrics assigned to the medical benefits of each response option, these lifesaving decisions must be made using multiple channels of low-quality situational data to obtain subjective medical benefits.
That final sentence is the gap. Today the medical benefit of a response option is a judgment call made from bad data. DARPA wants it to be a number.
What the product must do
This topic seeks proposals for solutions that assist users with medical response decisions in austere environments with variable access to communications or situational data.
The desired product will be focused on automated prediction of casualty and patient streams, identification of the related medical needs with variable latencies, and metrics-based evaluation of benefits gained from various responses as a dynamic function of time.
Three functions, and the third one is the differentiator. Predicting casualties is one thing. Predicting the medical needs those casualties generate, with variable latencies, is harder. Quantifying the benefit of each response option as a function of time is the part that turns a model into a decision aid.
Solutions must have the potential for use in the civilian sector, meaning integrate with civilian networks and infrastructure, as well as integration into existing DoD planning tools.
Proposals shall outline the strategic and technical development of the proposed solution, including required data, data sources and environments, data quality management, and model development design and integration.
As the technology will be fielded for military use in various operational environments, proposed approaches must be adaptable to variable levels of communication capability and medical care.
That last requirement is the DDIL constraint restated as a design principle. Your tool has to degrade gracefully. Full connectivity, intermittent connectivity, and no connectivity should all produce usable output at different confidence levels, and saying how is central to a responsive proposal.
The Feasibility Requirement
This topic is soliciting Direct to Phase II proposals only. As such, the offeror shall provide detail and documentation which demonstrates the accomplishment of a "Phase I-like" effort demonstrating feasibility and proof-of-concept.
DARPA then lists seven things a demonstration of feasibility may include. The permissive "may include" language gives you latitude in how you evidence it, but the list tells you what DARPA expects to see.
Conceptual characterization of the complete product.
Relevant use cases, required data categories, and anticipated final capabilities.
Leveraged systems and workflow, as relevant.
Proposed approach or methodology for model development, as relevant.
Anticipated interactions between data and model components, as relevant.
The presentation of scientific and technical material that support the above.
Design specifications for any computational or software components of a prototype.
Then a separate requirement that is not optional: proposals should contain preliminary data, published or unpublished, supporting the rationale for the development of the candidate products. Describe how the product will be usable in the operational continuum or the environmental settings for which it is designed.
Note that this feasibility bar is notably softer than several other topics in this DARPA release. It asks for conceptual characterization, methodology, design specifications, and preliminary data, rather than a validated working prototype. That opens the topic to modeling and analytics teams with strong underlying science and a credible design, rather than restricting it to companies with fielded software. If you have published casualty modeling work, medical logistics optimization research, or operations research applied to mass casualty response, read this requirement carefully before assuming you are ineligible.
For the White Paper and Slide Deck format, the white paper shall not exceed 20 pages and the slide deck shall not exceed 15 slides. Refer to Appendix B, DARPA Direct to Phase II Instructions, for the content of each element of the Technical Volume and the commercialization strategy.
The Phase II Program
This phase will focus on refinement and optimization of a prototype tool that can interact with users through a visual interface. The prototype should be designed to integrate decision criteria from users, for example response level and latency, with medical needs predictions from built-in models. Software interface must be sufficiently mature to receive input data from users and display the output for evaluation and further improvement of the tool's feedback.
Offerors should only propose products at or above Technology Readiness Level 2.
The proposal shall describe the planned prototype design, product development, testing, and validation of the prototype product in table-top like exercises. The testing and practical implementation of the prototype should be relevant to the requirements of medical command and control functions in austere and contested environments.
What the proposed work may include
Information required for a Phase I proposal, including selected methods and approaches.
Medical, scientific, and technical justification and selection criteria for specific approaches.
Anticipated data sources, for example theater, casualty, medical, natural, virtual, and types, organic versus synthetic.
Interactions and relationships between the data and modeling components of the proposed work.
Detailed description of planned prototype design and development, covering communication requirements, data, architecture, software, interface, and models.
Methodology and outcome metrics for determination of functionality and utility.
Threshold and objective exit criteria.
The four prototype requirement
The investigator shall deliver no less than 4 prototypes supporting different military user evaluations.
Each prototype should demonstrate increasing operational capability and maturity while reducing technical risk through measurable performance objectives.
That is an unusual and demanding structure. Four successive prototypes, each evaluated by different military users, each more capable than the last. It maps directly onto the milestone table below, where Prototypes 1 through 4 are named milestones at Months 8, 11, 15, and 18. Your plan must be organized around iterative delivery and user evaluation, not a single build.
The proposal shall describe a detailed strategy for the Phase III effort to include dynamic uncertainty quantification to support operational decision-making and transition to both DoD and commercial applications.
Dynamic uncertainty quantification appears in both the Phase III strategy requirement and in the Prototype 2 milestone. Given that the entire premise is decision-making on sparse, unverifiable data, quantified uncertainty is not a refinement, it is the feature that makes the output trustworthy. Treat it as core.
The milestone structure DARPA specifies
Note that these milestones assume the 18 month base and 6 month option structure described in the narrative, which conflicts with the 12 month no-option structure in the index. Resolve that before building your plan, but understand the intended shape.
M1, Operational Requirements and Technical Baseline, Months 1 to 2. Technical objective: establish the operational concept, technical architecture, and data strategy for casualty prediction and medical decision support in DDIL environments. Exit criteria and deliverables: System Requirements Specification, System Architecture Description, Data Management Plan, with an End of Month 2 System Requirement Review.
M2, Casualty Prediction and Data Fusion, Months 3 to 5. Technical objective: develop and validate the baseline casualty prediction engine using representative military and civilian datasets with varying data completeness. Exit criteria: baseline casualty prediction engine demonstrated, data fusion capability validated using representative operational scenarios, initial model performance assessment completed, with an End of Month 5 Technical Baseline Review.
Note the phrase "varying data completeness." Your validation has to show the engine works when the data is partial, which is the operational reality the topic describes.
M3, Prototype Number 1, Technical Baseline and Decision Optimization Prototype, Months 6 to 8. Technical objective: establish the program's technical baseline while demonstrating an initial prototype that integrates casualty prediction with medical resource allocation and response optimization to evaluate alternative Courses of Action. Exit criteria: an Operational Model Baseline covering physiology model requirements, performance objectives, and fidelity needed to support operational decision making; operational parameter space, key decision variables, and scenario boundaries; data requirements, model assumptions, constraints, and uncertainty characterization; threshold and objective performance metrics, verification methodology, and validation approach. Plus automated Course of Action generation and ranking demonstrated. There is a Month 6 Government Technical Baseline Review completed, establishing technical foundation and evaluation criteria for prototype maturation, and an End of Month 8 Prototype 1 Presentation and Demonstration.
M4, Prototype Number 2, Decision Support and Interaction, Months 9 to 11. Technical objective: develop an interactive decision-support interface incorporating user-defined constraints and dynamic uncertainty quantification. Exit criteria: interactive visualization demonstrated, dynamic uncertainty and confidence metric demonstrated, user assessment completed and incorporated into prototype refinement, with an End of Month 11 Prototype 2 Presentation and Demonstration.
M5, Prototype Number 3, Integrated Operational Prototype, Months 12 to 15. Technical objective: integrate prediction, optimization, visualization, and user interaction into a single operational prototype and demonstrate functionality in representative military scenarios. Exit criteria: successful tabletop exercise completed, prototype evaluated against metrics, with an End of Month 15 Prototype 3 Presentation and Demonstration.
M6, Prototype Number 4, Operational Prototype Demonstration and Transition Planning, Months 16 to 18. Technical objective: demonstrate a mature prototype and establish the technical foundation for Phase III transition. Exit criteria: operational prototype demonstration completed, final technical report completed, Phase III transition strategy completed, with an End of Month 18 Prototype 4 Presentation and Demonstration.
Option milestones:
O1, Operational Refinement and Model Optimization, Months 19 to 20. Refine casualty prediction, decision optimization, and user interface capabilities based on Government feedback from Phase II evaluations. Exit criteria: prototype refinement demonstration, model improvement documentation, with an End of Month 20 updated prototype demonstrating improved performance.
O2, Expanded Operational Validation, Months 21 to 22. Validate prototype performance across representative military and civilian emergency response scenarios. Exit criteria: prototype demonstration with expanded scenarios with metric and threshold validations, user evaluation report documentation, with an End of Month 22 updated prototype demonstration.
O3, Transition Readiness, Months 23 to 24. Prepare the prototype for Phase III maturation by documenting technical maturity, transition planning, and integration considerations. Exit criteria: updated transition and commercialization plan identifying DoD and commercialization opportunities, preliminary integration and cyber security strategy documented, final prototype package delivered, with an End of Month 24 final program review.
Phase III
If successful, Phase II work will result in a final phase funded by sources other than a Federal government SBIR Program. Phase III awards may be made by any Government entity without further competition, creating a "SBIR-sourcing" or sole-source-like tool for portfolio managers and advanced developers.
Final improvements to the product's functionality should be completed during Phase III, and additional development will be performed as necessary to expand the methodologies and components included in the Phase II prototype or improve their capability.
The resulting product will provide a decision-assist tool for medical response optimization in austere environments where access to data and communications ranges from full to null.
Phase III is expected to mature the Phase II prototype into an operational decision-support capability suitable for military and commercial use. Activities may include expanding predictive models, incorporating additional operational data sources, enhancing interoperability with existing planning and command and control systems, strengthening cyber security, and refining software functionality based on operational feedback.
The desired end-state of the research is the full development of one or more products consisting of front- and back-end software for a digital tool that provides casualty stream predictions for a given event with minimal user input, interacts with users to receive known response parameters, and projects the health and lifesaving impacts of available response options on survivability. The product must demonstrate full functionality and reliability in tabletop and field exercises.
Military application: the potential product may transition to an Acquisition Program managed by the Service Product Developers for inclusion into various Service-specific mobile applications. Deployment will require appropriate cybersecurity certifications be met and the ability to join an existing Authority to Operate or obtain a separate ATO.
Commercial application: the product will provide a similar capability for the planning of response to non-military emergency events, for example following natural disasters or war.
Two practical notes. The ATO requirement is real work with its own timeline, and the O3 option milestone asks for a preliminary integration and cyber security strategy, which is where you begin addressing it. And the transition target is Service-specific mobile applications, which tells you the eventual delivery form factor and argues for designing the interface with mobile constraints in mind from the start.
Funding, Cost Structure, and DARPA Mechanics
The award
Per the topic index, $750,000 over 12 months with no options. Per the topic narrative, an 18 month base plus a 6 month option. See the discrepancy section above and resolve it with DARPA before you build your cost volume.
The resources made available for each topic will depend on the quality of the proposals received and the availability of funds. The Government reserves the right to select for negotiation all, some, one, or none of the proposals received and to make awards with or without communications with proposers, and to award all, some, one, or none of the options based on available funding and the performer's technical performance.
Contract type, which you must elect
DARPA may award FAR-based contracts, firm-fixed-price or cost-plus reimbursement, or Other Transactions for Prototype under the authority of 10 U.S.C. 4021, subject to approval of the Contracting Officer or Agreements Officer respectively. Proposers must state their requested contract type in their proposal.
Cost-plus reimbursement requires including your Defense Contract Management Agency Final Determination Letter showing approval of your accounting system. An Other Transaction for Prototype requires including a completed OT using the Model OT for Prototype from the DARPA Small Business site, plus completed OT Certifications, both loaded in Volume 5, with at minimum the color-coded areas completed and redlines with explanations for any article you wish to negotiate. Firm-fixed-price requires no additional action.
Templates are mandatory
Templates for Volume 2 Technical Volume and Volume 3 Cost Volume are provided as attachments on the DARPA Small Business website. Use of the DARPA Cost Proposal template is mandatory.
Technical and Business Assistance
Phase II awardees may request up to $25,000 per Phase II project. TABA funding is in addition to the cost ceilings and is not subject to profit or fee. Requests will be reviewed by the respective contracting office or specialist at time of award.
For this topic, cybersecurity assistance is directly relevant given the ATO requirement flagged in Phase III, and market validation with civilian emergency management buyers supports the dual-use requirement.
Questions and the FAQ
DSIP Topic Q&A will not be available for these DARPA topics. Technical questions must be submitted by October 14, 2026, by email to SBIR_BAA@darpa.mil with the topic number in the subject line, including the name, email address, and telephone number of a point of contact. Questions submitted within seven calendar days of the proposal due date may not be answered. DARPA posts a consolidated Frequently Asked Questions document under the topic number summary on its Small Business site, updated on an ongoing basis until one week prior to the proposal due date.
DARPA will not accept late proposals.
Classification, marking, and registrations
All proposals are required to be UNCLASSIFIED or CUI. Do not include any classified information in your proposal submission. Do not include any proprietary information on the Proposal Coversheet in Volume 1.
Proposal titles, abstracts, anticipated benefits, and keywords of proposals selected for contract award will undergo a DARPA Policy and Security Review and are subject to revision or redaction by DARPA. Final approved versions may appear on the DoW SBIR/STTR awards website and the SBA's award website at sbir.gov/awards.
Proposers should ensure they have an accurate and active entity registration on SAM.gov. Those engaging in ITAR or CUI work for DARPA must have CMMC Level 2 certification, and the projected requirement for this topic is Level 2 with self-assessment. DARPA points to sprs.csd.disa.mil/nistsp.htm and notes Project Spectrum at projectspectrum.io as an assistance resource.
Venture capital, hedge fund, and private equity ownership
Proposers that are more than 50 percent owned by multiple venture capital operating companies, hedge funds, private equity firms, or any combination of these as set forth in 13 CFR 121.702 are eligible to submit proposals in response to DARPA topics advertised within this BAA. Three conditions apply: register with the SBA Company Registry Database before submitting; submit the Majority-Owned VCOC, HF, and PEF Certification, with the SBIR VC Certification available on the DARPA Small Business site, in Supporting Documents Volume 5; and immediately notify the Contracting Officer, register in the appropriate SBA database, and submit the required certification if you enter that ownership class after submitting but before receiving a funding agreement.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW SBIR Program BAA. DARPA will conduct an evaluation of each conforming proposal. Proposals that do not comply with the requirements detailed in this BAA and the research objectives of the corresponding topic are considered non-conforming and are therefore not evaluated nor considered for award.
Using the evaluation criteria, the Government will evaluate each proposal in its entirety, documenting the strengths and weaknesses relative to each evaluation criteria, and based on those will determine the proposal's overall selectability for funding. Proposals will not be evaluated against each other but on their own individual merit.
A selectable proposal is one where the strengths of the overall proposal outweigh its weaknesses, with no accumulated weaknesses that would require extensive negotiations or a resubmitted proposal. A non-selectable proposal is one where the strengths do not outweigh its weaknesses.
Proposing firms will be notified of selection or non-selection status within 90 calendar days of the closing date of the BAA. The Corporate Official indicated on the Proposal Cover Sheet will be notified by email. In accordance with the SBA SBIR/STTR Policy Directive, Appendix I, paragraph 4, subparagraph (d), DARPA will provide a technical evaluation narrative to the proposer for each proposal submitted in response to a topic. An informal feedback session may additionally be requested via email at sbir@darpa.mil, provided at the sole discretion of DARPA.
Company Commercialization Report information will not be considered by DARPA during proposal evaluations.
Protests regarding the selection decision should be submitted, as prescribed in FAR 33.106(b) and FAR 52.233-3, to DARPA Contracts Management Office, 675 N. Randolph Street, Arlington, VA 22203, by email to CMO_SBIRProtests@darpa.mil and sbir@darpa.mil.
Post-award support
DARPA provides Transition and Commercialization Support Program services to Phase II and DP2 awardees upon contract execution at no cost to awardees. Awardees may also be eligible for the Embedded Entrepreneurship Initiative, an invitation-only program at DARPA's sole discretion, typically no more than $310,000 per awardee over the duration of the award, supporting a Senior Commercialization Advisor relationship, investor working group connections, and hiring an embedded entrepreneur to execute a Go-to-Market strategy.
The References
Two, both from the military medical literature rather than the modeling literature.
Quinn M.T., et al. (2024). Automating the Survival Chain and Revolutionizing Combat Casualty Care, Human-Technology Teaming on the Future Battlefield. Military Review.
Butler FK, Bennett B, Wedmore CI. Tactical Combat Casualty Care and Wilderness Medicine: Advancing Trauma Care in Austere Environments. Emergency Medicine Clinics of North America. 2017 May;35(2):391-407.
That both references are clinical and doctrinal rather than computational is a signal. DARPA is not steering you toward a modeling method. It is telling you to understand combat casualty care and the survival chain, and leaving the operations research to you. Butler is a foundational Tactical Combat Casualty Care author, and the Quinn paper on human-technology teaming in the survival chain is the conceptual frame for a decision aid that assists rather than replaces the medical planner.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
Technical question deadline: October 14, 2026, to SBIR_BAA@darpa.mil with the topic number in the subject line
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Selection notification: within 90 calendar days of BAA close
Period of performance: unresolved. 12 months per the index, 18 months plus a 6 month option per the narrative
A working backward plan
Before September 23. Submit the period of performance question to DARPA immediately, since it affects everything downstream. Download the mandatory DARPA Volume 2 and Volume 3 templates. Read the FAQ and keep rechecking it. Read both cited references. Assemble your feasibility package against the seven-item list plus the preliminary data requirement. Identify your data sources and be specific about theater, casualty, medical, natural, and virtual data, and about organic versus synthetic. Line up military users for the four prototype evaluations, because four different user groups is a real coordination burden. Decide your contract type and prepare the corresponding documents. Confirm SAM registration and your CMMC Level 2 self-assessment in SPRS. If venture-backed, register with the SBA Company Registry and obtain the SBIR VC Certification.
September 23 through October 5. Draft the 20 page white paper and 15 slide deck. Structure the white paper around the DDIL constraint, the casualty and needs prediction engine, the response option benefit metric as a function of time, dynamic uncertainty quantification, and the four-prototype iterative plan. Address civilian integration and DoD planning tool integration, both of which are stated requirements.
October 6 through October 14. Build the cost volume in the mandatory template to whichever period of performance DARPA confirms. Price four prototype builds and four user evaluation events, data acquisition or synthesis, and the tabletop exercise support.
October 15 through October 18. Assemble Volume 5 with contract-type documents and certifications, complete Volume 7 and the Volume 4 CCR, and run compliance: 20 page white paper, 15 slide deck, unclassified or CUI only, no proprietary information on the coversheet, mandatory cost template.
October 19 through October 20. Submit and certify in DSIP.
Frequently Asked Questions
What is DARPA SBIR topic DPA26BZ06-DV027?
DPA26BZ06-DV027 is a DARPA SBIR Direct to Phase II topic titled "Casualty Operations and Resource Prediction Software (CORPS)," released under the DoW 2026 SBIR Broad Agency Announcement, Release 6. It seeks a software tool to predict casualty and patient streams and needs following large scale medical emergencies in austere environments, and to project the lifesaving potential of various response options with limited or no access to real-time data or communications.
How much funding is available, and for how long?
The topic index states $750,000 over 12 months with no options. However, the topic narrative and milestone tables describe an 18 month base period plus a 6 month option period with milestones running through Month 24. These conflict, and the discrepancy needs to be resolved with DARPA before you finalize a cost volume or milestone plan.
How should I handle the schedule conflict?
Submit the question to SBIR_BAA@darpa.mil with the topic number in the subject line, well before the October 14 deadline, and watch the consolidated FAQ. Because DARPA treats non-compliant proposals as non-conforming and does not evaluate them, this is not a cosmetic issue. Absent an answer, the safer construction is to propose to the index figures, since the index is the controlling award structure table, while showing how your plan maps onto DARPA's six-milestone base structure compressed to the funded period.
When is the proposal deadline?
October 21, 2026. DARPA will not accept late proposals. Note the separate technical question deadline of October 14, 2026.
Can I submit a Phase I proposal?
No. This topic is soliciting Direct to Phase II proposals only, and the offeror shall provide detail and documentation demonstrating the accomplishment of a "Phase I-like" effort demonstrating feasibility and proof-of-concept.
What feasibility documentation is required?
DARPA lists seven things a demonstration of feasibility may include: conceptual characterization of the complete product; relevant use cases, required data categories, and anticipated final capabilities; leveraged systems and workflow as relevant; proposed approach or methodology for model development as relevant; anticipated interactions between data and model components as relevant; the presentation of scientific and technical material supporting the above; and design specifications for any computational or software components of a prototype. Separately and not optionally, proposals should contain preliminary data, published or unpublished, supporting the rationale for the development of the candidate products.
Is the feasibility bar lower than on other DARPA topics in this release?
Notably so. This topic asks for conceptual characterization, methodology, design specifications, and preliminary data rather than a validated working prototype or fielded system. That opens it to modeling, analytics, and operations research teams with strong underlying science and a credible design. If you have published casualty modeling, medical logistics optimization, or mass casualty operations research, read the requirement before assuming you are out of scope.
What is the DDIL constraint?
Crisis action decisions must be made in much shorter time frames than strategic planning, using sparse, low-quality situational data that cannot be verified due to lack of secure communications and denied, disrupted, intermittent, and limited-bandwidth states. The topic requires that proposed approaches be adaptable to variable levels of communication capability and medical care, which means your tool must degrade gracefully and produce usable output at different confidence levels depending on connectivity.
What are the three core functions?
Automated prediction of casualty and patient streams. Identification of the related medical needs with variable latencies. And metrics-based evaluation of benefits gained from various responses as a dynamic function of time. The third is the differentiator, because it converts a subjective judgment about medical benefit into a quantitative one.
Does the tool have to work in the civilian sector?
Yes. Solutions must have the potential for use in the civilian sector, meaning integration with civilian networks and infrastructure, as well as integration into existing DoD planning tools. Both integrations are stated requirements, not commercialization aspirations.
What technology readiness level must I start at?
Offerors should only propose products at or above Technology Readiness Level 2.
How many prototypes must I deliver?
No less than 4 prototypes supporting different military user evaluations. Each prototype should demonstrate increasing operational capability and maturity while reducing technical risk through measurable performance objectives. The milestone table names Prototype 1 through Prototype 4 at Months 8, 11, 15, and 18 respectively, each with a presentation and demonstration.
What are the four prototypes for?
Per the milestone table, Prototype 1 is the technical baseline and decision optimization prototype integrating casualty prediction with medical resource allocation and response optimization to evaluate alternative Courses of Action. Prototype 2 is decision support and interaction, adding an interactive interface with user-defined constraints and dynamic uncertainty quantification. Prototype 3 is the integrated operational prototype demonstrated in a tabletop exercise. Prototype 4 is the operational prototype demonstration with the Phase III transition strategy.
What is dynamic uncertainty quantification and why does it matter?
It appears both in the Prototype 2 milestone, which requires a dynamic uncertainty and confidence metric demonstrated, and in the Phase III strategy requirement. Because the whole premise is decision-making on sparse, unverifiable data, quantified uncertainty is what makes the output trustworthy rather than a false-precision number. Treat it as a core feature rather than a refinement.
What data sources should I plan for?
The topic names anticipated data sources including theater, casualty, medical, natural, and virtual, and types including organic versus synthetic. It also requires you to outline required data, data sources and environments, data quality management, and model development design and integration. The M2 milestone requires validating the baseline casualty prediction engine using representative military and civilian datasets with varying data completeness.
What testing is expected?
The proposal shall describe planned prototype design, product development, testing, and validation of the prototype product in table-top like exercises, relevant to the requirements of medical command and control functions in austere and contested environments. The M5 milestone requires a successful tabletop exercise completed.
How long can my technical volume be?
This topic uses the White Paper and Slide Deck format. The white paper shall not exceed 20 pages and the slide deck shall not exceed 15 slides. Refer to Appendix B, DARPA Direct to Phase II Instructions, for the content of each element and the commercialization strategy.
What happens in Phase III?
Phase II work will result in a final phase funded by sources other than a Federal government SBIR Program, and Phase III awards may be made by any Government entity without further competition, creating a sole-source-like pathway. Phase III matures the prototype into an operational decision-support capability, potentially transitioning to an Acquisition Program managed by Service Product Developers for inclusion into Service-specific mobile applications. Deployment will require appropriate cybersecurity certifications and the ability to join an existing Authority to Operate or obtain a separate ATO.
Do I need to think about an ATO?
Yes, though not in Phase II. Phase III deployment requires appropriate cybersecurity certifications and the ability to join an existing ATO or obtain a separate one, and the O3 option milestone asks for a preliminary integration and cyber security strategy documented. Building toward that from the start is better than retrofitting.
Can I ask questions through DSIP Topic Q&A?
No. DARPA states DSIP Topic Q&A will not be available for these topics. Technical questions go to SBIR_BAA@darpa.mil with the topic number in the subject line by October 14, 2026. Questions submitted within seven calendar days of the due date may not be answered. DARPA posts a consolidated FAQ, updated until one week before the due date.
Do I have to choose a contract type?
Yes. Proposers must state their requested contract type. DARPA may award FAR-based firm-fixed-price or cost-plus reimbursement contracts, or Other Transactions for Prototype under 10 U.S.C. 4021. Cost-plus requires your DCMA Final Determination Letter showing accounting system approval. An OT requires a completed Model OT plus OT Certifications in Volume 5. Firm-fixed-price requires no additional action.
Is the cost template mandatory?
Yes. Templates for Volume 2 and Volume 3 are on the DARPA Small Business website, and use of the DARPA Cost Proposal template is mandatory.
Are venture capital backed companies eligible?
Yes. Proposers more than 50 percent owned by multiple venture capital operating companies, hedge funds, private equity firms, or any combination as set forth in 13 CFR 121.702 are eligible, subject to registering with the SBA Company Registry Database before submitting, submitting the Majority-Owned VCOC, HF, and PEF Certification in Volume 5, and notifying the Contracting Officer if you enter that class after submitting but before award.
How will my proposal be evaluated?
Against the evaluation criteria in the DoW SBIR Program BAA. DARPA evaluates each conforming proposal in its entirety, documenting strengths and weaknesses relative to each criterion, then determines overall selectability. Proposals are not evaluated against each other. A selectable proposal is one where strengths outweigh weaknesses with no accumulated weaknesses requiring extensive negotiations or resubmission.
Will I get feedback if not selected?
Yes. DARPA will provide a technical evaluation narrative to the proposer for each proposal submitted in response to a topic, per the SBA SBIR/STTR Policy Directive. An informal feedback session may additionally be requested via sbir@darpa.mil, at DARPA's sole discretion.
What is the commercial application?
The product will provide a similar capability for the planning of response to non-military emergency events, for example following natural disasters or war. Combined with the stated requirement to integrate with civilian networks and infrastructure, the buyers are emergency management agencies, hospital systems and trauma networks doing surge planning, and disaster response organizations.
Who do I contact with questions?
The DARPA Small Business Programs Office at SBIR_BAA@darpa.mil for both program administration and topic technical questions, with the topic number in the subject line. DSIP technical support at DoDSBIRSupport@reisystems.com with a copy to SBIR_BAA@darpa.mil, Monday through Friday 9:00 a.m. to 5:00 p.m. ET. DARPA also offers free resources through DARPAConnect at DARPAConnect.us.
Positioning Advice for Companies Considering This Topic
Resolve the period of performance first. Twelve months with no options versus eighteen months plus a six month option is a factor-of-two difference in scope, staffing, and price. Ask DARPA before you write, because a proposal built to the wrong structure risks being found non-conforming.
Recognize that this is a softer feasibility bar than its neighbors. Several topics in this DARPA release demand validated fielded prototypes. This one asks for conceptual characterization, methodology, design specifications, and preliminary data. Operations research groups, casualty modeling teams, and medical logistics analytics companies who would be ineligible elsewhere should look here.
Make the benefit metric the intellectual core. Predicting casualties is the entry price. Quantifying the lifesaving benefit of each response option as a dynamic function of time is what the topic actually lacks and what decision makers need. Define your metric, ground it in survivability science, and explain how it lets a planner compare a fast partial response against a slower complete one.
Design for degradation, explicitly. The DDIL constraint means your tool must work at full connectivity, at intermittent connectivity, and at none. Show three operating modes with the confidence bounds each produces. A proposal that assumes data availability has misread the topic's central premise.
Make uncertainty quantification visible in the interface. Prototype 2 requires a dynamic uncertainty and confidence metric demonstrated, and Phase III requires dynamic uncertainty quantification to support operational decision-making. A planner who cannot see how much to trust a projection will not use it. This is an interface problem as much as a statistical one.
Line up four different military user groups now. The four-prototype requirement specifies different military user evaluations, and each prototype has a presentation and demonstration milestone. Access to military medical planners for four separate evaluation events is a coordination burden that will not appear from nowhere at Month 8. Naming your evaluation partners strengthens the proposal considerably.
Answer both integration requirements. Civilian networks and infrastructure, and existing DoD planning tools. These are stated as requirements. Naming specific DoD planning systems you would interoperate with, and specific civilian emergency management systems, converts a generic claim into a credible one.
Ground the work in TCCC and the survival chain. Both references are clinical and doctrinal, not computational. Butler is foundational Tactical Combat Casualty Care, and the Quinn paper frames human-technology teaming in the survival chain. Using that vocabulary correctly signals that you understand the medical decision the tool supports, rather than treating it as an abstract allocation problem.
Design for a mobile end state. Phase III transition is to Service-specific mobile applications. A desktop-first interface will need rework. Thinking about screen real estate, offline operation, and low-bandwidth sync from Prototype 2 onward is cheap now and expensive later.
Start the cybersecurity story early. CMMC Level 2 self-assessment is the projected requirement, the option period asks for a preliminary cyber security strategy, and Phase III requires ATO. For a tool that will handle casualty data and join a Service network, security architecture is not a late-stage concern.
DARPA SBIR DPA26BZ06-DV026: Influence Benchmarks for AI Systems
Deadline: October 21, 2026
Funding Award Size: $1.8m
Description: Complete guide to DARPA SBIR topic DPA26BZ06-DV026, influence benchmarks for AI systems using simulated markets. Phase I $300K or DP2 $1.8M. Closes October 21, 2026.
Quick Answer
DPA26BZ06-DV026 is a DARPA SBIR topic under the DoW 2026 SBIR Broad Agency Announcement, Release 6, and one of only two topics in this release accepting both Phase I and Direct to Phase II proposals. DARPA wants a simulated market, an auction environment, used as a testbed to characterize the latent behavioral preferences of AI systems. The premise is that economic frameworks let you measure how an AI agent actually behaves under incentives, including whether it deceives or collaborates, without needing a prior definition of what counts as harmful. Phase I is $300,000 over 12 months. Direct to Phase II is $1,800,000 over 24 months with no options, which makes it the largest single-tranche award in this DARPA release. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The design constraint that makes this topic interesting is stated in one sentence: such a testbed shall rely only on queries and outputs of the subject AI system, rather than direct access to the model itself. You are building a black-box behavioral assay. You do not get weights, gradients, or internals. You get to ask the model things and watch what it does in a market.
Topic At a Glance
Topic number: DPA26BZ06-DV026
Title: Influence Benchmarks for AI Systems
Agency: Defense Advanced Research Projects Agency (DARPA)
Solicitation: DoW 2026 Small Business Innovation Research Broad Agency Announcement, Release 6, DARPA Proposal Submission Instructions
Program types accepted: both Phase I and Direct to Phase II
Phase I award: $300,000 over 12 months. Technical volume is a 10 page white paper plus a 5 page slide deck
Direct to Phase II award: $1,800,000 over 24 months. No Option 1 and no Option 2
Direct to Phase II technical volume format: Standard Proposal Format, 35 pages
OUSD (R&E) Critical Technology Area: Applied Artificial Intelligence
Component Technology Priority Area: Human-Machine Interfaces
Projected CMMC level requirement: Level 1
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic
Access model: black box only. The testbed shall rely only on queries and outputs of the subject AI system, not direct access to the model
Technical and Business Assistance: up to $6,500 for Phase I awardees, up to $25,000 per Phase II project, in addition to the cost ceilings
Topic Q&A: DSIP Topic Q&A is not available for DARPA topics. Technical questions go to SBIR_BAA@darpa.mil by October 14, 2026
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Submission portal: DSIP at dodsbirsttr.mil
Keywords: AI, LLM, cognitive bias, economic model, market, information
Two Entry Points, and the Unusual Award Shape
This topic appears in both the Phase I and the Direct to Phase II award structure tables.
Phase I: $300,000 over 12 months, with a 10 page white paper and a 5 page slide deck. Note that 12 months is a long Phase I by DoW standards, and the topic's Phase I description is structured around month-by-month milestones running to Month 12, so the schedule is intentional rather than incidental.
Direct to Phase II: $1,800,000 over 24 months, Standard Proposal Format, 35 pages. There are no options on this topic, which is worth noticing. Most DP2 awards in this release use a base plus one or two priced options, giving DARPA off-ramps. Here the entire $1,800,000 is committed as a single 24 month effort. That means DARPA is buying a complete program in one decision, and the proposal has to justify all 24 months up front rather than relying on option exercise to prove itself.
The DP2 requirement is stated plainly: proposers submitting Direct to Phase II should be able to demonstrate an existing test environment that meets the criteria outlined in Phase I. So the Phase I description doubles as the DP2 feasibility specification. Read it that way.
What DARPA Is Actually Looking For
The objective
Develop a testbed utilizing economic frameworks to benchmark AI biases in dynamic environments. This tool will evaluate how agents adapt, collaborate, or deceive within a simulated marketplace. Ultimately, these assessments will ensure safe and effective AI integration in high-stakes decision-making.
The threat model
As warfighters increasingly engage with AI systems, particularly agents that rely on large language models, concern has arisen that the systems may encourage cognitive behavior in users that impart hidden biases, impair judgement, and ultimately degrade warfighting capacity.
DARPA gives three concrete mechanisms, each with a citation. Overreliance on AI-powered decision support tools may induce users to accept erroneous suggestions or change from correct decisions to incorrect decisions, citing Bucinca et al. 2021. LLM use may also induce novel cognitive biases, citing Alessa et al. 2025. And LLM use may amplify delusional beliefs, citing Dohnany et al. 2025.
Compounding the threat, deceptive strategies frequently emerge among interacting AI systems, citing Ying et al. 2026. Few tests of AI systems account for their adaption to dynamic data, which obfuscates such adaptive strategies.
To detect and combat these risks, the DoW requires a universal approach to elicit, characterize, and compare the behavior of AI agents amid changing contexts, citing Li et al. 2026.
That word "universal" is doing work. DARPA is not asking for a test of one model family. It is asking for an instrument that can be pointed at any AI system and produce comparable measurements.
Why economics, and why a market
Economic frameworks permit the measurement and comparison of decisions. As the basis of extensive prior research, models of auctions, markets, and other economic arenas provide a critical baseline of organic patterns of human behavior, citing Hausch 1986 and Martinez-Saito 2019.
Emerging open-source tools, such as Magentic Marketplace, citing Bansal et al. 2026, have shown promise in revealing behavioral variations in market-focused contexts.
Then the argument that distinguishes this approach from mainstream AI safety evaluation: furthermore, unlike existing assessments of AI risk, economic frameworks do not rely on a priori definitions of "harmful" or "helpful" traits, citing Vijayvargiya et al. 2026, and these testbeds allow for diverse social strategies such as deception and collaboration.
That is the intellectual core of the topic. Most AI safety benchmarks require you to define the bad behavior in advance and then test for it. A market does not: it reveals preferences through revealed choices under incentives, and it has decades of human baseline data to compare against. If you write only one thing well in this proposal, make it your grasp of this argument.
The black-box constraint
Such a testbed shall rely only on queries and outputs of the subject AI system, rather than direct access to the model itself.
This is a hard architectural requirement and it shapes everything. You cannot inspect activations, probe internal representations, or fine-tune the subject model. Your instrument observes behavior in a market and infers latent preferences from it. That is both the constraint and the reason the approach generalizes to any commercial or adversary model you can query.
The Phase I Program, Which Is Also the DP2 Feasibility Bar
The goal of Phase I is to develop and define the architecture of the test environment and demonstrate a functional proof of concept for a simulated auction or market, referred to throughout as the market.
The test environment must be able to accommodate different models of auctions and markets to evaluate agents on different measures of market efficiency. Proposers must also develop classifiers to detect biases in AI agent behavior and outcomes in the market environment. This test environment should be a sandbox in which multiple AI agents may operate.
The month-by-month Phase I requirements
By Month 2, proposers should have defined the core mechanics of the test environment, such as market structures, payout schemes, measures of market efficiency and consumer utility, and a dynamic "news feed" that informs the behavior of AI agents within the market. The news feed is meant to simulate public information relevant to the market.
By Month 6, proposers should establish a strategy for scaling issues and define how AI agents interact through their decision space: a set of bids, each of which is defined by the time, asset, quantity, and price offered or requested on the market.
Note that four-tuple. A bid is time, asset, quantity, and price. That is the atomic unit of observable behavior in your instrument, and defining the decision space this precisely is what makes cross-model comparison possible.
By Month 9, proposers must demonstrate a suite of "stock" AI agents for use within the test environment. These AI agents shall only be used within the test environment sandbox and shall not be used on other systems.
By Month 12, proposers should have a proof-of-concept test environment for assessing behavior preferences of AI systems amid dynamic data. In the proof-of-concept, the test environment must be able to replicate bidding patterns and allocation efficiencies observed in markets and demonstrate allocative efficiencies of greater than 90 percent. The environment and associated data must be sufficient for comparison to models of human behavior in auctions or markets, as well as to compare agents to one another.
That greater than 90 percent allocative efficiency figure is the only hard numeric performance target in the topic. It is a validation criterion: if your simulated market does not clear efficiently, it is not a credible market and any behavioral inference from it is suspect. Address it directly.
Phase I fixed payable milestones
Month 2: report on initial architecture and core mechanics of test environment and classifiers of AI agent behavior.
Month 6: report on scaling and AI agent interaction within the test environment.
Month 9: demonstration of a suite of stock AI agents, drawn from at least 10 different LLMs, operating within the test environment.
Month 12: demonstration of test environment and AI agent behavior within the marketplace, with allocative efficiencies of greater than 90 percent.
The Month 9 milestone contains a requirement the narrative does not state: stock AI agents drawn from at least 10 different LLMs. Ten distinct models is a meaningful integration and cost burden, covering API access, prompt harnesses, rate limits, and version drift across a dozen vendors and open-weight families. Budget for it and name your intended model set.
The Phase II Program
Proposers submitting Direct to Phase II should be able to demonstrate an existing test environment that meets the criteria outlined in Phase I.
In Phase II, proposers should develop a range of AI agents as stock reference models against which the performance of a test AI agent may be compared.
Proposers must define an error function that describes differences between AI agent decisions and expected human results.
The stock AI agents should help identify and bound agent behaviors of concern, such as hidden biases. At least one AI agent should represent the human user of the test AI agent, to explore how the test AI agent may drive human behaviors of concern such as encouraging delusions or eroding military discipline.
That last sentence is the most striking requirement in the topic. You are not only benchmarking the AI agent's market behavior. You are instantiating a simulated human user as an agent in the market, so you can observe whether the test AI influences that user toward delusion or indiscipline. The topic's title is Influence Benchmarks, and this is where influence is actually measured.
Proposers should define analytics to compare the behavior of the AI agents.
The Phase II timeline in narrative form
By Month 6, proposers should have a suite of such stock AI agents, representing both human and native AI agent behavior. Agents should be able to interact with other agents in the test environment to develop strategic bids in an attempt to influence the behavior of other market actors to maximize their expected profit.
By Month 12, proposers should demonstrate and measure how each AI agent responds to the news feed and the observable behavior of other agents, which alters each agent's expectation of future asset values and behaviors of other agents.
By Month 15, proposers should assess social interactions among AI agents to model behaviors including, but not limited to, collaboration and deception. Proposers must provide quantitative metrics on how social behaviors between AI agents alter market efficiency and outcomes and generate testable hypotheses for how latent AI agent preferences affect social interactions, dynamic responses to external stimuli, and ultimately market outcomes.
By Month 18, proposers should test their own hypotheses across a range of AI agents and quantify differences between human-like agents and AI agents.
By Month 21, proposers should demonstrate the test environment to DARPA, and deliver software and data for testing by U.S. Government partners to ensure that the environment may be used to reliably characterize the latent preferences of a test AI agent.
By Month 24, the proposers should deliver a final report describing the performance of the test environment and AI agents, and include a transition plan.
Phase II fixed payable milestones
Month 2: a report on new capabilities that will be added to the test environment to enable analysis of dynamic AI agent behavior and outcomes. Provide definitions of an error function to compare AI agent decisions versus human decisions, as well as measures for comparing distributions of human and AI market outcomes.
Month 6: demonstration of "human" stock AI agents that replicate human preferences and behavior in the market, and a report on how a "human" stock AI agent simulates expected human interaction in the market. Test Phase 1 classifier's ability to distinguish "human" versus AI agents in the market environment, and propose improvements to classification algorithm or alternative test statistics.
Month 12: report on dynamic behavior of AI agents in response to news events and observed actions of other agents within the test environment. Evaluate agent behavior using improved classifiers.
Month 15: report on modeling social interactions between AI agents, including impact on bidding behavior of AI agents.
Month 18: report on how latent AI preferences impact social interactions between AI agents, dynamic responses, and market outcomes.
Month 21: final software delivery, both object and source code, for operation by DARPA or other U.S. Government personnel for additional demonstrations, with suitable documentation in a contractor proposed format.
Month 24: final report, including quantitative metrics on AI agent behavior and fidelity of the test environment. The report must also discuss advances in LLM models and AI agents that have occurred during the period of performance and their potential impact on the evaluation of AI models.
Two of these deserve emphasis. The Month 21 delivery is object and source code for operation by DARPA or other U.S. Government personnel. That is a source code delivery to the Government, and it has direct implications for how you structure intellectual property and what you build on. Address data rights deliberately.
And the Month 24 requirement to discuss advances in LLM models during the period of performance and their impact on the evaluation approach is an unusual and telling ask. DARPA is acknowledging that the measurement target will move underneath you across 24 months, and asking you to reason about whether your instrument survives that. A proposal that addresses model drift and instrument durability up front is answering a question DARPA has already flagged.
Phase III Dual Use
DARPA's commercial argument here is specific and unusually persuasive, and it is worth quoting closely because it is your commercialization strategy handed to you.
The potential risk of AI systems inducing harmful behaviors is a major commercial concern due to the associated legal and reputational risk to their developers. AI firms are actively seeking mechanisms to argue the limits of their systems' biases and impacts on user behavior, including measuring deviations from expected human norms of behavior, in response to lawsuits. The impact on a firm's stock value of establishing the scope of liability far exceeds the already-robust immediate financial consequences of these court cases, suggesting that the AI firms have strong market demand for the capability to benchmark the influence of their systems.
Read that as a liability-driven market thesis. The buyer is an AI developer facing litigation over user harm, and the product is a defensible, quantitative measurement of how far its system's influence deviates from human norms. That is a market with willingness to pay set by legal exposure rather than by research budgets, which is a much better commercial position than most AI evaluation tooling occupies.
Extend it naturally to insurers underwriting AI liability, to regulators needing a measurement standard, and to enterprise buyers performing vendor due diligence. But lead with DARPA's own framing.
Funding, Cost Structure, and DARPA Mechanics
The awards
Phase I: $300,000 over 12 months, 10 page white paper plus 5 page slide deck.
Direct to Phase II: $1,800,000 over 24 months, Standard Proposal Format 35 pages, no options.
The resources made available for each topic will depend on the quality of the proposals received and the availability of funds. The Government reserves the right to select for negotiation all, some, one, or none of the proposals received and to make awards with or without communications with proposers. Because this topic has no options, the option-exercise language elsewhere in the instructions does not apply here.
The Standard Proposal Format structure for DP2
DP2 feasibility documentation shall not exceed 10 pages. The DP2 technical proposal shall not exceed 20 pages. The Phase II commercialization strategy shall not exceed five pages, and this should be the last section of the technical volume. Those three total the 35 page limit. Refer to Appendix B, DARPA Direct to Phase II Instructions, for the content of each element.
Contract type, which you must elect
DARPA may award FAR-based contracts, firm-fixed-price or cost-plus reimbursement, or Other Transactions for Prototype under the authority of 10 U.S.C. 4021, subject to approval of the Contracting Officer or Agreements Officer respectively. Proposers must state their requested contract type in their proposal.
Cost-plus reimbursement requires including your Defense Contract Management Agency Final Determination Letter showing approval of your accounting system. An Other Transaction for Prototype requires including a completed OT using the Model OT for Prototype from the DARPA Small Business site, plus completed OT Certifications, both loaded in Volume 5, with at minimum the color-coded areas completed and redlines with explanations for any article you wish to negotiate. Firm-fixed-price requires no additional action.
For a software program with fixed payable milestones and a source code delivery, the choice of instrument interacts with your intellectual property position. Think about it early rather than defaulting.
Templates are mandatory
Templates for Volume 2 Technical Volume and Volume 3 Cost Volume are provided as attachments on the DARPA Small Business website. Use of the DARPA Cost Proposal template is mandatory.
Technical and Business Assistance
Phase I awardees may request up to $6,500. Phase II awardees may request up to $25,000 per Phase II project. TABA funding is in addition to the cost ceilings and is not subject to profit or fee. Requests will be reviewed by the respective contracting office or specialist at time of award.
For this topic, intellectual property protections and market validation are both directly useful, given the source code delivery requirement and the liability-driven commercial thesis.
Questions and the FAQ
DSIP Topic Q&A will not be available for these DARPA topics. Technical questions must be submitted by October 14, 2026, by email to SBIR_BAA@darpa.mil with the topic number in the subject line, including the name, email address, and telephone number of a point of contact. Questions submitted within seven calendar days of the proposal due date may not be answered. DARPA posts a consolidated Frequently Asked Questions document under the topic number summary on its Small Business site, updated on an ongoing basis until one week prior to the proposal due date.
DARPA will not accept late proposals.
Classification, marking, and registrations
All proposals are required to be UNCLASSIFIED or CUI. Do not include any classified information in your proposal submission. Do not include any proprietary information on the Proposal Coversheet in Volume 1.
Proposal titles, abstracts, anticipated benefits, and keywords of proposals selected for contract award will undergo a DARPA Policy and Security Review and are subject to revision or redaction by DARPA. Final approved versions may appear on the DoW SBIR/STTR awards website and the SBA's award website at sbir.gov/awards.
Proposers should ensure they have an accurate and active entity registration on SAM.gov. Those engaging in ITAR or CUI work for DARPA must have CMMC Level 2 certification, though the projected requirement for this topic is Level 1. DARPA points to sprs.csd.disa.mil/nistsp.htm and notes Project Spectrum at projectspectrum.io as an assistance resource.
Venture capital, hedge fund, and private equity ownership
Proposers that are more than 50 percent owned by multiple venture capital operating companies, hedge funds, private equity firms, or any combination of these as set forth in 13 CFR 121.702 are eligible to submit proposals in response to DARPA topics advertised within this BAA. Three conditions apply: register with the SBA Company Registry Database before submitting; submit the Majority-Owned VCOC, HF, and PEF Certification, with the SBIR VC Certification available on the DARPA Small Business site, in Supporting Documents Volume 5; and immediately notify the Contracting Officer, register in the appropriate SBA database, and submit the required certification if you enter that ownership class after submitting but before receiving a funding agreement.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW SBIR Program BAA. DARPA will conduct an evaluation of each conforming proposal. Proposals that do not comply with the requirements detailed in this BAA and the research objectives of the corresponding topic are considered non-conforming and are therefore not evaluated nor considered for award.
Using the evaluation criteria, the Government will evaluate each proposal in its entirety, documenting the strengths and weaknesses relative to each evaluation criteria, and based on those will determine the proposal's overall selectability for funding. Proposals will not be evaluated against each other but on their own individual merit.
A selectable proposal is one where the strengths of the overall proposal outweigh its weaknesses, with no accumulated weaknesses that would require extensive negotiations or a resubmitted proposal. A non-selectable proposal is one where the strengths do not outweigh its weaknesses.
Proposing firms will be notified of selection or non-selection status for a Phase I or Direct to Phase II award within 90 calendar days of the closing date of the BAA. The Corporate Official indicated on the Proposal Cover Sheet will be notified by email. In accordance with the SBA SBIR/STTR Policy Directive, Appendix I, paragraph 4, subparagraph (d), DARPA will provide a technical evaluation narrative to the proposer for each proposal submitted in response to a topic. An informal feedback session may additionally be requested via email at sbir@darpa.mil, provided at the sole discretion of DARPA.
Company Commercialization Report information will not be considered by DARPA during proposal evaluations.
Protests regarding the selection decision should be submitted, as prescribed in FAR 33.106(b) and FAR 52.233-3, to DARPA Contracts Management Office, 675 N. Randolph Street, Arlington, VA 22203, by email to CMO_SBIRProtests@darpa.mil and sbir@darpa.mil.
Post-award support
DARPA provides Transition and Commercialization Support Program services to Phase II and DP2 awardees upon contract execution at no cost to awardees. Awardees may also be eligible for the Embedded Entrepreneurship Initiative, an invitation-only program at DARPA's sole discretion, typically no more than $310,000 per awardee over the duration of the award, supporting a Senior Commercialization Advisor relationship, investor working group connections, and hiring an embedded entrepreneur to execute a Go-to-Market strategy. Given that the commercial thesis here is a liability-driven enterprise sale to AI developers, an embedded entrepreneur with that market's access would be materially useful.
The Reference List, Which Is the Intellectual Map
Eight references, and they divide cleanly into three groups that together define the topic's intellectual position.
The AI harm evidence. Bucinca Z., Malaya M., Gajos K. (2021), To Trust or to Think: Cognitive Forcing Functions Can Reduce Overreliance on AI in AI-assisted Decision-making, Proceedings of the ACM on Human-computer Interaction 5, CSCW, 1-21. Alessa A., Somane P., Lakshminarasimhan A., Skirzynski J., McAuley J., Echterhoff J. (2025), Quantifying Cognitive Bias Induction in LLM-Generated Content. Dohnany S., Kurth-Nelson Z., Spens E., Luettgau L., Reid A., Gabriel I., Arulkumaran K., and Nour M. M. (2025), Technological folie a deux: Feedback loops between AI chatbots and mental illness, arXiv:2507.19218.
The economics baseline. Hausch DB. (1986), Multi-Object Auctions: Sequential vs. Simultaneous Sales, Management Science 32(12):1599-1610. Martinez-Saito M., Konovalov R., Piradov MA., Shestakova A., Gutkin B., Klucharev V. (2019), Action in auctions: neural and computational mechanisms of bidding behaviour, European Journal of Neuroscience 50(8): 3327-3348.
The AI agent evaluation tooling. Li E., Bellotti V., Sessions N., Kao R. (2026), Mastering Agentic Techniques: AI Agent Evaluation. Bansal G. et al. (2026), Magentic Marketplace: An Open-Source Environment for Studying Agentic Markets, arXiv:2510.25779v1. Vijayvargiya S., Bharat Soni A., Zhou X., Zhiruo Z., Wang Z.Z. (2026), OpenAgentSafety: A Comprehensive Framework for Evaluating Real-World AI Agent Safety, arXiv:2507.06134v2.
Three things follow. The Hausch and Martinez-Saito references are your human baseline: sequential versus simultaneous multi-object auction theory and the neural and computational mechanisms of human bidding. If you cannot speak to human bidding behavior in auctions, you cannot build the comparison the topic requires.
Magentic Marketplace is named in the narrative as an emerging open-source tool showing promise, which is as close as DARPA comes to a build-on recommendation. Understand it and state your relationship to it, whether you extend it, differentiate from it, or replace it.
And OpenAgentSafety is cited precisely as the thing this approach improves on, because it relies on a priori definitions of harmful traits. Engaging with why revealed-preference measurement in a market is better than checklist safety evaluation is the argument that makes your proposal look expert rather than derivative.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
Technical question deadline: October 14, 2026, to SBIR_BAA@darpa.mil with the topic number in the subject line
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Selection notification: within 90 calendar days of BAA close
Phase I period: 12 months
Direct to Phase II period: 24 months, no options
A working backward plan
Before September 23. Decide your entry point: DP2 requires an existing test environment meeting the Phase I criteria, including the greater than 90 percent allocative efficiency demonstration and stock agents from at least 10 LLMs. Download the mandatory DARPA Volume 2 and Volume 3 templates. Read the FAQ and keep rechecking it. Read the eight references, particularly Hausch on multi-object auctions, Martinez-Saito on human bidding mechanisms, Magentic Marketplace, and OpenAgentSafety. Settle your market design: structures, payout schemes, efficiency measures, consumer utility, and the news feed mechanism. Enumerate your intended 10-plus LLMs and confirm API access, terms of service compatibility, and cost. Work out your data rights position given the Month 21 object and source code delivery to the Government. Decide your contract type and prepare the corresponding documents. Confirm SAM registration. If venture-backed, register with the SBA Company Registry and obtain the SBIR VC Certification. Submit technical questions before October 14.
September 23 through October 5. Draft to your format. For Phase I, a 10 page white paper plus 5 slides structured on the Month 2, 6, 9, and 12 milestones. For DP2, 10 pages of feasibility documentation evidencing your existing environment, 20 pages of technical proposal built on the Month 2 through 24 milestone structure, and 5 pages of commercialization strategy built on DARPA's liability thesis. Address the black-box constraint, the error function definition, the simulated human user agent, and the allocative efficiency target explicitly.
October 6 through October 14. Build the cost volume in the mandatory template. Price LLM API consumption honestly, since a market simulation with many agents across a dozen models over 24 months is a substantial inference bill. Price the source code delivery and documentation work at Month 21.
October 15 through October 18. Assemble Volume 5 with contract-type documents and certifications, complete Volume 7 and the Volume 4 CCR, and run compliance: page and slide limits for your format, unclassified or CUI only, no proprietary information on the coversheet, mandatory cost template.
October 19 through October 20. Submit and certify in DSIP.
Frequently Asked Questions
What is DARPA SBIR topic DPA26BZ06-DV026?
DPA26BZ06-DV026 is a DARPA SBIR topic titled "Influence Benchmarks for AI Systems," released under the DoW 2026 SBIR Broad Agency Announcement, Release 6. It seeks a testbed utilizing economic frameworks to benchmark AI biases in dynamic environments, evaluating how agents adapt, collaborate, or deceive within a simulated marketplace, to ensure safe and effective AI integration in high-stakes decision-making.
Can I submit either a Phase I or a Direct to Phase II proposal?
Yes. This topic appears in both award structure tables. Phase I is $300,000 over 12 months with a 10 page white paper and 5 page slide deck. Direct to Phase II is $1,800,000 over 24 months using the Standard Proposal Format at 35 pages.
Are there options on the Direct to Phase II award?
No. The award structure table lists no Option 1 and no Option 2 for this topic. The full $1,800,000 is a single 24 month commitment, which makes it the largest single-tranche award in this DARPA release and means your proposal must justify all 24 months up front.
How do I know if I qualify for the Direct to Phase II path?
The topic states that proposers submitting Direct to Phase II should be able to demonstrate an existing test environment that meets the criteria outlined in Phase I. So the Phase I description is the DP2 feasibility specification, including the greater than 90 percent allocative efficiency demonstration and the suite of stock AI agents drawn from at least 10 different LLMs.
Can I access the model I am testing?
No. The topic states that such a testbed shall rely only on queries and outputs of the subject AI system, rather than direct access to the model itself. You are building a black-box behavioral assay. That is both the constraint and the reason the approach generalizes to commercial and adversary models you can only query.
Why use economic frameworks instead of a safety benchmark?
Because economic frameworks permit measurement and comparison of decisions against decades of human baseline data from auction and market research, and because unlike existing assessments of AI risk they do not rely on a priori definitions of harmful or helpful traits. Markets also allow for diverse social strategies such as deception and collaboration to emerge rather than needing to be specified in advance.
What is the hard performance target?
Greater than 90 percent allocative efficiency in the proof-of-concept test environment by Month 12, along with the ability to replicate bidding patterns and allocation efficiencies observed in markets. It is the only numeric target in the topic and it functions as a validity check: an inefficient simulated market undermines any behavioral inference drawn from it.
How many LLMs do I need to integrate?
At least 10. The Month 9 Phase I fixed payable milestone requires demonstration of a suite of stock AI agents drawn from at least 10 different LLMs operating within the test environment. Budget for API access, prompt harnesses, rate limits, and version drift across that many models.
How is a bid defined?
By Month 6, proposers should define how AI agents interact through their decision space: a set of bids, each of which is defined by the time, asset, quantity, and price offered or requested on the market. That four-tuple is the atomic unit of observable behavior and what makes cross-model comparison possible.
What is the news feed?
A dynamic mechanism, defined by Month 2, that informs the behavior of AI agents within the market and is meant to simulate public information relevant to the market. By Month 12 of Phase II, proposers must demonstrate and measure how each AI agent responds to the news feed and the observable behavior of other agents, which alters each agent's expectation of future asset values and behaviors of other agents.
What is the simulated human user requirement?
In Phase II, at least one AI agent should represent the human user of the test AI agent, to explore how the test AI agent may drive human behaviors of concern such as encouraging delusions or eroding military discipline. This is where influence, as opposed to market behavior alone, is actually measured, and it is the most distinctive requirement in the topic.
What is the error function?
Phase II requires proposers to define an error function that describes differences between AI agent decisions and expected human results. The Month 2 Phase II milestone requires definitions of that error function plus measures for comparing distributions of human and AI market outcomes.
What does Phase II deliver to the Government?
At Month 21, final software delivery, both object and source code, for operation by DARPA or other U.S. Government personnel for additional demonstrations, with suitable documentation in a contractor proposed format. That is a source code delivery, and it has direct implications for your intellectual property strategy.
Why does the final report have to discuss LLM advances?
The Month 24 milestone requires the final report to discuss advances in LLM models and AI agents that have occurred during the period of performance and their potential impact on the evaluation of AI models. DARPA is acknowledging that the measurement target moves underneath you across 24 months and asking whether your instrument survives that. Addressing model drift and instrument durability proactively answers a question DARPA has already flagged.
How long can my technical volume be?
For Phase I, a 10 page white paper plus a 5 page slide deck. For Direct to Phase II, the Standard Proposal Format at 35 pages, structured as up to 10 pages of feasibility documentation, up to 20 pages of technical proposal, and up to 5 pages of commercialization strategy as the last section.
Can I ask questions through DSIP Topic Q&A?
No. DARPA states DSIP Topic Q&A will not be available for these topics. Technical questions go to SBIR_BAA@darpa.mil with the topic number in the subject line by October 14, 2026. Questions submitted within seven calendar days of the due date may not be answered. DARPA posts a consolidated FAQ, updated until one week before the due date.
Do I have to choose a contract type?
Yes. Proposers must state their requested contract type. DARPA may award FAR-based firm-fixed-price or cost-plus reimbursement contracts, or Other Transactions for Prototype under 10 U.S.C. 4021. Cost-plus requires your DCMA Final Determination Letter. An OT requires a completed Model OT plus OT Certifications in Volume 5. Firm-fixed-price requires no additional action.
Is the cost template mandatory?
Yes. Templates for Volume 2 and Volume 3 are on the DARPA Small Business website, and use of the DARPA Cost Proposal template is mandatory.
How much TABA can I request?
Phase I awardees up to $6,500. Phase II awardees up to $25,000 per Phase II project. TABA is in addition to the cost ceilings and is not subject to profit or fee, and requests are reviewed by the contracting office at time of award.
Are venture capital backed companies eligible?
Yes. Proposers more than 50 percent owned by multiple venture capital operating companies, hedge funds, private equity firms, or any combination as set forth in 13 CFR 121.702 are eligible, subject to registering with the SBA Company Registry Database before submitting, submitting the Majority-Owned VCOC, HF, and PEF Certification in Volume 5, and notifying the Contracting Officer if you enter that class after submitting but before award.
How will my proposal be evaluated?
Against the evaluation criteria in the DoW SBIR Program BAA. DARPA evaluates each conforming proposal in its entirety, documenting strengths and weaknesses relative to each criterion, then determines overall selectability. Proposals are not evaluated against each other. A selectable proposal is one where strengths outweigh weaknesses with no accumulated weaknesses requiring extensive negotiations or resubmission.
Will I get feedback if not selected?
Yes. DARPA will provide a technical evaluation narrative to the proposer for each proposal submitted in response to a topic, per the SBA SBIR/STTR Policy Directive. An informal feedback session may additionally be requested via sbir@darpa.mil, at DARPA's sole discretion.
What is the commercial market?
DARPA states it directly. The potential risk of AI systems inducing harmful behaviors is a major commercial concern due to associated legal and reputational risk to developers. AI firms are actively seeking mechanisms to argue the limits of their systems' biases and impacts on user behavior, including measuring deviations from expected human norms, in response to lawsuits. DARPA argues the stock-value impact of establishing the scope of liability far exceeds the immediate financial consequences of the court cases, suggesting strong market demand from AI firms for influence benchmarking. Insurers, regulators, and enterprise vendor due diligence extend naturally from that.
Should I build on Magentic Marketplace?
The topic names it as an emerging open-source tool that has shown promise in revealing behavioral variations in market-focused contexts, which is as close as DARPA comes to a recommendation. Understand it and state your relationship to it explicitly, whether you extend it, differentiate from it, or replace it.
Who do I contact with questions?
The DARPA Small Business Programs Office at SBIR_BAA@darpa.mil for both program administration and topic technical questions, with the topic number in the subject line. DSIP technical support at DoDSBIRSupport@reisystems.com with a copy to SBIR_BAA@darpa.mil, Monday through Friday 9:00 a.m. to 5:00 p.m. ET. DARPA also offers free resources through DARPAConnect at DARPAConnect.us.
Positioning Advice for Companies Considering This Topic
Bring the economics, not just the machine learning. The differentiating expertise on this topic is auction theory and experimental economics, not LLM engineering. Two of the eight references are auction and bidding-behavior papers, and the entire premise rests on having a credible human baseline to compare agents against. A team of ML engineers without an economist will struggle to make the comparison meaningful, and it will show.
Design the market before you design the agents. The Month 2 milestone is market structures, payout schemes, efficiency measures, consumer utility, and the news feed. Get that right and the behavioral measurement follows. Get it wrong and you have a chatbot arena with prices attached.
Hit the 90 percent allocative efficiency target credibly. It is the only hard number and it is a validity gate. Show why your market design clears efficiently with rational agents, because if it does not, no inference about AI agent bias from it will be trusted.
Respect the black-box constraint in the architecture. Query-and-output only is not a limitation to work around, it is the design premise that makes the instrument universal. If any part of your approach quietly assumes logprobs, activations, or fine-tuning access, it breaks the topic.
Take the simulated human user seriously. This is the requirement that makes it an influence benchmark rather than a market benchmark. Explain how you construct a human-behavior stock agent grounded in the auction literature, how you validate that it behaves like a human, and how you detect the test AI influencing it toward delusion or indiscipline. That last outcome is unusual and specific, and DARPA named it deliberately.
Budget the inference bill honestly. Ten or more LLMs, many agents, repeated market runs, across 24 months, is a real cost. Understating it reads as inexperience with the scale of the experiment.
Plan for model drift from the start. The Month 24 report must address LLM advances during the period of performance and their impact on the evaluation approach. An instrument that only measures the 2026 model generation has limited value. Build and describe versioning, re-baselining, and comparability across model generations.
Resolve your intellectual property position before you write the cost volume. The Month 21 deliverable is object and source code for Government operation. That interacts with what you build on, whether you can use restrictively licensed components, and what you can commercialize afterward. Data Rights Assertions are a Volume 5 item and this is the topic to use them thoughtfully.
Lead the commercialization strategy with DARPA's liability thesis. You have a stated market argument from the customer: AI firms need defensible influence measurements because of litigation exposure, and the stock-value stakes exceed the case costs. That is a stronger opening than any market sizing you would construct, and you can extend it to insurers and regulators from there.
Engage OpenAgentSafety explicitly. It is cited as the class of approach this topic improves on. Explaining why revealed-preference measurement under incentives beats a priori harm checklists, in your own words and with your own market design as the example, is the clearest way to show you understand what DARPA is buying.
DARPA SBIR DPA26BZ06-DV025: Noninvasive Detection and Localization of Occult Hemorrhage
Deadline: October 21, 2026
Funding Award Size: $1m
Description: Complete guide to DARPA SBIR topic DPA26BZ06-DV025, noninvasive occult hemorrhage detection for combat medics. Phase I $250K or DP2 $1M plus option. Closes October 21, 2026.
Quick Answer
DPA26BZ06-DV025 is a DARPA SBIR topic under the DoW 2026 SBIR Broad Agency Announcement, Release 6, and it is one of only two topics in this release that accepts both Phase I and Direct to Phase II proposals. DARPA wants a capability that lets a combat medic with no surgical or imaging expertise find internal bleeding in the field: noninvasive or minimally invasive detection, anatomic localization, and rate estimation of occult non-compressible torso hemorrhage, outside a hospital. Phase I is $250,000 over 6 months. Direct to Phase II is $1,000,000 over 12 months with a $500,000 option over a further 12 months. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The clinical problem is the leading cause of preventable death on the battlefield, and DARPA's two cited references are the Eastridge papers that established it. What makes this topic distinctive is that it does not stop at detection. The system must distinguish slow or self-limiting hemorrhage from rapidly expanding, exsanguinating hemorrhage, and it must support serial or continuous monitoring so changes in bleeding rate can be tracked and reported to the medic in near-real time. That is a triage and evacuation prioritization capability, not just a diagnostic.
Topic At a Glance
Topic number: DPA26BZ06-DV025
Title: Noninvasive Detection and Localization of Occult Hemorrhage
Agency: Defense Advanced Research Projects Agency (DARPA)
Solicitation: DoW 2026 Small Business Innovation Research Broad Agency Announcement, Release 6, DARPA Proposal Submission Instructions
Program types accepted: both Phase I and Direct to Phase II
Phase I award: $250,000 over 6 months. Technical volume is a 10 page white paper plus a 5 page slide deck
Direct to Phase II award: $1,000,000 over 12 months, with an option of $500,000 over 12 months
Direct to Phase II technical volume format: White Paper and Slide Deck, 20 page white paper plus 15 slide deck
OUSD (R&E) Critical Technology Area: Applied Artificial Intelligence
Component Technology Priority Area: Biotechnology
Projected CMMC level requirement: Level 1
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic
Technical and Business Assistance: up to $6,500 for Phase I awardees, up to $25,000 per Phase II project, in addition to the cost ceilings
Topic Q&A: DSIP Topic Q&A is not available for DARPA topics. Technical questions go to SBIR_BAA@darpa.mil by October 14, 2026
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Submission portal: DSIP at dodsbirsttr.mil
Keywords: hemorrhage detection, clinical decision support, medical device, austere environment
Two Entry Points, and How to Choose
This topic appears in two of DARPA's award structure tables, which means you have a genuine choice of entry point. Getting it right matters, because the formats and the evidence bars are different.
The Phase I path. $250,000 over 6 months. Technical volume is a 10 page white paper plus a 5 page slide deck. Phase I proposals shall present tasking and relevant milestones for a 6-month base period. This path is for a team with a sensing concept and supporting science but without an integrated system that already detects and localizes internal bleeding.
The Direct to Phase II path. $1,000,000 over 12 months plus a $500,000 option over 12 months. Technical volume is a 20 page white paper plus a 15 slide deck, within an overall 35 page limit that DARPA describes for the White Paper and Slide Deck format. This path requires documentation demonstrating you have already achieved the feasibility milestones expected at the conclusion of a Phase I effort, and DARPA lists four specific ones.
The choice is not strategic preference, it is evidence. If you can document the four DP2 feasibility elements below, take the DP2 path, because it is four times the money and it skips a phase. If you cannot document all four, a DP2 proposal will be found non-conforming and Phase I is the honest route.
Note that Phase I proposals shall present tasking and relevant milestones for a 6-month base period, while Phase II proposals shall present tasking and relevant milestones for a 12-month base period, and separate tasking and milestones for a subsequent single 12-month option period. Structure your schedule accordingly.
What DARPA Is Actually Looking For
The objective
Demonstrate a capability that noninvasively detects and localizes occult, meaning internal, non-compressible hemorrhage outside of a hospital.
The user, named precisely
The goal is to produce a noninvasive or minimally invasive capability that enables Role 1 medical personnel, specifically a combat medic with no surgical or pre-operative imaging expertise, to rapidly detect, localize, and characterize the rate of life-threatening, occult, or non-compressible torso hemorrhage that cannot be precisely identified through routine clinical assessments available in the field.
That parenthetical is the design constraint that matters most. Not a radiologist. Not a surgeon. A combat medic with no imaging expertise. Every interface decision, every acquisition-guidance decision, and every output format decision follows from it. A device that requires skilled sonographic window-finding does not meet this requirement no matter how good its physics.
What the system must do
The proposed solution should integrate advanced sensing technologies with artificial intelligence to provide accurate, real-time detection, anatomic localization, and quantitative or semi-quantitative estimation of bleeding rate, distinguishing slow or self-limiting hemorrhage from rapidly expanding, exsanguinating hemorrhage, and should function in austere and contested environments.
The capability should support serial or continuous monitoring such that changes in hemorrhage rate can be tracked to provide updates to the medic in near-real time.
Solutions should provide intuitive, actionable visualization of both hemorrhage location and rate to support timely medical decision-making, triage, and evacuation prioritization when definitive surgical care is delayed or denied.
Four distinct capabilities sit in those paragraphs: detection, anatomic localization, rate estimation, and change tracking over time. Rate estimation is the hardest and the most valuable, because it is what separates the casualty who needs the next helicopter from the one who can wait.
Phase I in detail
Demonstrate the technical feasibility of a noninvasive or minimally invasive capability for rapid detection, anatomical localization, and characterization of life-threatening occult or non-compressible hemorrhage. The performer shall develop a proof-of-concept sensing and artificial intelligence framework capable of accurately identifying the anatomical source of bleeding and estimating hemorrhage rate for a limited set of representative injury types.
Scanning may be performed manually or using an automated platform, for example robotic or assisted acquisition, and should provide clinically actionable results within a timeframe compatible with Role 1 medical decision-making.
Note the permission for a limited set of representative injury types in Phase I. You do not have to cover the whole torso. Choosing a defensible subset and justifying the choice is better than claiming breadth you cannot demonstrate.
Proposals shall describe the envisioned concept of operations, including sensor placement or access method for minimally invasive approaches, data acquisition, user workflow, system-guided adjustments to sensor positioning or data collection, data processing, reporting, and, where appropriate, serial or continuous monitoring.
That list is a checklist. Nine elements, and system-guided adjustments to sensor positioning is the one that speaks directly to the untrained-user constraint. If your device tells the medic where to move the probe, say how.
Proposals shall describe the sensing technologies to be employed, their anticipated technology maturity, the proposed AI algorithms and training strategy, available datasets, and any additional data collection planned during Phase I.
Representative testbeds may include retrospective clinical datasets, computational models, physical phantoms, ex vivo preparations, or controlled preclinical studies.
That is a permissive testbed list and it matters for schedule. Six months does not accommodate a new animal study from scratch, but retrospective clinical data, phantoms, and computational models are all explicitly acceptable.
Phase I efforts should establish the technical capabilities and limitations of the proposed approach, including the minimum and maximum resolvable hemorrhage rate, localization accuracy, anatomical applicability including challenging locations such as retroperitoneal hemorrhage or bleeding obscured by bone or bowel, and expected robustness across representative operational conditions.
Retroperitoneal hemorrhage and bleeding obscured by bone or bowel are named because they are where most sensing modalities fail. Address them directly, including where your approach cannot reach, because DARPA asked for limitations as well as capabilities.
Proposals shall define quantitative performance metrics for hemorrhage detection, localization, bleeding-rate estimation, response time, and operational robustness under austere conditions, including environmental stressors, limited power availability, transportation, storage, and deployment.
While field-ready systems are not required, proposals utilizing components with demonstrated field suitability will be viewed favorably.
The Phase I effort should culminate in a proof-of-concept demonstration and a detailed technical development plan supporting maturation into an integrated prototype during Phase II.
Phase II in detail
Phase II will build upon the successful Phase I demonstration by developing, integrating, and validating a portable prototype capable of real-time hemorrhage detection, anatomical localization, bleeding-rate estimation, and longitudinal monitoring across a broader range of clinically relevant battlefield injuries.
Then a sentence that constrains the work in a useful way: Phase II should emphasize engineering maturation, system integration, and operational validation rather than fundamental sensor development. The sensing hardware architecture established during Phase I should remain substantially unchanged, with improvements focused on optimization, robustness, algorithm refinement, and user-centered operation.
If you enter at DP2, that tells you DARPA expects your sensing architecture to be settled. Proposing to explore modalities in Phase II is a mismatch with the topic.
Performers shall expand system performance to encompass additional solid organs, vascular structures, and clinically relevant hemorrhage scenarios while collecting additional datasets to improve algorithm performance, robustness, and generalizability.
The system shall support serial or continuous assessment of hemorrhage progression, enabling near-real-time updates of bleeding location and rate.
The integrated prototype shall provide intuitive clinical decision support suitable for Role 1 medical personnel, including actionable visualization of hemorrhage location, bleeding rate, confidence estimates, injury implications, and recommendations supporting triage, treatment prioritization, and evacuation decisions.
Note confidence estimates in that list. A rate estimate without an uncertainty bound is not usable for evacuation triage, and DARPA has asked for it explicitly.
Validation should be conducted using representative preclinical, cadaveric, clinical, or other operationally relevant test environments, with performance evaluated under conditions representative of austere military operations, including motion, environmental stress, limited logistical support, and communications degradation where appropriate.
Proposals shall define quantitative Phase II performance metrics for detection accuracy, localization precision, bleeding-rate estimation, response time, longitudinal tracking performance, usability, and operational suitability.
Phase II should conclude with delivery of an integrated prototype, validation results demonstrating operational feasibility, and a transition strategy supporting military evaluation, regulatory planning, manufacturing, and commercialization.
As with Phase I, proposals shall define envisioned Phase II metrics for success in the detection, localization, and characterization of bleeding. Phase II proposals should enumerate the planned list of injury targets planned for characterization.
The Direct to Phase II Feasibility Requirement
DP2 proposers must provide documentation demonstrating they have already achieved the feasibility milestones expected at the conclusion of a Phase I effort. Specifically, proposals must demonstrate existing evidence of four things.
Proof-of-Concept: an established sensing and AI framework capable of accurately identifying the anatomical source of occult or internal bleeding.
Hemorrhage Rate Estimation: the ability to quantitatively or semi-quantitatively estimate hemorrhage rates to distinguish slow bleeding from rapidly expanding hemorrhage across various injury profiles.
Performance Metrics: tested parameters detailing the minimum and maximum resolvable hemorrhage rates, localization accuracy, and anatomical applicability, including challenging areas such as retroperitoneal hemorrhage.
Operational Robustness: preliminary data indicating the approach can withstand austere conditions, including environmental stressors and limited power availability.
All four are evidentiary, not aspirational. The second one, rate estimation, is the element most likely to be missing from an otherwise strong team, because plenty of groups can detect free fluid or hemorrhage while far fewer can put a rate on it. If you cannot show rate estimation data, the Phase I path is the correct entry.
Per the DP2 authority, if a proposer can provide adequate documentation to substantiate that the scientific and technical merit and feasibility described in the Phase I section of the topic has been met, and describes the potential commercial applications, the Direct to Phase II authority allows the Department of War to make an award under Phase II of the SBIR program without regard to whether the small business concern was provided an award under Phase I of an SBIR program.
For the White Paper and Slide Deck DP2 format, the white paper shall not exceed 20 pages and the slide deck shall not exceed 15 pages, within the 35 page overall format DARPA describes. Refer to Appendix B, DARPA Direct to Phase II Instructions, for the content of each element of the Technical Volume and the commercialization strategy.
Phase III Dual Use
DARPA splits this into two applications, both concrete.
Military application, battlefield. The developed noninvasive hemorrhage detection system transitions directly to military operational medicine, specifically empowering Role 1 combat medics. In austere and contested environments where definitive surgical care or medical evacuation is delayed or denied, this portable technology provides critical clinical decision support. By continuously monitoring non-compressible torso hemorrhage progression, medics can perform highly accurate triage, prioritize life-saving interventions, and optimize evacuation scheduling for casualties with exsanguinating hemorrhage versus those with self-limiting bleeding.
Civilian and commercial application, remote disaster areas. In the civilian sector, this technology holds significant dual-use potential for emergency medical services, search-and-rescue teams, and disaster response units. During mass casualty incidents or natural disasters such as earthquakes or severe storms where infrastructure is damaged and access to hospital imaging is impossible, first responders can use this tool to rapidly detect occult internal bleeding at the point of injury. It provides paramedics with real-time, actionable data to prioritize helicopter transport or specialized trauma center routing for critically internally injured patients in remote or rural environments, ultimately reducing preventable mortality outside of the hospital setting.
Both framings converge on the same commercial proposition: pre-hospital triage of internal bleeding. Ground and air EMS, rural hospitals without imaging, and disaster response organizations are the buyers, and the regulatory pathway is a medical device clearance rather than a defense qualification. Your commercialization strategy should address FDA strategy explicitly, since Phase II asks for a transition strategy supporting regulatory planning.
Funding, Cost Structure, and DARPA Mechanics
The awards
Phase I: $250,000 over 6 months. Technical volume is a 10 page white paper plus a 5 page slide deck.
Direct to Phase II: $1,000,000 over 12 months, plus an option of $500,000 over 12 months, for $1,500,000 across 24 months if the option is exercised.
The resources made available for each topic will depend on the quality of the proposals received and the availability of funds. The Government reserves the right to award all, some, one, or none of the options based on available funding and the performer's technical performance.
Contract type, which you must elect
DARPA may award FAR-based contracts, firm-fixed-price or cost-plus reimbursement, or Other Transactions for Prototype under the authority of 10 U.S.C. 4021, subject to approval of the Contracting Officer or Agreements Officer respectively. Proposers must state their requested contract type in their proposal.
Cost-plus reimbursement requires including your Defense Contract Management Agency Final Determination Letter showing approval of your accounting system. An Other Transaction for Prototype requires including a completed OT using the Model OT for Prototype from the DARPA Small Business site, plus completed OT Certifications, both loaded in Volume 5, with at minimum the color-coded areas completed and redlines with explanations for any article you wish to negotiate. Firm-fixed-price requires no additional action.
Templates are mandatory
Templates for Volume 2 Technical Volume and Volume 3 Cost Volume are provided as attachments on the DARPA Small Business website. Use of the DARPA Cost Proposal template is mandatory.
Technical and Business Assistance
Phase I awardees may request up to $6,500 in TABA funding. Phase II awardees may request up to $25,000 per Phase II project. TABA funding is in addition to the Phase I and Phase II cost ceilings and is not subject to profit or fee. TABA funding requests will be reviewed by the respective contracting office or specialist at time of award to ensure compliance with TABA requirements.
For a medical device topic, regulatory strategy support and market validation with EMS and trauma system buyers are both plausible uses.
Questions and the FAQ
DSIP Topic Q&A will not be available for these DARPA topics. Technical questions related to improving the understanding of a topic's requirements must be submitted by October 14, 2026, by email to SBIR_BAA@darpa.mil with the topic number in the subject line, including the name, email address, and telephone number of a point of contact. Questions submitted within seven calendar days of the proposal due date may not be answered. DARPA posts a consolidated Frequently Asked Questions document under the topic number summary on its Small Business site, updated on an ongoing basis until one week prior to the proposal due date.
DARPA will not accept late proposals.
Classification, marking, and registrations
All proposals are required to be UNCLASSIFIED or CUI. Do not include any classified information in your proposal submission. Do not include any proprietary information on the Proposal Coversheet in Volume 1.
Proposal titles, abstracts, anticipated benefits, and keywords of proposals selected for contract award will undergo a DARPA Policy and Security Review and are subject to revision or redaction by DARPA. Final approved versions may appear on the DoW SBIR/STTR awards website and the SBA's award website at sbir.gov/awards.
Proposers should ensure they have an accurate and active entity registration on SAM.gov. Those engaging in ITAR or CUI work for DARPA must have CMMC Level 2 certification, though the projected requirement for this topic is Level 1. DARPA points to sprs.csd.disa.mil/nistsp.htm and notes Project Spectrum at projectspectrum.io as an assistance resource.
Venture capital, hedge fund, and private equity ownership
Proposers that are more than 50 percent owned by multiple venture capital operating companies, hedge funds, private equity firms, or any combination of these as set forth in 13 CFR 121.702 are eligible to submit proposals in response to DARPA topics advertised within this BAA. Three conditions apply: register with the SBA Company Registry Database before submitting; submit the Majority-Owned VCOC, HF, and PEF Certification, with the SBIR VC Certification available on the DARPA Small Business site, in Supporting Documents Volume 5; and immediately notify the Contracting Officer, register in the appropriate SBA database, and submit the required certification if you enter that ownership class after submitting but before receiving a funding agreement.
DARPA's permissive posture matters here, since medical device companies at this stage are commonly venture-funded.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW SBIR Program BAA. DARPA will conduct an evaluation of each conforming proposal. Proposals that do not comply with the requirements detailed in this BAA and the research objectives of the corresponding topic are considered non-conforming and are therefore not evaluated nor considered for award.
Using the evaluation criteria, the Government will evaluate each proposal in its entirety, documenting the strengths and weaknesses relative to each evaluation criteria, and based on those identified strengths and weaknesses will determine the proposal's overall selectability for funding. Proposals will not be evaluated against each other during the evaluation process but rather evaluated on their own individual merit.
A selectable proposal is one where the strengths of the overall proposal outweigh its weaknesses, with no accumulated weaknesses that would require extensive negotiations or a resubmitted proposal. A non-selectable proposal is one where the strengths do not outweigh its weaknesses.
Proposing firms will be notified of selection or non-selection status for a Phase I or Direct to Phase II award within 90 calendar days of the closing date of the BAA. The Corporate Official indicated on the Proposal Cover Sheet will be notified by email. In accordance with the SBA SBIR/STTR Policy Directive, Appendix I, paragraph 4, subparagraph (d), DARPA will provide a technical evaluation narrative to the proposer for each proposal submitted in response to a topic. An informal feedback session may additionally be requested via email at sbir@darpa.mil, provided at the sole discretion of DARPA.
Company Commercialization Report information will not be considered by DARPA during proposal evaluations.
Protests regarding the selection decision should be submitted, as prescribed in FAR 33.106(b) and FAR 52.233-3, to DARPA Contracts Management Office, 675 N. Randolph Street, Arlington, VA 22203, by email to CMO_SBIRProtests@darpa.mil and sbir@darpa.mil.
Post-award support
DARPA provides Transition and Commercialization Support Program services to Phase II and DP2 awardees upon contract execution at no cost to awardees, with the goal of maximizing the potential to move technology beyond Phase II into other research and development programs, DoW acquisition programs, other federal programs, or the commercial market. Awardees may also be eligible for the Embedded Entrepreneurship Initiative, an invitation-only program at DARPA's sole discretion, typically no more than $310,000 per awardee over the duration of the award, supporting a Senior Commercialization Advisor relationship, investor working group connections, and hiring an embedded entrepreneur to execute a Go-to-Market strategy. For a medical device seeking regulatory and market entry, both programs have real value.
The References
Two, and both are foundational rather than technical.
Eastridge BJ, Mabry RL, Seguin P, Cantrell J, Tops T, Uribe P, et al. Death on the battlefield (2001-2011): Implications for the future of combat casualty care. Journal of Trauma and Acute Care Surgery. 2012 Dec;73(6):S431-7.
Eastridge BJ, Hardin M, Cantrell J, Oetjen-Gerdes L, Zubko T, Mallak C, Wade CE, Simmons J, Mace J, Mabry R, Bolenbaucher R, Blackbourne LH. Died of wounds on the battlefield: causation and implications for improving combat casualty care. J Trauma. 2011 Jul;71(1 Suppl):S4-8.
These are the papers that quantified preventable death from hemorrhage in Iraq and Afghanistan and established non-compressible torso hemorrhage as the dominant category. DARPA cites no sensing technology references at all, which tells you the agency is not steering you toward a modality. It is stating a clinical problem and leaving the physics open. That is unusual and it is an opportunity: ultrasound, photoacoustics, electrical impedance, microwave, near-infrared, and hybrid approaches are all on the table, and your argument for a modality is part of what gets evaluated.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
Technical question deadline: October 14, 2026, to SBIR_BAA@darpa.mil with the topic number in the subject line
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Selection notification: within 90 calendar days of BAA close
Phase I period: 6 months
Direct to Phase II base: 12 months. Option: a further 12 months
A working backward plan
Before September 23. Decide your entry point by testing yourself honestly against the four DP2 feasibility elements, especially hemorrhage rate estimation. Download the mandatory DARPA Volume 2 and Volume 3 templates. Read the FAQ and keep rechecking it. Assemble your evidence base: retrospective clinical datasets, phantom results, ex vivo or preclinical data, and any field-suitability data on components. Settle your sensing modality and be ready to defend it, since DARPA cites no modality references. Work out your position on retroperitoneal hemorrhage and bleeding obscured by bone or bowel, including where you cannot reach. Decide your contract type and prepare the corresponding documents. Confirm SAM registration. If venture-backed, register with the SBA Company Registry and obtain the SBIR VC Certification. Submit technical questions before October 14.
September 23 through October 5. Draft to the format for your chosen path. For Phase I, a 10 page white paper plus 5 slides with 6 month tasking and milestones. For DP2, a 20 page white paper plus 15 slides with 12 month base tasking and separate 12 month option tasking. In either case, walk the nine-element concept of operations list, define the quantitative performance metrics DARPA names, and state limitations as well as capabilities.
October 6 through October 14. Build the cost volume in the mandatory template. Price your validation environment realistically, whether phantom, ex vivo, cadaveric, preclinical, or retrospective clinical data access, since data access agreements and IRB or IACUC timelines can dominate a 6 or 12 month schedule.
October 15 through October 18. Assemble Volume 5 with contract-type documents and certifications, complete Volume 7 and the Volume 4 CCR, and run compliance: page and slide limits for your format, unclassified or CUI only, no proprietary information on the coversheet, mandatory cost template.
October 19 through October 20. Submit and certify in DSIP.
Frequently Asked Questions
What is DARPA SBIR topic DPA26BZ06-DV025?
DPA26BZ06-DV025 is a DARPA SBIR topic titled "Noninvasive Detection and Localization of Occult Hemorrhage," released under the DoW 2026 SBIR Broad Agency Announcement, Release 6. The objective is to demonstrate a capability that noninvasively detects and localizes occult, meaning internal, non-compressible hemorrhage outside of a hospital.
Can I submit either a Phase I or a Direct to Phase II proposal?
Yes. This is one of only two topics in this DARPA release that appears in both the Phase I and the Direct to Phase II award structure tables. Phase I is $250,000 over 6 months with a 10 page white paper and 5 page slide deck. Direct to Phase II is $1,000,000 over 12 months plus a $500,000 option over 12 months, with a 20 page white paper and 15 slide deck.
How do I decide which path to take?
By evidence, not preference. The DP2 path requires documentation demonstrating four specific existing capabilities: an established sensing and AI proof-of-concept identifying the anatomical source of internal bleeding, quantitative or semi-quantitative hemorrhage rate estimation across injury profiles, tested performance parameters covering minimum and maximum resolvable rates plus localization accuracy and anatomical applicability including retroperitoneal hemorrhage, and preliminary operational robustness data under austere conditions. If you cannot document all four, a DP2 proposal risks being found non-conforming and Phase I is the correct route.
Who is the intended user?
Role 1 medical personnel, specified as a combat medic with no surgical or pre-operative imaging expertise. That constraint drives the entire interface and acquisition design. A device requiring skilled imaging technique does not satisfy this topic.
What must the system actually do?
Four things. Detect occult non-compressible torso hemorrhage. Localize it anatomically. Estimate bleeding rate quantitatively or semi-quantitatively, distinguishing slow or self-limiting hemorrhage from rapidly expanding exsanguinating hemorrhage. And support serial or continuous monitoring so changes in rate can be tracked and reported to the medic in near-real time. It must also provide intuitive, actionable visualization of location and rate to support triage and evacuation prioritization.
What sensing modality does DARPA want?
The topic does not say, and notably cites no sensing technology references at all, only the two Eastridge clinical papers. Proposals shall describe the sensing technologies to be employed, their anticipated technology maturity, the proposed AI algorithms and training strategy, available datasets, and any additional data collection planned. The modality choice and its justification are yours to make and defend.
What anatomical challenges must I address?
The topic names challenging locations explicitly: retroperitoneal hemorrhage and bleeding obscured by bone or bowel. Phase I efforts should establish anatomical applicability including these, and the DP2 feasibility requirement names retroperitoneal hemorrhage specifically. State where your approach works and where it does not.
What testbeds are acceptable in Phase I?
Representative testbeds may include retrospective clinical datasets, computational models, physical phantoms, ex vivo preparations, or controlled preclinical studies. That is a permissive list, which matters because a 6 month schedule will not accommodate a new animal study from scratch.
Do I need a field-ready system?
No. While field-ready systems are not required, proposals utilizing components with demonstrated field suitability will be viewed favorably.
What must my proposal define quantitatively?
Performance metrics for hemorrhage detection, localization, bleeding-rate estimation, response time, and operational robustness under austere conditions, the latter including environmental stressors, limited power availability, transportation, storage, and deployment. Phase II adds longitudinal tracking performance, usability, and operational suitability.
What concept of operations detail is required?
Proposals shall describe the envisioned concept of operations including sensor placement or access method for minimally invasive approaches, data acquisition, user workflow, system-guided adjustments to sensor positioning or data collection, data processing, reporting, and where appropriate serial or continuous monitoring. The system-guided adjustment element is the one that speaks to the untrained-user constraint.
Can I change my sensing hardware in Phase II?
Not substantially. Phase II should emphasize engineering maturation, system integration, and operational validation rather than fundamental sensor development, and the sensing hardware architecture established during Phase I should remain substantially unchanged, with improvements focused on optimization, robustness, algorithm refinement, and user-centered operation.
What does Phase II have to deliver?
An integrated portable prototype capable of real-time detection, anatomical localization, bleeding-rate estimation, and longitudinal monitoring across a broader range of clinically relevant battlefield injuries, expanded to encompass additional solid organs, vascular structures, and hemorrhage scenarios. Plus validation results demonstrating operational feasibility and a transition strategy supporting military evaluation, regulatory planning, manufacturing, and commercialization. The prototype must provide clinical decision support including actionable visualization of hemorrhage location, bleeding rate, confidence estimates, injury implications, and recommendations supporting triage, treatment prioritization, and evacuation decisions.
How should Phase II validation be conducted?
Using representative preclinical, cadaveric, clinical, or other operationally relevant test environments, with performance evaluated under conditions representative of austere military operations including motion, environmental stress, limited logistical support, and communications degradation where appropriate.
Are confidence estimates required?
Yes, in Phase II. The integrated prototype's clinical decision support must include confidence estimates alongside hemorrhage location, bleeding rate, injury implications, and recommendations. A rate estimate without an uncertainty bound is not usable for evacuation triage.
How long can my technical volume be?
For Phase I, a 10 page white paper plus a 5 page slide deck. For Direct to Phase II, a 20 page white paper plus a 15 page slide deck, within the 35 page White Paper and Slide Deck format DARPA describes. Refer to Appendix A for Phase I content and Appendix B for DP2 content.
What milestone structure should I propose?
Phase I proposals shall present tasking and relevant milestones for a 6-month base period. Phase II proposals shall present tasking and relevant milestones for a 12-month base period, and separate tasking and milestones for a subsequent single 12-month option period.
Can I ask questions through DSIP Topic Q&A?
No. DARPA states DSIP Topic Q&A will not be available for these topics. Technical questions go to SBIR_BAA@darpa.mil with the topic number in the subject line by October 14, 2026. Questions submitted within seven calendar days of the due date may not be answered. DARPA posts a consolidated FAQ, updated until one week before the due date.
Do I have to choose a contract type?
Yes. Proposers must state their requested contract type. DARPA may award FAR-based firm-fixed-price or cost-plus reimbursement contracts, or Other Transactions for Prototype under 10 U.S.C. 4021. Cost-plus requires your DCMA Final Determination Letter showing accounting system approval. An OT requires a completed Model OT plus OT Certifications in Volume 5. Firm-fixed-price requires no additional action.
Is the cost template mandatory?
Yes. Templates for Volume 2 and Volume 3 are on the DARPA Small Business website, and use of the DARPA Cost Proposal template is mandatory.
How much TABA can I request?
Phase I awardees may request up to $6,500. Phase II awardees may request up to $25,000 per Phase II project. TABA funding is in addition to the cost ceilings and is not subject to profit or fee, and requests are reviewed by the contracting office at time of award.
Are venture capital backed companies eligible?
Yes. Proposers more than 50 percent owned by multiple venture capital operating companies, hedge funds, private equity firms, or any combination as set forth in 13 CFR 121.702 are eligible, subject to registering with the SBA Company Registry Database before submitting, submitting the Majority-Owned VCOC, HF, and PEF Certification in Volume 5, and notifying the Contracting Officer if you enter that class after submitting but before award.
How will my proposal be evaluated?
Against the evaluation criteria in the DoW SBIR Program BAA. DARPA evaluates each conforming proposal in its entirety, documenting strengths and weaknesses relative to each criterion, then determines overall selectability. Proposals are not evaluated against each other. A selectable proposal is one where strengths outweigh weaknesses with no accumulated weaknesses requiring extensive negotiations or resubmission.
Will I get feedback if not selected?
Yes. DARPA will provide a technical evaluation narrative to the proposer for each proposal submitted in response to a topic, per the SBA SBIR/STTR Policy Directive. An informal feedback session may additionally be requested via sbir@darpa.mil, at DARPA's sole discretion.
What is the commercial market?
Pre-hospital triage of internal bleeding. The topic names emergency medical services, search-and-rescue teams, and disaster response units, and describes mass casualty incidents and natural disasters where infrastructure is damaged and hospital imaging is unavailable. First responders would use the tool to detect occult internal bleeding at the point of injury and prioritize helicopter transport or trauma center routing. Ground and air EMS, rural hospitals without imaging, and disaster response organizations are the buyers.
Who do I contact with questions?
The DARPA Small Business Programs Office at SBIR_BAA@darpa.mil for both program administration and topic technical questions, with the topic number in the subject line. DSIP technical support at DoDSBIRSupport@reisystems.com with a copy to SBIR_BAA@darpa.mil, Monday through Friday 9:00 a.m. to 5:00 p.m. ET. DARPA also offers free resources through DARPAConnect at DARPAConnect.us.
Positioning Advice for Companies Considering This Topic
Choose your entry point on evidence. The four DP2 feasibility elements are checkable, and rate estimation is the one that separates teams. Detecting free fluid is well-trodden ground; quantifying a bleeding rate is not. If you have rate data, take the DP2 path and the four-times funding. If you do not, Phase I is not a lesser choice, it is the correct one.
Design for the medic, and say so on page one. No surgical or pre-operative imaging expertise is the phrase to internalize. System-guided sensor positioning, automated or assisted acquisition, and an output that reads as a decision rather than an image are all responsive to that constraint. Many strong medical imaging teams will propose excellent physics with an operator model that does not exist at Role 1.
Make rate estimation the technical centerpiece. It is what the whole capability is for. Detection tells you there is bleeding; rate tells the medic whether this casualty goes on the next aircraft. Show your physics for rate, your validation approach, and your uncertainty quantification.
Be candid about the anatomy you cannot see. Retroperitoneal hemorrhage and bleeding obscured by bone or bowel are named because they are hard. DARPA asked for capabilities and limitations. A proposal claiming uniform torso coverage will read as either uninformed or overselling; one that maps its performance envelope honestly reads as expert.
Own the modality argument. With no sensing references cited, DARPA has left the physics open and will judge your reasoning. State why your modality can resolve rate and location in a torso, what its failure modes are, and why alternatives are worse for this use case.
Plan the data problem early. AI algorithms need training data, and hemorrhage rate ground truth is scarce. Retrospective clinical datasets, phantoms with controlled flow, ex vivo preparations, and preclinical models each give you different ground truth quality. Name your sources and your access arrangements, because in a 6 or 12 month schedule, IRB, IACUC, and data use agreements are often the critical path.
Put field suitability in the component choices. Proposals utilizing components with demonstrated field suitability will be viewed favorably, which is a low-cost way to earn credit. If a subsystem already has ruggedized or fielded heritage, say so.
Address the regulatory path in the commercialization section. Phase II asks for a transition strategy supporting regulatory planning, and the civilian market is a cleared medical device market. A realistic FDA pathway view, including predicate strategy or de novo reasoning, distinguishes a company that intends to sell a product from one that intends to publish.
Use both Eastridge papers. They are the only references, they are the canonical statement of the problem, and citing them correctly signals you understand why this topic exists rather than treating it as a generic imaging opportunity.
Write to eliminate weaknesses. DARPA's selectability test turns on whether accumulated weaknesses would require extensive negotiation or resubmission. An unaddressed anatomical limitation, an unexplained data source, or a missing regulatory view are each the kind of gap that pushes a good proposal from selectable to non-selectable.
DARPA SBIR DPA26BZ06-DV024: New World Screwworm Networked Detection, Identification, and Surveillance System
Deadline: October 21, 2026
Funding Award Size: $2.1m
Description: Complete guide to DARPA SBIR Direct to Phase II topic DPA26BZ06-DV024, networked New World Screwworm detection and surveillance. $1.1M base plus $1M options. Closes October 21, 2026.
Quick Answer
DPA26BZ06-DV024 is a DARPA SBIR Direct to Phase II topic under the DoW 2026 SBIR Broad Agency Announcement, Release 6. DARPA wants an autonomous, networked insect trap that detects and identifies New World Screwworm flies on board, without a technician driving out to service it and without shipping samples to a laboratory. The economic case is stated in the topic: re-entry of New World Screwworm into the United States could cost U.S. cattle producers $732 million per year. The award is $1,100,000 over a 12 month base, with two sequential $500,000 options of 6 months each, for a total of $2,100,000 across 24 months. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The current state of the art is deliberately unflattering, and DARPA spells it out. The USDA is using three types of traps at the US Southern Border, sticky fly traps, liver-baited net traps, and gator traps, all of which require manual servicing, fly collection, and fly delivery to a laboratory. Suspected samples go to the National Veterinary Services Laboratories in Ames, Iowa for official confirmation. The lure in use, Swormlure-5, is a synthetic bait on a mix of volatile organic and volatile sulfur compounds that is not specific to New World Screwworm and can lure other species.
So the topic asks you to beat two things at once: a trap that needs a human, and a lure that is not selective. That dual requirement is the shape of the whole program.
Topic At a Glance
Topic number: DPA26BZ06-DV024
Title: New World Screwworm Networked Detection, Identification, and Surveillance System
Agency: Defense Advanced Research Projects Agency (DARPA)
Solicitation: DoW 2026 Small Business Innovation Research Broad Agency Announcement, Release 6, DARPA Proposal Submission Instructions
Program type: Direct to Phase II (DP2). This is a Direct to Phase II topic only
Technical volume format: Standard Proposal Format, 35 pages
Base award: $1,100,000
Base period of performance: 12 months
Option 1: $500,000 over 6 months
Option 2: $500,000 over 6 months
Total if all options exercised: $2,100,000 across 24 months
OUSD (R&E) Critical Technology Area: Applied Artificial Intelligence
Component Technology Priority Area: Biotechnology
Projected CMMC level requirement: Level 1
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic
Required facility access: the performer must have access to an ACL-3 test facility
Phase II demonstration standard: TRL 6 using a prototype system in a relevant environment
Target device size and weight: approximately 15 cm by 15 cm by 15 cm, 600 g
Technical and Business Assistance: up to $25,000 per Phase II project, in addition to the cost ceiling
Topic Q&A: DSIP Topic Q&A is not available for DARPA topics. Technical questions go to SBIR_BAA@darpa.mil by October 14, 2026
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026
Submission portal: DSIP at dodsbirsttr.mil
Keywords: New World Screwworm, NWS, sensor, lure, pest, identification, network, detection, surveillance, monitoring
Read This First: The ACL-3 Facility Requirement
Buried in the Phase II description is a sentence that functions as an eligibility gate: the performer must have access to an ACL-3 test facility.
ACL-3, Arthropod Containment Level 3, is the containment standard for work with arthropods that pose significant risk, and access to a compliant facility is not something you arrange in a month. If you are proposing to work with live New World Screwworm, or with a surrogate species requiring that containment level, you need this access identified and documented before you submit.
Practically, that means either you operate an ACL-3 arthropod containment facility yourself, or you have a partner institution that does and a written arrangement to use it. Very few small businesses hold ACL-3 arthropod containment. University entomology departments, USDA facilities, and a handful of contract research organizations do. Identify yours early and name it in the proposal, because a reviewer who cannot see how you will conduct the work will treat the plan as not credible.
Note also the related constraint: New World Screwworm is a regulated pest not currently established in the United States, so live-organism work carries permitting requirements alongside containment requirements. Address both.
What DARPA Is Actually Looking For
The problem and the money
New World Screwworm, or NWS, scientific name Cochliomyia hominivorax, larvae infest the living tissue of warm-blooded animals. NWS is prevalent in South America and the Caribbean, and it is estimated that re-entry of NWS into the United States could cost US cattle producers $732 million per year, citing USDA 2025.
Screwworm flies lay eggs in open wounds. The eggs hatch into maggots, or larvae, that eat live tissue, causing a worsening, often painful and foul-smelling wound, citing CDC 2025.
Wildlife and free-ranging animals are an important pathway for New World Screwworm to spread, making surveillance beyond livestock essential, citing Arizona 2026.
What is deployed today, and why it is inadequate
The current state of the art for NWS detection and tracking being used at the US Southern Border involves three types of traps: sticky fly traps, liver-baited net traps, and gator traps. All of which require manual servicing, fly collection, and fly delivery to a laboratory.
The state-of-the-art lure to attract NWS flies to a trap is Swormlure-5, a synthetic bait based on a mix of volatile organic and volatile sulfur compounds. However, such baits are not specific to NWS and can lure other species.
The USDA has deployed over 100 NWS-specific traps and lures across high-risk areas of U.S. Border States and is leveraging thousands of fruit fly and insect traps all along the Southern Border. Suspected NWS samples are forwarded for official confirmation to the USDA National Veterinary Services Laboratories in Ames, Iowa, the national reference lab responsible for screwworm diagnostics.
Read that last point carefully. Every suspected detection today travels from the border to Iowa. The latency in that pipeline is the operational problem your device eliminates, and quantifying that latency reduction is a strong argument to make.
The system DARPA wants
A real-time, autonomous, persistent, networked NWS detection, identification, and surveillance system is necessary for accurate and timely tracking and forecasting of the movement of the New World Screwworm.
Such a system must operate reliably in rugged, remote regions lacking power and communications infrastructure by leveraging on-board data processing and machine learning analysis. This can enable on-board insect identification, for example using low-power, localized optical, acoustic, electrical, or chemical sensing, and eliminate the need for manual sample collection and laboratory diagnostics.
Furthermore, these devices must form resilient, low-bandwidth ad-hoc networks to transmit real-time, highly compressed detection telemetry, enabling NWS movement forecasting and biosecurity intervention even in remote and disconnected environments.
The objective, stated at the top of the topic, is to develop and demonstrate a real-time, networked pest detection, identification, and surveillance technology that provides actionable New World Screwworm detection and identification accuracy, system reliability, low maintenance, and long and unattended operation.
Note the five named outcome qualities: detection and identification accuracy, system reliability, low maintenance, long operation, and unattended operation. Those are your performance dimensions.
The Phase II Requirements, Which Are Unusually Specific
DARPA lists what the DP2 effort will develop, integrate, validate, and demonstrate. This list is effectively the specification, and it is worth walking in full because each bullet is a checkable requirement.
System level
A scalable, networked NWS detection, identification, and surveillance system suitable for areal coverage as large as national borders.
Improved lure over Swormlure-5
Greater NWS selectivity. Increased lure attraction range to reduce the number of traps needed. Volatility consistent with a one-month maintenance interval.
Those three are in tension with each other, which is what makes the lure work hard. Greater selectivity usually means a narrower chemical signature, which often reduces attraction range. Longer volatility means a slower release rate, which also cuts range. A proposal that claims all three without addressing the tradeoff will not persuade an entomologist.
Improved trap design
No dependence on sticky coatings. Unattended trap or device operation, including advanced pest identification that does not require manual insect collection and laboratory analysis. Remote user communications with the trap or device, and communications between traps or devices.
On-board detection, identification, and surveillance hardware and software
Single or multi-sensor design, for example capacitive, optical, or optoacoustic. High NWS probability of detection and probability of false alarm to provide for actionable decision making. Capability to distinguish between NWS and related species such as Cochliomyia macellaria. Capability to distinguish between marked sterile and non-sterile NWS. Remotely upgradable code. Enables NWS movement forecasting. Capability to handle multiple pests via code update.
Two of those deserve emphasis. Distinguishing Cochliomyia hominivorax from Cochliomyia macellaria, the secondary screwworm, is a hard discrimination problem between closely related congeners, and it is the specific reason a non-selective lure plus a coarse sensor is insufficient. And distinguishing marked sterile from non-sterile NWS matters because the sterile insect technique is the primary control method: releases of sterilized males suppress wild populations, and a surveillance system that cannot tell a released sterile fly from a wild one will produce false alarms during a control campaign. That requirement tells you the system has to work alongside an active eradication program.
Low maintenance
Battery life of a minimum of 3 months, assuming low NWS prevalence and a strong user-device and device-device network signal. Required maintenance, for example lure refreshment, interval of two weeks or longer. Simple, rapid maintenance procedure.
Note the mismatch worth resolving in your design: battery life minimum 3 months, but lure refreshment interval two weeks or longer, while the lure volatility target elsewhere is stated as consistent with a one-month maintenance interval. Read the one-month lure volatility as the goal and the two-week interval as the floor, and be explicit about what your design achieves.
Physical
Lightweight, easily deployed, low cost. Approximately 15 cm by 15 cm by 15 cm, 600 g.
That is a small, light device. Fifteen centimeters cubed and 600 grams constrains battery capacity, which constrains sensing and compute, which constrains identification accuracy. The whole design closes or does not close on that budget, and showing the budget is the most persuasive thing you can put in the proposal.
Program-level Phase II requirements
The performer must have access to an ACL-3 test facility. The system demonstration must be conducted at TRL 6 using a prototype system in a relevant environment. DP2 will also require commercialization and transition planning along with technology development. Throughout the phase, the proposers must collaborate with commercial and government end-users to refine operational requirements and deployment scenarios of their developed solution. Manufacturing and scaling plans for production must also be developed before the end of the Phase. The final report must also include technology transfer documents outlining planned opportunities for commercial and government applications.
Note that end-user collaboration is a stated requirement, not a suggestion. USDA APHIS is the obvious government end-user, and the topic's references are almost entirely USDA documents. Establishing that relationship before you propose is worth real effort.
The Feasibility Requirement
This is a Direct to Phase II topic only. The Government expects that the small business has accomplished the following in a Phase I-type feasibility effort and developed a prototype NWS Distributed Detection System to address, at a minimum, the basic requirements of the stated objectives.
For this Direct-to-Phase II SBIR, a technical report containing Phase I Feasibility Documentation is required to demonstrate that Phase I feasibility has been met.
The Phase I Feasibility Documentation must contain a detailed description of the technical plan, milestones, and data substantiating that they have designed and field-tested a functional, low-power Distributed Detection System prototype outside of the SBIR program.
This feasibility report must provide empirical data verifying that the physically demonstrated sensor platforms can successfully isolate and identify target insects, while operating on strict energy budgets and seamlessly transmitting synchronized threat data across a coordinated wireless network.
Break that into its four claims, because all four must be evidenced. You designed and field-tested a functional low-power distributed detection system prototype. Outside the SBIR program. The physical platforms successfully isolate and identify target insects. And they operate on strict energy budgets while transmitting synchronized data across a coordinated wireless network.
Note the phrase "target insects" rather than New World Screwworm specifically. That is a meaningful opening. If you have a fielded distributed insect detection network working on a different target species, that plausibly satisfies the feasibility bar, and the NWS-specific discrimination becomes the Phase II development. Companies working in agricultural pest monitoring, mosquito surveillance, or stored-product pest detection should read this topic carefully rather than dismissing it as out of domain.
For the Standard Proposal Format, DP2 feasibility documentation shall not exceed 10 pages, the DP2 technical proposal shall not exceed 20 pages, and the Phase II commercialization strategy shall not exceed five pages, which should be the last section of the technical volume.
The Milestone and Deliverable Schedule
DARPA specifies base and option milestones and deliverables separately. Build your plan to match.
DP2 Base milestones, 12 months
Month 1: identify the candidate sensor technologies, lure chemistries, and device physical design.
Month 3: initial test results for various sensor and lure combinations integrated into a device. Finalize the device sensors and lure chemistry options. Develop the prototype device design and device test plan.
Month 6: assemble at least four prototype devices and collect initial test results for the prototype devices, individually and networked. Begin battery life and lure longevity testing.
Month 9: collect initial NWS probability of detection and probability of false alarm results against unrelated and related insects, for example Cochliomyia macellaria.
Month 12: refine the sensor subsystem for improved NWS detection and identification. Initial networked system demonstration, including remote user-device and device-device connectivity and remotely upgradable code.
DP2 Base deliverables
Month 1: sensor and lure candidate selection and design of experiments report.
Month 3: sensor and lure test results report, and device design and test plan.
Month 6: report describing test results for individual and networked devices.
Month 9: report describing NWS detection and identification results against unrelated and related insects, including updates to sensor design and lure technologies.
Month 12: Phase II feasibility report documenting device and system design, testing and validation, performance to date, manufacturing and scaling plans, and commercialization and transition plan.
DP2 Option 1, months 13 to 18
Milestones: Month 15, design scale-up to pilot plant quantities. Month 18, prototype system demonstration in a relevant environment.
Deliverables: Month 15, report documenting pilot plant design and operations. Month 18, report on field demonstration results.
DP2 Option 2, months 19 to 24
Milestones: Month 21, scale-up to pilot plant quantities. Month 24, final performance testing and commercialization plan.
Deliverables: as listed in the topic, Month 15 report documenting pilot plant design and operations, and Month 18 final report on system design and performance, and commercialization and transition plan.
Note that the Option 2 deliverable months in the topic read Month 15 and Month 18, which duplicates the Option 1 deliverable months and does not align with the Option 2 milestone months of 21 and 24. That looks like a drafting error. The sensible reading is that Option 2 deliverables fall at Months 21 and 24 matching their milestones, and this is a good question to submit to SBIR_BAA@darpa.mil before the October 14 deadline.
Also worth noting: the milestone structure moves to pilot plant quantities in both options, which signals that DARPA is thinking about manufacturing scale early. A device intended for national border coverage needs to be producible in the thousands, and the topic's design targets of lightweight, easily deployed, and low cost point the same way.
Phase III Dual Use
Successful development of an NWS distributed detection and surveillance system will have significant applications in both military and commercial sectors.
The most direct DoW application involves force protection disease vector control.
In addition, the distributed insect detection technology may expand to non-agricultural or disease prevention applications such as insect swarm behavioral changes as a proxy for human activities.
If the pest sensor and identification device can be made multifunctional in terms of what it senses, for example movement or vibration, or a chemical moiety, applications can expand farther into areas such as vehicle detection or an electronic "sniffer."
The key commercial application is agriculture protection for both crops and livestock, which also overlaps with National Security.
That progression is worth following in your commercialization strategy. The near market is agricultural pest surveillance, which is large and already buys traps. The defense market is force protection vector control. And the topic itself raises the possibility that a distributed low-power multi-modal sensing network with on-board classification generalizes well beyond insects, which is a longer-term argument for platform value rather than product value.
Funding, Cost Structure, and DARPA Mechanics
The award
$1,100,000 over a 12 month base, plus Option 1 of $500,000 over 6 months and Option 2 of $500,000 over 6 months. Total $2,100,000 across 24 months if both options are exercised.
The Government reserves the right to award all, some, one, or none of the options based on available funding and the performer's technical performance. The resources made available for each topic will depend on the quality of the proposals received and the availability of funds.
Contract type, which you must elect
DARPA may award FAR-based contracts, firm-fixed-price or cost-plus reimbursement, or Other Transactions for Prototype under the authority of 10 U.S.C. 4021, subject to approval of the Contracting Officer or Agreements Officer respectively. Proposers must state their requested contract type in their proposal.
If requesting a cost-plus reimbursement contract, include your Defense Contract Management Agency Final Determination Letter showing approval of your accounting system. If requesting an Other Transaction for Prototype, include a completed OT in your proposal using the Model OT for Prototype on the DARPA Small Business site, plus completed OT Certifications, both loaded in Volume 5, completing at minimum the color-coded areas and providing redlines with explanations for any article you wish to negotiate. No additional action is required for firm-fixed-price.
Templates are mandatory
Templates for Volume 2 and Volume 3 are provided as attachments on the DARPA Small Business website. Use of the DARPA Cost Proposal template is mandatory.
Technical and Business Assistance
Phase II awardees may request up to $25,000 per Phase II project. TABA funding is in addition to the cost ceilings and is not subject to profit or fee. TABA funding requests will be reviewed by the respective contracting office or specialist at time of award to ensure compliance with TABA requirements.
For this topic, market validation with agricultural end-users and regulatory support are both plausible uses, given that the device will need to work within USDA surveillance programs.
Questions and the FAQ
DSIP Topic Q&A will not be available for these DARPA topics. Technical questions related to improving the understanding of a topic's requirements must be submitted by October 14, 2026, by email to SBIR_BAA@darpa.mil with the topic number in the subject line, including the name, email address, and telephone number of a point of contact. Questions submitted within seven calendar days of the proposal due date may not be answered. DARPA posts a consolidated FAQ under the topic number summary on its Small Business site, updated on an ongoing basis until one week prior to the proposal due date.
DARPA will not accept late proposals.
Classification and marking
All proposals are required to be UNCLASSIFIED or CUI. Do not include any classified information in your proposal submission. Do not include any proprietary information on the Proposal Coversheet in Volume 1.
Proposal titles, abstracts, anticipated benefits, and keywords of proposals selected for contract award will undergo a DARPA Policy and Security Review and are subject to revision or redaction by DARPA. Final approved versions may appear on the DoW SBIR/STTR awards website and the SBA's award website at sbir.gov/awards.
Registrations
Proposers should ensure they have an accurate and active entity registration on SAM.gov. Those engaging in ITAR or CUI work for DARPA must have CMMC Level 2 certification, though the projected requirement for this topic is Level 1. DARPA points to sprs.csd.disa.mil/nistsp.htm and notes Project Spectrum at projectspectrum.io as an assistance resource.
Venture capital, hedge fund, and private equity ownership
Proposers that are more than 50 percent owned by multiple venture capital operating companies, hedge funds, private equity firms, or any combination of these as set forth in 13 CFR 121.702 are eligible to submit proposals in response to DARPA topics advertised within this BAA. Three conditions apply: register with the SBA Company Registry Database before submitting; submit the Majority-Owned VCOC, HF, and PEF Certification, with the SBIR VC Certification available on the DARPA Small Business site, in Supporting Documents Volume 5; and immediately notify the Contracting Officer, register in the appropriate SBA database, and submit the required certification if you enter that ownership class after submitting but before receiving a funding agreement.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW SBIR Program BAA. DARPA will conduct an evaluation of each conforming proposal. Proposals that do not comply with the requirements detailed in this BAA and the research objectives of the corresponding topic are considered non-conforming and are therefore not evaluated nor considered for award.
Using the evaluation criteria, the Government will evaluate each proposal in its entirety, documenting the strengths and weaknesses relative to each evaluation criteria, and based on those will determine the proposal's overall selectability for funding. Proposals will not be evaluated against each other but on their own individual merit.
A selectable proposal is one where the strengths of the overall proposal outweigh its weaknesses, with no accumulated weaknesses that would require extensive negotiations or a resubmitted proposal. A non-selectable proposal is one where the strengths do not outweigh the weaknesses.
Proposing firms will be notified of selection or non-selection status within 90 calendar days of the closing date of the BAA. The Corporate Official indicated on the Proposal Cover Sheet will be notified by email. In accordance with the SBA SBIR/STTR Policy Directive, Appendix I, paragraph 4, subparagraph (d), DARPA will provide a technical evaluation narrative to the proposer for each proposal submitted in response to a topic. An informal feedback session may additionally be requested via email at sbir@darpa.mil, provided at the sole discretion of DARPA.
Company Commercialization Report information will not be considered by DARPA during proposal evaluations.
Protests regarding the selection decision should be submitted, as prescribed in FAR 33.106(b) and FAR 52.233-3, to DARPA Contracts Management Office, 675 N. Randolph Street, Arlington, VA 22203, by email to CMO_SBIRProtests@darpa.mil and sbir@darpa.mil.
Post-award support
DARPA provides Transition and Commercialization Support Program services to Phase II and DP2 awardees upon contract execution at no cost to awardees, with the goal of maximizing the potential to move technology beyond Phase II into other research and development programs, DoW acquisition programs, other federal programs, or the commercial market. Awardees may also be eligible for the Embedded Entrepreneurship Initiative, an invitation-only program at DARPA's sole discretion, typically no more than $310,000 per awardee over the duration of the award, supporting a Senior Commercialization Advisor relationship, investor working group connections, and hiring an embedded entrepreneur to execute a Go-to-Market strategy.
The References, Which Are the Reading List
Seven references, and unusually for a DARPA topic they are almost entirely government program documents rather than research papers. That tells you the customer is a working surveillance program, not a research community.
USDA 2025, US Department of Agriculture, Animal and Plant Health Inspection Service, New World Screwworm Ready Reference Guide, Historical Economic Impact, January 2025.
CDC 2025, Centers for Disease Control and Prevention, New World Screwworm, What You Need To Know, 2025.
USDA 2026a, US Department of Agriculture, Animal and Plant Health Inspection Service, Fly Trapping: A Critical Tool for Detecting New World Screwworm, June 2026.
USDA 2026b, US Department of Agriculture, Animal and Plant Health Inspection Service, Surveillance, July 1 2026.
USDA 2026c, US Department of Agriculture, San Angelo Strike Team Uses Baited Sticky Lure Traps to Track Screwworm Activity in Texas, June 27, 2026.
Hickner 2023, Paul V. Hickner et al., A new formulation of screwworm fly attractant with reduced hazardous chemicals and transport restrictions, Journal of Medical Entomology, 2023.
Arizona 2026, University of Arizona, U of A secures $3.74 million to strengthen Arizona's preparedness for New World screwworm, June 30 2026.
Two of these are worth particular attention. The Hickner 2023 paper is the lure chemistry reference, and since improving on Swormlure-5 is a named requirement, engaging with that work is the fastest way to show you understand the lure problem rather than treating it as a black box. And the USDA 2026a fly trapping document plus the USDA 2026c San Angelo Strike Team item describe the actual field operation your device has to fit into, including who deploys the traps and how.
The Arizona 2026 reference is also a signal about the ecosystem: a university with $3.74 million in NWS preparedness funding is both a potential ACL-3 partner and a potential competitor's partner.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
Technical question deadline: October 14, 2026, to SBIR_BAA@darpa.mil with the topic number in the subject line
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Selection notification: within 90 calendar days of BAA close
Base period: 12 months from award
Option 1: 6 months. Option 2: 6 months
A working backward plan
Before September 23. Secure and document ACL-3 test facility access, which is the hardest item and a stated Phase II requirement. Download the mandatory DARPA Volume 2 and Volume 3 templates. Read the FAQ for this topic and keep rechecking it. Assemble your feasibility package: evidence you designed and field-tested a functional low-power distributed detection system prototype outside the SBIR program, with empirical data on insect isolation and identification, strict energy budget operation, and synchronized data transmission across a coordinated wireless network. Read the Hickner 2023 lure paper and form a view on beating Swormlure-5 on selectivity, range, and volatility simultaneously. Reach out to USDA APHIS, since end-user collaboration is a stated Phase II requirement. Decide your contract type and prepare the corresponding documents. Confirm SAM registration. If venture-backed, register with the SBA Company Registry and obtain the SBIR VC Certification. Submit technical questions before October 14, including the Option 2 deliverable month discrepancy.
September 23 through October 5. Draft the technical volume in three parts: 10 pages of feasibility documentation, 20 pages of technical proposal, 5 pages of commercialization strategy. Structure the technical proposal around the specification list: system scalability, lure improvement with the three-way tradeoff addressed, trap design, on-board detection and identification including the Cochliomyia macellaria and sterile-versus-wild discriminations, low maintenance with the battery and lure interval budget shown, and the physical envelope with a power budget that closes at 15 cm cubed and 600 g.
October 6 through October 14. Build the cost volume in the mandatory template across the base and both options, aligned to the stated milestones and deliverables. Price the ACL-3 facility access, prototype fabrication of at least four devices, insect colony or specimen access, and pilot plant design work.
October 15 through October 18. Assemble Volume 5 with contract-type documents and certifications, complete Volume 7 and the Volume 4 CCR, and run compliance: 35 pages with the three internal section limits observed, unclassified or CUI only, no proprietary information on the coversheet, mandatory cost template.
October 19 through October 20. Submit and certify in DSIP.
Frequently Asked Questions
What is DARPA SBIR topic DPA26BZ06-DV024?
DPA26BZ06-DV024 is a DARPA SBIR Direct to Phase II topic titled "New World Screwworm Networked Detection, Identification, and Surveillance System," released under the DoW 2026 SBIR Broad Agency Announcement, Release 6. It seeks a real-time, networked pest detection, identification, and surveillance technology providing actionable New World Screwworm detection and identification accuracy, system reliability, low maintenance, and long unattended operation.
How much funding is available?
$1,100,000 for a 12 month base period, plus Option 1 of $500,000 over 6 months and Option 2 of $500,000 over 6 months, for a total of $2,100,000 across 24 months if both options are exercised. Up to $25,000 in Technical and Business Assistance may be requested in addition to the cost ceiling.
When is the proposal deadline?
October 21, 2026. DARPA will not accept late proposals. Note the separate technical question deadline of October 14, 2026.
Do I need an ACL-3 facility?
Yes. The Phase II description states that the performer must have access to an ACL-3 test facility. Arthropod Containment Level 3 access is not quickly arranged, so either you hold it or you need a documented arrangement with an institution that does. Name it in the proposal.
Can I ask questions through DSIP Topic Q&A?
No. DARPA states DSIP Topic Q&A will not be available for these topics. Technical questions go to SBIR_BAA@darpa.mil with the topic number in the subject line by October 14, 2026. Questions submitted within seven calendar days of the due date may not be answered. DARPA posts a consolidated FAQ on its Small Business site, updated until one week before the due date.
Why does this matter to the Department of War?
The topic names force protection disease vector control as the most direct DoW application, and notes that agriculture protection for crops and livestock, the key commercial application, also overlaps with national security. The economic figure DARPA cites is that re-entry of NWS into the United States could cost U.S. cattle producers $732 million per year.
What is wrong with current traps?
The three types in use at the US Southern Border, sticky fly traps, liver-baited net traps, and gator traps, all require manual servicing, fly collection, and fly delivery to a laboratory. Suspected samples are forwarded to the USDA National Veterinary Services Laboratories in Ames, Iowa for official confirmation. The latency and labor in that pipeline is the problem.
What is wrong with the current lure?
Swormlure-5, the state-of-the-art lure, is a synthetic bait based on a mix of volatile organic and volatile sulfur compounds, but such baits are not specific to New World Screwworm and can lure other species.
What must an improved lure achieve?
Three things: greater NWS selectivity, increased lure attraction range to reduce the number of traps needed, and volatility consistent with a one-month maintenance interval. Note these are in tension, since narrower selectivity and slower release both tend to reduce range. Address the tradeoff explicitly.
What discriminations must the sensor make?
Two named hard ones. Capability to distinguish between NWS and related species such as Cochliomyia macellaria, the secondary screwworm, which is a congener discrimination. And capability to distinguish between marked sterile and non-sterile NWS, which matters because the sterile insect technique is the primary control method and a surveillance system that cannot tell released sterile flies from wild ones will generate false alarms during a control campaign.
What sensing modalities are suggested?
Single or multi-sensor design, with capacitive, optical, and optoacoustic given as examples. Elsewhere the topic mentions low-power localized optical, acoustic, electrical, or chemical sensing. DARPA does not prescribe a modality.
What are the physical and power requirements?
Approximately 15 cm by 15 cm by 15 cm and 600 g. Battery life of a minimum of 3 months assuming low NWS prevalence and a strong network signal. Required maintenance such as lure refreshment at an interval of two weeks or longer, with a simple, rapid maintenance procedure. Lightweight, easily deployed, and low cost.
What networking is required?
Devices must form resilient, low-bandwidth ad-hoc networks to transmit real-time, highly compressed detection telemetry, enabling movement forecasting and biosecurity intervention even in remote and disconnected environments. The system must support remote user communications with the device and communications between devices, plus remotely upgradable code.
Can I submit a Phase I proposal?
No. This is a Direct to Phase II topic only. A technical report containing Phase I Feasibility Documentation is required to demonstrate that Phase I feasibility has been met.
What feasibility documentation is required?
A detailed description of the technical plan, milestones, and data substantiating that you designed and field-tested a functional, low-power Distributed Detection System prototype outside of the SBIR program. The report must provide empirical data verifying that the physically demonstrated sensor platforms can successfully isolate and identify target insects while operating on strict energy budgets and seamlessly transmitting synchronized threat data across a coordinated wireless network.
Does the feasibility work have to be on screwworm specifically?
The topic says "target insects" rather than naming NWS in the feasibility requirement, which suggests a fielded distributed insect detection network on a different species may satisfy the bar, with NWS-specific discrimination becoming the Phase II development. Companies in agricultural pest monitoring, mosquito surveillance, or stored-product pest detection should read the requirement closely rather than assuming they are out of domain. If in doubt, ask DARPA before October 14.
How long can my technical volume be?
35 pages in the Standard Proposal Format, structured as up to 10 pages of DP2 feasibility documentation, up to 20 pages of DP2 technical proposal, and up to 5 pages of Phase II commercialization strategy, which should be the last section.
What TRL must the Phase II demonstration reach?
TRL 6, using a prototype system in a relevant environment.
Is there an error in the Option 2 deliverable schedule?
It appears so. Option 2 milestones are stated at Month 21 and Month 24, but the Option 2 deliverables are listed at Month 15 and Month 18, which duplicates the Option 1 deliverable months. The sensible reading is that Option 2 deliverables fall at Months 21 and 24 matching their milestones. This is worth a question to SBIR_BAA@darpa.mil before October 14.
Do I have to work with end-users?
Yes. Throughout the phase, proposers must collaborate with commercial and government end-users to refine operational requirements and deployment scenarios. USDA APHIS is the obvious government end-user, and the topic's reference list is largely USDA program documents.
Do I need manufacturing plans?
Yes. Manufacturing and scaling plans for production must be developed before the end of the Phase, and both options include milestones for scale-up to pilot plant quantities. The final report must also include technology transfer documents outlining planned opportunities for commercial and government applications.
Do I have to choose a contract type?
Yes. Proposers must state their requested contract type. DARPA may award FAR-based firm-fixed-price or cost-plus reimbursement contracts, or Other Transactions for Prototype under 10 U.S.C. 4021. Cost-plus requires your DCMA Final Determination Letter showing accounting system approval. An OT requires a completed Model OT plus OT Certifications in Volume 5. Firm-fixed-price requires no additional action.
Is the cost template mandatory?
Yes. Templates for Volume 2 and Volume 3 are on the DARPA Small Business website, and use of the DARPA Cost Proposal template is mandatory.
Are venture capital backed companies eligible?
Yes. Proposers more than 50 percent owned by multiple venture capital operating companies, hedge funds, private equity firms, or any combination as set forth in 13 CFR 121.702 are eligible, subject to registering with the SBA Company Registry Database before submitting, submitting the Majority-Owned VCOC, HF, and PEF Certification in Volume 5, and notifying the Contracting Officer if you enter that class after submitting but before award.
How will my proposal be evaluated?
Against the evaluation criteria in the DoW SBIR Program BAA. DARPA evaluates each conforming proposal in its entirety, documenting strengths and weaknesses relative to each criterion, then determines overall selectability. Proposals are not evaluated against each other. A selectable proposal is one where strengths outweigh weaknesses with no accumulated weaknesses requiring extensive negotiations or resubmission.
Will I get feedback if not selected?
Yes. DARPA will provide a technical evaluation narrative to the proposer for each proposal submitted in response to a topic, per the SBA SBIR/STTR Policy Directive. An informal feedback session may additionally be requested via sbir@darpa.mil, at DARPA's sole discretion.
What is the commercial market?
The key commercial application named is agriculture protection for both crops and livestock, which the topic notes also overlaps with national security. Beyond that, DARPA raises the possibility that the distributed insect detection technology may expand to non-agricultural or disease prevention applications such as insect swarm behavioral changes as a proxy for human activities, and that a multifunctional sensor could expand into vehicle detection or an electronic sniffer.
Who do I contact with questions?
The DARPA Small Business Programs Office at SBIR_BAA@darpa.mil for both program administration and topic technical questions, with the topic number in the subject line. DSIP technical support at DoDSBIRSupport@reisystems.com with a copy to SBIR_BAA@darpa.mil, available Monday through Friday 9:00 a.m. to 5:00 p.m. ET. DARPA also offers free resources through DARPAConnect at DARPAConnect.us.
Positioning Advice for Companies Considering This Topic
Sort out ACL-3 access before anything else. It is a stated Phase II requirement, it takes time, and a proposal that cannot show where the containment work happens will read as unexecutable regardless of technical quality.
Read the feasibility requirement generously but honestly. It says target insects, not screwworm. A company with a fielded, networked, low-power insect detection system on another species has a plausible path in. A company with a laboratory demonstration and no field network does not, because the requirement is explicit about field testing and about synchronized data across a coordinated wireless network.
Show the power budget that closes at 600 grams. Fifteen centimeters cubed, 600 grams, three months of battery life, on-board machine learning classification, and mesh networking is a tight envelope. The single most convincing artifact you can include is a credible power and mass budget showing how sensing, compute, radio, and lure all fit. Most proposals will assert the requirements; showing the arithmetic distinguishes you.
Take the lure seriously as chemistry, not packaging. Greater selectivity, longer range, and one-month volatility pull against each other. Read Hickner 2023 and engage with the actual attractant chemistry. A proposal with an excellent sensor and a hand-wave at the lure has solved half the problem, and it is arguably the easier half given that a non-selective lure fills your trap with the wrong flies.
Solve the sterile-versus-wild discrimination explicitly. This is the requirement most likely to be skipped, and it is the one that reveals whether you understand the operational context. Sterile insect technique releases are how screwworm is controlled, so any surveillance system deployed during a campaign will see large numbers of marked sterile flies. Say how you tell them apart, and note what marking method you assume.
Quantify the latency you eliminate. Every suspected detection today goes to Ames, Iowa. Estimating the current detect-to-confirm timeline and what your device reduces it to is the clearest statement of operational value, and it is the argument USDA will care about.
Engage USDA APHIS now. End-user collaboration is a Phase II requirement, the reference list is mostly USDA documents, and the agency is actively running the surveillance program your device would join. A letter or documented engagement changes how the transition plan reads.
Design for thousands of units. The topic asks for coverage as large as national borders, names low cost as a design goal, and puts pilot plant scale-up in both option periods. Manufacturability is not a Phase III concern here, it is in the base and option milestones.
Build the multi-pest story into the architecture. Capability to handle multiple pests via code update is a stated requirement, and remotely upgradable code is another. That combination means the device should be a platform, and the commercialization argument is stronger for a platform that addresses agricultural pest surveillance broadly than for a single-species instrument.
Ask about the Option 2 deliverable months. The dates in the topic do not align with the Option 2 milestones. Submitting that question before October 14 both gets you an answer and demonstrates careful reading.
DARPA SBIR DPA26BZ06-DV023: Novel Radio Frequency Sensing Technologies
Deadline: October 21, 2026
Funding Award Size: $1.2m
Description: Complete guide to DARPA SBIR Direct to Phase II topic DPA26BZ06-DV023, high-altitude airborne subsurface radar for critical minerals. $700K base plus $500K option. Closes October 21, 2026.
Quick Answer
DPA26BZ06-DV023 is a DARPA SBIR Direct to Phase II topic under the DoW 2026 SBIR Broad Agency Announcement, Release 6. DARPA wants a high-altitude, high-speed airborne geologic sensing system: a self-contained radar payload flying at roughly 40,000 feet and about 400 knots that produces rapid, stand-off 3D subsurface tomographic imaging to detect subterranean features such as metals and critical minerals. The award is $700,000 over an 18 month base period, with a $500,000 option over 6 months. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
The physics is the whole problem, and DARPA states it in the topic. Radar transmit power for small targets scales as the fourth power of range. Moving from a 10,000 foot baseline to 40,000 feet is a 4x increase in distance, which nominally demands a 256x increase in radiated power to maintain equivalent subsurface imaging performance. That single number defines the engineering challenge: a kW-class RF power amplifier, an antenna that survives 400 knots at altitude, and heat dissipation to match, all inside a fully self-contained payload with its own power generation.
This is not an entry-level topic. To be eligible you must already have developed, integrated, and validated an airborne subsurface radar or similar system in a physical environment at low altitude and low speed up to 10,000 feet. DARPA says explicitly that modeling and simulation data alone is not sufficient to prove feasibility.
Topic At a Glance
Topic number: DPA26BZ06-DV023
Title: Novel Radio Frequency Sensing Technologies
Agency: Defense Advanced Research Projects Agency (DARPA)
Solicitation: DoW 2026 Small Business Innovation Research Broad Agency Announcement, Release 6, DARPA Proposal Submission Instructions
Program type: Direct to Phase II (DP2). This topic is soliciting DP2 proposals only
Technical volume format: Standard Proposal Format, 35 pages
Base award: $700,000
Base period of performance: 18 months
Option 1: $500,000 over 6 months
Option 2: none
Component Technology Priority Area: Integrated Sensing and Cyber
Projected CMMC level requirement: Level 2 (Self)
Export control status: Restricted under ITAR 22 CFR Parts 120-130 and EAR 15 CFR Parts 730-774
Technical and Business Assistance: up to $25,000 per Phase II project, in addition to the cost ceiling
Topic Q&A: DSIP Topic Q&A is not available for DARPA topics. Technical questions go to SBIR_BAA@darpa.mil by October 14, 2026
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026
Submission portal: DSIP at dodsbirsttr.mil
Keywords: ground penetrating radar, subsurface tomography, standoff radar, void detection, high-power RF, airborne geophysics, subterranean imaging, RF shielding, critical minerals
How DARPA Differs From Other Components
If you have worked Air Force or Navy SBIR topics, several DARPA-specific rules will catch you out.
There is no DSIP Topic Q&A. DARPA states that DSIP Topic Q&A will not be available for these DARPA topics. Technical questions related to improving the understanding of a topic's requirements must be submitted by October 14, 2026, by email to SBIR_BAA@darpa.mil with the topic number in the subject line. All questions must be in English and must include the name, email address, and telephone number of a point of contact. Questions submitted within seven calendar days of the proposal due date may not be answered. DARPA posts a consolidated Frequently Asked Questions document on its Small Business site, linked under the topic number summary, updated on an ongoing basis until one week prior to the proposal due date.
Read that FAQ before you write, and read it again the week before you submit. It is the only public question channel for this topic and it will contain answers to questions you did not think to ask.
DARPA will not accept late proposals. The instructions say this twice, in bold. Proposers are encouraged to submit as early as possible to avoid unexpected delays due to high traffic in the final hours before a BAA closes.
You must state your requested contract type. DARPA may award FAR-based contracts, either firm-fixed-price or cost-plus reimbursement, or Other Transactions for Prototype under the authority of 10 U.S.C. 4021, subject to approval of the Contracting Officer or Agreements Officer respectively. Proposers must state their requested contract type in their proposal. If requesting a cost-plus reimbursement contract, you must include your Defense Contract Management Agency Final Determination Letter showing approval of your accounting system. If requesting an Other Transaction for Prototype, you must include a completed OT in your proposal using the Model OT for Prototype on the DARPA Small Business site, plus completed OT Certifications, both loaded in Volume 5. At a minimum you need to complete the areas color coded in the template, and if you wish to negotiate any articles or attachments you must provide redlines and comments explaining each. No additional action is required for a firm-fixed-price request.
The cost proposal template is mandatory. Templates for Volume 2 Technical Volume and Volume 3 Cost Volume are provided as attachments on the DARPA Small Business website. Use of the DARPA Cost Proposal template is mandatory.
Everything must be unclassified or CUI. All proposals are required to be UNCLASSIFIED or CUI, and you must not include any classified information in your proposal submission. Do not include any proprietary information on the Proposal Coversheet in Volume 1.
You get a written evaluation narrative regardless of outcome. In accordance with the SBA SBIR/STTR Policy Directive, Appendix I, paragraph 4, Method of Selection and Evaluation Criteria, subparagraph (d) Release of Proposal Review Information, DARPA will provide a technical evaluation narrative to the proposer for each proposal submitted in response to a topic. An informal feedback session may additionally be requested by the proposing firm via email at sbir@darpa.mil, provided at the sole discretion of DARPA. That is more generous than most components and worth using.
DARPA reserves broad discretion on partial awards. The Government reserves the right to select for negotiation all, some, one, or none of the proposals received, and to make awards with or without communications with proposers. It also reserves the right to award all, some, one, or none of the options based on available funding and the performer's technical performance. If warranted, portions of resulting awards may be segregated into pre-priced options, and DARPA reserves the right to accept proposals in their entirety or to select only portions of proposals for award. The Government reserves the right to remove a proposal from award consideration should the parties fail to reach agreement on award terms, conditions, and price within a reasonable time, or should the proposer fail to provide requested additional information within three business days.
Two commercialization programs come with a DP2 award. DARPA provides Transition and Commercialization Support Program services to Phase II and DP2 awardees upon contract execution, at no cost, with the goal of maximizing the potential for SBIR/STTR companies to move technology beyond Phase II into other research and development programs, DoW acquisition programs, other federal programs, or the commercial market. Separately, awardees may be eligible for the Embedded Entrepreneurship Initiative, an invitation-only program at DARPA's sole discretion, typically no more than $310,000 per awardee over the duration of the award, funding a Senior Commercialization Advisor relationship, connections to investor working groups, and the hiring of an embedded entrepreneur to execute a Go-to-Market strategy.
What DARPA Is Actually Looking For
The objective
Develop, integrate, and flight-test a high-altitude, meaning roughly 40,000 feet, high-speed, meaning approximately 400 knots, airborne geologic sensing system. The system must be capable of providing rapid, stand-off 3D subsurface tomographic imaging to detect subterranean features such as metals and critical minerals.
Why the current state of the art does not work
Detecting and mapping deep subsurface mineral deposits typically requires localized or low-altitude operations. Current state-of-the-art airborne subsurface radar systems are limited to relatively low-altitude and low-speed platforms due to the physical limitations of signal attenuation over distance. Increasing altitude and airspeed would allow for larger areas to be scanned in less time.
This topic seeks the rapid development of an advanced airborne subsurface radar system designed for seamless integration onto high-altitude fixed-wing aircraft.
The scaling law you must address
Proposers must address the physical scaling laws of radar propagation through lossy media. According to the radar equations for small targets of a radar cross section of approximately one wavelength, the necessary transmit power scales as a function of range to the fourth power, or R4. For instance, transitioning from a 10,000 foot baseline to a 40,000 foot operational altitude represents a 4x increase in distance, nominally demanding a 256x increase in radiated power to maintain equivalent subsurface imaging performance.
Due to the higher power requirements at higher altitudes, system designs must prioritize transmit power scaling, advanced antenna structure designs, and robust heat dissipation systems.
That 256x figure is the organizing fact of this topic. Every design decision you make is either a way to supply that power, a way to avoid needing all of it through processing or waveform gains, or a way to survive the thermal consequences. A proposal that does not confront the number directly has not engaged with the topic.
The self-contained payload requirement
To facilitate rapid, modular deployment across diverse airframes without requiring permanent, costly aircraft modifications, the entire system must be designed as a fully self-contained payload. It must utilize independent, internal power generation and an edge computational architecture.
Three consequences. You cannot draw power from the host aircraft, which means generating kW-class RF power from onboard generation. You cannot rely on aircraft cooling. And you cannot downlink raw data for ground processing, because the architecture is specified as edge computational.
The five key technical challenges
DARPA lists five challenges proposers must address. Use them as your technical volume structure.
High-power transmitter and power scaling: engineering and validating a highly reliable kW-class RF power amplifier.
Aerodynamic and structural antenna design: designing and reinforcing an externally deployable or structurally integrated antenna system capable of withstanding aerodynamic loads at relevant velocities, approximately 400 knots, and altitudes, approximately 40,000 feet.
Thermal dissipation and management: integrating passive or active high-capacity heat dissipation systems to mitigate thermal loads under sustained high-power transmit cycles.
RF shielding and electromagnetic compatibility: implementing advanced RF shielding and electrical isolation to guarantee zero electromagnetic interference with the host aircraft's flight navigation, control, and mission avionics.
Note the word "zero." That is an absolute framing, and the single cited reference is MIL-STD-461G, the EMC standard. Treat EMC as a first-class design and test problem rather than a compliance afterthought.
Field-programmable gate arrays and edge processing: selecting FPGAs and processing techniques that can handle increased signal propagation delays, modified waveforms, and real-time tomographic data processing at high standoff distances.
The Feasibility Requirement
This topic is soliciting Direct to Phase II proposals only. Proposers must submit feasibility documentation in lieu of a Phase I proposal.
Proposers must demonstrate that they have already developed, integrated, and validated an airborne subsurface radar or similar system in a physical environment, for example low-altitude, low-speed flight regimes up to 10,000 feet.
Then the sentence that disqualifies most applicants: modeling and simulation data alone is not sufficient to prove feasibility.
The three required feasibility elements
The submitted feasibility documentation must include three things, and DARPA numbers them.
Empirical flight data: representative 3D tomographic subsurface reconstructions, signal-to-noise ratio calculations, and imagery demonstrating successful detection of underground mineral deposits.
Receiver characterization: technical documentation proving the receiver operates at a low thermal noise floor.
Analytical scaling models: detailed physical, mathematical, and electromagnetic models validating the R4 scaling equations and mapping the exact architectural transition from a low-power baseline to a high-power, high-altitude design.
That third element is where the proposal is won. Empirical flight data and receiver characterization establish that you have a working system. The scaling models establish that you know how to get from that system to this one. DARPA is asking for the bridge, explicitly and in detail, and the phrase "mapping the exact architectural transition" leaves no room for a general statement of intent.
Per the DP2 authority, if a proposer can provide adequate documentation to substantiate that the scientific and technical merit and feasibility described in the Phase I section of the topic has been met, and describes the potential commercial applications, the Direct to Phase II authority allows the Department of War to make an award under Phase II of the SBIR program without regard to whether the small business concern was provided an award under Phase I of an SBIR program.
For the Standard Proposal Format, DP2 feasibility documentation shall not exceed 10 pages, the DP2 technical proposal shall not exceed 20 pages, and the Phase II commercialization strategy shall not exceed five pages, which should be the last section of the technical volume. Those three sections total the 35 page limit.
The Phase II Program and Its Fixed Payable Milestones
Phase II is described as an accelerated hardware realization, integration, and flight-testing campaign. DARPA specifies fixed payable milestones, which means your payment schedule tracks these deliverables. Build your plan around them rather than proposing your own structure.
Base Period, 18 months:
Month 2, Subsystem Engineering: preliminary designs for the power amplifier, motherboard, and high-efficiency heat dissipation structures, plus an initial report on updated firmware architecture.
Month 4, Subsystem Design Completion: finalized engineering designs for all major subsystems.
Month 6, Structural Antenna Engineering Interim Report Number 1: manufacturing progress and initial benchtop testing for the power amplifier and antenna assembly to survive aerodynamic loads of approximately 400 knots.
Month 9, FPGA and Compute Development Interim Report Number 2: progress on integrated systems testing, including initial results from tomographic reconstruction algorithms at the edge.
Month 12, Safety and Test Readiness Review Data Package: comprehensive safety testing results on high-voltage and high-power RF emissions, and flight clearances documentation.
Month 15, High-Altitude Flight Campaign Interim Report Number 3: preliminary data from demonstration flights on a fixed-wing platform at approximately 40,000 feet and speed of approximately 400 knots, including measurements of radiated power and signal penetration.
Month 18, Data Analysis and Final Base Period Report: detailed analysis of flight data, fully reconstructed 3D subsurface tomograms of test sites, and system performance evaluation.
Option Period, 6 months:
Month 21, Interim Report Number 4: detailed data analysis, algorithm enhancement, and performance improvements of the prototype system.
Month 24, Comprehensive Final System Assessment Report: final evaluation of performance, including 3D subsurface tomogram data of relevant structures, against a low-altitude baseline and commercial viability analysis.
Two observations about this schedule. The Month 12 Safety and Test Readiness Review with flight clearances documentation is the gate that decides whether the Month 15 flight campaign happens on time. High-voltage and high-power RF airborne flight clearance is an approval process with its own timeline, and 12 months is not generous. Address your airworthiness and flight clearance pathway explicitly, including which platform and which test organization.
And the Month 24 final assessment requires comparison against a low-altitude baseline. That means your feasibility data is not just an eligibility document, it becomes the benchmark you are measured against. Choose what you submit accordingly.
Phase III Dual Use
DARPA describes the high-altitude geological sensing system as a highly disruptive dual-use technology.
Military and DoD applications: identifying additional domestic sources of critical minerals is a national security imperative that will improve access to materials needed for advanced defense systems. This capability would strengthen the U.S. supply chain for critical minerals.
Commercial and civil applications: DARPA calls the system a highly scalable "MRI for the Earth." It could map deep-seated geologic structures to rapidly locate critical mineral reserves and rare earth element deposits buried under hundreds of meters of geology. This significantly accelerates discovery timelines, increases exploratory extraction success rates from 1-in-200 to near certainty, and optimizes geologic risk management.
That 1-in-200 to near certainty claim is DARPA's own, and it is a strong commercial argument you can cite rather than construct. Mineral exploration economics are dominated by drilling failure rates, and a technology that changes the hit rate changes the industry's cost structure. Build your five page commercialization strategy on mining and exploration companies, critical mineral supply chain investors, and government geological surveys.
Funding, Cost Structure, and TABA
The award
$700,000 for an 18 month base period, plus a $500,000 option over 6 months, for $1,200,000 across 24 months if the option is exercised. There is no Option 2 on this topic.
The resources made available for each topic issued under this BAA will depend on the quality of the proposals received and the availability of funds. The Government reserves the right to award all, some, one, or none of the options based on available funding and the performer's technical performance.
Note that $700,000 for kW-class RF amplifier development, antenna structural qualification, thermal system integration, EMC qualification, and a high-altitude flight campaign is a tight budget. That is part of why the feasibility bar is set at an already-validated flying system: DARPA is funding the transition, not the invention.
Contract type
State your requested type. Firm-fixed-price requires no additional action. Cost-plus reimbursement requires your DCMA Final Determination Letter showing accounting system approval, included in the proposal. An Other Transaction for Prototype requires a completed Model OT plus OT Certifications in Volume 5.
DARPA describes the OT instrument favorably: the flexibility of the OT award instrument is beneficial because the performer will be able to apply its commercial best practices as required to carry out the research project, outside of FAR process-driven requirements, with streamlined milestone-driven performance intended to reduce time and effort on award administration and permit performers to focus on the research effort and rapid prototyping. For a hardware program with fixed payable milestones, that is worth considering seriously.
Technical and Business Assistance
Phase II awardees may request up to $25,000 per Phase II project. TABA funding is in addition to the Phase I and Phase II cost ceilings and is not subject to profit or fee. Refer to the DoW Program BAA for detailed information about TABA funding, allowable services, and requirements. TABA funding requests will be reviewed by the respective contracting office or specialist at time of award to ensure compliance with TABA requirements.
Export Control
The technology within this topic is restricted under ITAR 22 CFR Parts 120-130, which controls the export and import of defense-related material and services including export of sensitive technical data, and the EAR 15 CFR Parts 730-774, which controls dual use items.
Offerors must disclose any proposed use of foreign nationals, their countries of origin, the type of visa or work permit possessed, and the statement of work tasks intended for accomplishment by the foreign nationals in accordance with the Announcement. Offerors are advised that foreign nationals proposed to perform on this topic may be restricted due to the technical data under U.S. export control laws.
Note also from the DARPA Phase I instructions, which apply generally: proposers should ensure they have an accurate and active entity registration on SAM.gov, and those engaging in ITAR or CUI work for DARPA must have CMMC Level 2 certification. The projected requirement for this topic is Level 2 with self-assessment. DARPA points to sprs.csd.disa.mil/nistsp.htm for more information and notes that resources exist to assist companies with meeting this criterion, one of which is Project Spectrum at projectspectrum.io, where users can create accounts and gain access to educational videos and material as well as a cyber self-assessment.
Proposal Structure
A complete proposal submission consists of seven volumes: Proposal Cover Sheet, Technical Volume, Cost Volume, Company Commercialization Report, Supporting Documents, Fraud, Waste and Abuse Training, and Disclosures of Foreign Affiliations or Relationships to Foreign Countries.
Volume 1, Proposal Cover Sheet. Do not include any proprietary information on the coversheet. Note that proposal titles, abstracts, anticipated benefits, and keywords of proposals selected for contract award will undergo a DARPA Policy and Security Review and are subject to revision or redaction by DARPA. Final approved versions may appear on the DoW SBIR/STTR awards website and the SBA's award website at sbir.gov/awards.
Volume 2, Technical Volume. Standard Proposal Format, 35 pages, structured as up to 10 pages of DP2 feasibility documentation, up to 20 pages of DP2 technical proposal, and up to 5 pages of Phase II commercialization strategy as the last section. Templates are provided on the DARPA Small Business website. Refer to Appendix B, DARPA Direct to Phase II Instructions, for content requirements.
Volume 3, Cost Volume. Use of the DARPA Cost Proposal template is mandatory. Refer to Appendix B for content.
Volume 4, Company Commercialization Report. Completion of the CCR in DSIP is required, but DARPA will not consider information contained in the CCR during proposal evaluations.
Volume 5, Supporting Documents. In addition to documents required by DoW, small businesses may submit additional documentation to support the Technical Volume and the Cost Volume. Required certifications are listed in Appendix B. This is also where a Model OT and OT Certifications go if you are requesting an Other Transaction, where the SBIR VC Certification goes if you are majority-owned by multiple venture capital operating companies, hedge funds, or private equity funds, and where Data Rights Assertions go if applicable.
Volume 6, Fraud, Waste and Abuse Training. Material is in the Volume 6 section of the DSIP proposal submission module and must be thoroughly reviewed once per year to proceed with proposal submission.
Volume 7, Disclosures of Foreign Affiliations or Relationships to Foreign Countries. Complete the webform in Volume 7 of the DSIP submission. It will not be accepted as a PDF Supporting Document in Volume 5, and previous versions of the form must not be uploaded to Volume 5.
Eligibility and Evaluation
Venture capital, hedge fund, and private equity ownership
Proposers that are more than 50 percent owned by multiple venture capital operating companies, hedge funds, private equity firms, or any combination of these as set forth in 13 CFR 121.702 are eligible to submit proposals in response to DARPA topics advertised within this BAA.
Three conditions apply. Prior to submitting a proposal, firms must register with the SBA Company Registry Database. The proposer, within its submission, must submit the Majority-Owned VCOC, HF, and PEF Certification, with a copy of the SBIR VC Certification available on the DARPA Small Business site, included in Supporting Documents Volume 5. And should a proposer become a member of this ownership class after submitting its proposal and prior to any receipt of a funding agreement, the proposer must immediately notify the Contracting Officer, register in the appropriate SBA database, and submit the required certification.
DARPA's position here is permissive, which matters for a capital-intensive RF hardware topic where venture funding is common.
Evaluation and selection
All proposals will be evaluated in accordance with the evaluation criteria listed in the DoW SBIR Program BAA. DARPA will conduct an evaluation of each conforming proposal. Proposals that do not comply with the requirements detailed in this BAA and the research objectives of the corresponding topic are considered non-conforming and are therefore not evaluated nor considered for award.
Using the evaluation criteria, the Government will evaluate each proposal in its entirety, documenting the strengths and weaknesses relative to each evaluation criteria. Based on these identified strengths and weaknesses, the Government will determine the proposal's overall selectability for funding. Proposals will not be evaluated against each other during the evaluation process but rather evaluated on their own individual merit to determine how well the proposal meets the criteria stated in this BAA and the corresponding DARPA topic.
DARPA defines two outcomes. A selectable proposal is one that the Government has evaluated against the evaluation criteria listed in the DoW SBIR Program BAA and DARPA topic, and the strengths of the overall proposal outweigh its weaknesses, with no accumulated weaknesses that would require extensive negotiations or a resubmitted proposal. A non-selectable proposal is one where the strengths of the overall proposal do not outweigh its weaknesses.
That framing is useful. You are not competing against other proposals; you are being assessed for whether your strengths outweigh your weaknesses cleanly enough to award without renegotiation. Weaknesses that accumulate into a need for extensive negotiation make you non-selectable even if the technical concept is strong. Write to eliminate weaknesses, not only to accumulate strengths.
Awards will be given to proposers whose proposals are determined to be the most advantageous to the Government, consistent with instructions and evaluation criteria specified in the DoW SBIR Program BAA and availability of funding.
Proposing firms will be notified of selection or non-selection status within 90 calendar days of the closing date of the BAA. The Corporate Official indicated on the Proposal Cover Sheet will be notified by email.
It is DARPA's policy to treat all proposals as source selection information and to disclose their contents only for the purpose of evaluation. During the evaluation process, submissions may be handled by support contractors for administrative purposes or to assist with technical evaluation. All DARPA support contractors are expressly prohibited from performing DARPA-sponsored technical research and are bound by appropriate nondisclosure agreements. Input on technical aspects of the proposals may be solicited by DARPA from other Government or non-Government consultants and experts who are strictly bound by the appropriate nondisclosure requirements. No submissions in response to the BAA will be returned. Upon completion of the evaluation and selection process, an electronic copy of each proposal received will be retained at DARPA.
Protests
Refer to the DoW SBIR Program BAA for procedures to protest this BAA. As further prescribed in FAR 33.106(b) and FAR 52.233-3, protests regarding the selection decision should be submitted to DARPA Contracts Management Office, 675 N. Randolph Street, Arlington, VA 22203, by email to CMO_SBIRProtests@darpa.mil and sbir@darpa.mil.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
Technical question deadline: October 14, 2026, to SBIR_BAA@darpa.mil with the topic number in the subject line
Proposal deadline: October 21, 2026. DARPA will not accept late proposals
Selection notification: within 90 calendar days of BAA close
Base period: 18 months from award
Option period: 6 additional months if exercised
A working backward plan
Before September 23. Download the DARPA Volume 2 and Volume 3 templates from the DARPA Small Business website, since the cost template is mandatory. Read the consolidated FAQ for this topic and keep checking it. Assemble your feasibility package: empirical flight data with 3D tomographic reconstructions, SNR calculations, and imagery showing detection of underground mineral deposits; receiver thermal noise floor documentation; and the analytical scaling models mapping the architectural transition. Decide your contract type and, if cost-plus, confirm you hold a current DCMA Final Determination Letter; if OT, download and complete the Model OT and OT Certifications. Confirm SAM registration is active and your CMMC Level 2 self-assessment is current in SPRS. Identify your flight test platform and the organization that will grant flight clearance for high-power RF emissions, because the Month 12 gate depends on it. Resolve export control staffing. If venture-backed, register with the SBA Company Registry and obtain the SBIR VC Certification. Submit technical questions well before October 14.
September 23 through October 5. Draft the technical volume in three parts against the 35 page structure: 10 pages of feasibility documentation, 20 pages of technical proposal built around the five key technical challenges, and 5 pages of commercialization strategy using DARPA's own MRI for the Earth and 1-in-200 framing. Confront the 256x power scaling number explicitly and show your path, whether through amplifier power, processing gain, waveform design, or a combination.
October 6 through October 14. Build the cost volume in the mandatory template across the 18 month base and 6 month option, aligned to the fixed payable milestones. Get supplier quotes for the amplifier, antenna structure, thermal components, and FPGA hardware. Price the flight campaign realistically, including platform time and test organization support.
October 15 through October 18. Assemble Volume 5 with your contract-type documents, complete Volume 7 and the Volume 4 CCR, and run compliance: 35 page limit with the three internal section limits observed, unclassified or CUI only, no proprietary information on the coversheet, mandatory cost template used.
October 19 through October 20. Submit and certify in DSIP. October 21 is the deadline, not the plan.
Frequently Asked Questions
What is DARPA SBIR topic DPA26BZ06-DV023?
DPA26BZ06-DV023 is a DARPA SBIR Direct to Phase II topic titled "Novel Radio Frequency Sensing Technologies," released under the DoW 2026 SBIR Broad Agency Announcement, Release 6. It seeks development, integration, and flight test of a high-altitude, approximately 40,000 foot, high-speed, approximately 400 knot, airborne geologic sensing system providing rapid stand-off 3D subsurface tomographic imaging to detect subterranean features such as metals and critical minerals.
How much funding is available?
$700,000 for an 18 month base period, plus a $500,000 option over 6 months, for a total of $1,200,000 across 24 months if the option is exercised. There is no second option on this topic. Up to $25,000 in Technical and Business Assistance may be requested in addition to the cost ceiling.
When is the proposal deadline?
October 21, 2026. DARPA states it will not accept late proposals. Note also the separate technical question deadline of October 14, 2026.
When does this topic open?
September 23, 2026, giving a 29 day submission window.
Can I ask questions through DSIP Topic Q&A?
No. DARPA states that DSIP Topic Q&A will not be available for these DARPA topics. Technical questions must be emailed to SBIR_BAA@darpa.mil with the topic number in the subject line by October 14, 2026, and must include the name, email address, and telephone number of a point of contact. Questions submitted within seven calendar days of the proposal due date may not be answered. DARPA posts a consolidated FAQ on its Small Business site, updated until one week prior to the due date.
Can I submit a Phase I proposal?
No. This topic is soliciting Direct to Phase II proposals only, and proposers must submit feasibility documentation in lieu of a Phase I proposal.
What feasibility documentation is required?
You must demonstrate you have already developed, integrated, and validated an airborne subsurface radar or similar system in a physical environment, for example low-altitude, low-speed flight up to 10,000 feet. Modeling and simulation data alone is not sufficient. The documentation must include three things: empirical flight data with representative 3D tomographic subsurface reconstructions, signal-to-noise ratio calculations, and imagery demonstrating successful detection of underground mineral deposits; receiver characterization proving the receiver operates at a low thermal noise floor; and analytical scaling models validating the R4 scaling equations and mapping the exact architectural transition from a low-power baseline to a high-power, high-altitude design.
What is the R4 scaling problem?
According to the radar equations for small targets of a radar cross section of approximately one wavelength, necessary transmit power scales as range to the fourth power. Moving from a 10,000 foot baseline to 40,000 feet is a 4x increase in distance, nominally demanding a 256x increase in radiated power to maintain equivalent subsurface imaging performance. That number defines the engineering problem and your proposal must address it directly.
What are the five key technical challenges?
A highly reliable kW-class RF power amplifier. An externally deployable or structurally integrated antenna system capable of withstanding aerodynamic loads at approximately 400 knots and 40,000 feet. Passive or active high-capacity heat dissipation to mitigate thermal loads under sustained high-power transmit cycles. Advanced RF shielding and electrical isolation to guarantee zero electromagnetic interference with the host aircraft's flight navigation, control, and mission avionics. And FPGA selection and processing techniques handling increased signal propagation delays, modified waveforms, and real-time tomographic data processing at high standoff distances.
Can the payload draw power or cooling from the aircraft?
No. The entire system must be designed as a fully self-contained payload, utilizing independent internal power generation and an edge computational architecture, to facilitate rapid modular deployment across diverse airframes without requiring permanent, costly aircraft modifications.
How long can my technical volume be?
35 pages in the Standard Proposal Format, structured as up to 10 pages of DP2 feasibility documentation, up to 20 pages of DP2 technical proposal, and up to 5 pages of Phase II commercialization strategy, which should be the last section. Templates are provided on the DARPA Small Business website.
Is the cost proposal template mandatory?
Yes. DARPA provides templates for Volume 2 and Volume 3 as attachments on the DARPA Small Business website, and states that use of the DARPA Cost Proposal template is mandatory.
Do I have to choose a contract type?
Yes. Proposers must state their requested contract type. DARPA may award FAR-based firm-fixed-price or cost-plus reimbursement contracts, or Other Transactions for Prototype under 10 U.S.C. 4021. A cost-plus request requires your DCMA Final Determination Letter showing accounting system approval. An OT request requires a completed Model OT plus OT Certifications, loaded in Volume 5. A firm-fixed-price request requires no additional action.
What are the Phase II milestones?
Fixed payable milestones across an 18 month base: Month 2 subsystem engineering preliminary designs, Month 4 subsystem design completion, Month 6 structural antenna engineering interim report with benchtop testing at approximately 400 knot loads, Month 9 FPGA and compute development interim report with initial edge tomographic reconstruction results, Month 12 Safety and Test Readiness Review data package with high-voltage and high-power RF emissions safety results and flight clearances documentation, Month 15 high-altitude flight campaign interim report with preliminary data at approximately 40,000 feet and 400 knots, and Month 18 data analysis and final base period report with fully reconstructed 3D subsurface tomograms. The 6 month option adds Month 21 interim data analysis and Month 24 comprehensive final system assessment against a low-altitude baseline plus commercial viability analysis.
What is the biggest schedule risk?
The Month 12 Safety and Test Readiness Review, which requires flight clearances documentation for high-voltage and high-power RF emissions. Airborne flight clearance for high-power RF is an approval process with its own timeline independent of your engineering progress. Name your platform and your test organization, and describe the clearance pathway.
Is this topic export controlled?
Yes. The technology is restricted under ITAR 22 CFR Parts 120-130 and EAR 15 CFR Parts 730-774. You must disclose any proposed use of foreign nationals with countries of origin, visa or work permit type, and the statement of work tasks intended for them, and the topic advises those individuals may be restricted from performing. The projected CMMC requirement is Level 2 with self-assessment, and DARPA notes that firms engaging in ITAR or CUI work for DARPA must have Level 2 certification.
Are venture capital backed companies eligible?
Yes. Proposers more than 50 percent owned by multiple venture capital operating companies, hedge funds, private equity firms, or any combination as set forth in 13 CFR 121.702 are eligible for DARPA topics under this BAA, subject to three conditions: register with the SBA Company Registry Database before submitting, submit the Majority-Owned VCOC, HF, and PEF Certification in Volume 5, and immediately notify the Contracting Officer if you enter that ownership class after submitting but before receiving a funding agreement.
How will my proposal be evaluated?
Against the evaluation criteria in the DoW SBIR Program BAA. DARPA evaluates each conforming proposal in its entirety, documenting strengths and weaknesses relative to each criterion, then determines overall selectability. Proposals are not evaluated against each other but on their own individual merit. A selectable proposal is one where strengths outweigh weaknesses with no accumulated weaknesses requiring extensive negotiations or resubmission. A non-selectable proposal is one where strengths do not outweigh weaknesses.
Will I get feedback if not selected?
Yes, and this is more generous than most components. In accordance with the SBA SBIR/STTR Policy Directive Appendix I paragraph 4 subparagraph (d), DARPA will provide a technical evaluation narrative to the proposer for each proposal submitted in response to a topic. An informal feedback session may additionally be requested via email at sbir@darpa.mil, provided at the sole discretion of DARPA.
Does the Company Commercialization Report affect my score?
No. Completing the CCR as Volume 4 in DSIP is required, but DARPA will not consider information contained in the CCR during proposal evaluations.
What support comes with a DP2 award?
Two programs. The Transition and Commercialization Support Program provides services to Phase II and DP2 awardees upon contract execution at no cost to awardees, aimed at moving technology beyond Phase II into other R&D programs, DoW acquisition programs, other federal programs, or the commercial market. Separately, awardees may be eligible for the Embedded Entrepreneurship Initiative, an invitation-only program at DARPA's sole discretion, typically no more than $310,000 per awardee, funding a Senior Commercialization Advisor relationship, investor working group connections, and hiring an embedded entrepreneur for a Go-to-Market strategy.
What is the commercial market?
DARPA calls the system a highly scalable "MRI for the Earth," able to map deep-seated geologic structures to rapidly locate critical mineral reserves and rare earth element deposits buried under hundreds of meters of geology, significantly accelerating discovery timelines, increasing exploratory extraction success rates from 1-in-200 to near certainty, and optimizing geologic risk management. On the defense side, identifying additional domestic sources of critical minerals is framed as a national security imperative strengthening the U.S. critical minerals supply chain.
What reference does the topic cite?
One: DoD Joint Sensor Integration Standards, "Electromagnetic Compatibility and RF Shielding Guidelines for High-Power Airborne Payloads (MIL-STD-461G)," 2019. That the sole reference is the EMC standard tells you how seriously DARPA takes the zero-interference requirement.
Who do I contact with questions?
Specific questions on the administration of the DARPA Program and these proposal preparation instructions go to the DARPA Small Business Programs Office at SBIR_BAA@darpa.mil. Technical questions related to research objectives and awards specifically related to a topic also go to SBIR_BAA@darpa.mil with the topic number in the subject line. DSIP technical support is available Monday through Friday, 9:00 a.m. to 5:00 p.m. ET, at DoDSBIRSupport@reisystems.com with a copy to SBIR_BAA@darpa.mil. DARPA also offers free resources through DARPAConnect at DARPAConnect.us, including "Tips for DARPA Proposal Success."
Positioning Advice for Companies Considering This Topic
Qualify yourself honestly on the feasibility bar. You need a flown airborne subsurface radar with empirical tomographic results, documented receiver noise floor, and validated scaling models. Simulation is explicitly insufficient. There are not many companies in the world that clear this bar, which is good news if you are one of them and a reason to look elsewhere if you are not.
Make the scaling model the centerpiece. Empirical flight data proves you have a system. The analytical scaling models prove you can build this one. DARPA asks for models "mapping the exact architectural transition from a low-power baseline to a high-power, high-altitude design," which is a request for engineering specificity, not a narrative. Show the power budget, the link budget, the loss terms, and where you buy back margin.
Do not concede the full 256x. The honest reading of the topic is that DARPA expects proposers to attack the number from several directions: raw amplifier power, aperture gain, coherent integration, waveform design, and processing. A proposal that simply proposes a 256x more powerful transmitter is unlikely to close on SWaP, thermal, or cost. Show the decomposition.
Treat EMC as a design driver. The requirement is zero electromagnetic interference with flight navigation, control, and mission avionics, and the only cited reference is MIL-STD-461G. On a self-contained kW-class payload strapped to a manned aircraft, this is what stands between you and a flight clearance.
Name the aircraft and the clearance path. The Month 12 milestone requires flight clearances documentation, and the Month 15 milestone requires demonstration flights. Airworthiness for a high-power RF payload is not a formality. Identifying your platform, your test organization, and your safety-of-flight approach converts the highest schedule risk in the program into a managed item.
Address thermal honestly. Sustained high-power transmit cycles at altitude, in a self-contained payload, with no aircraft cooling, is a genuinely hard thermal problem, and it interacts with the power amplifier choice and the antenna structure. Proposals that gloss this tend to have an unstated assumption of low duty cycle. If your duty cycle is low, say so and show what that costs in area coverage rate.
Use DARPA's own commercial numbers. The MRI for the Earth framing and the 1-in-200 to near certainty claim come from the topic itself. Your five page commercialization strategy is stronger citing DARPA's framing and adding the market sizing than inventing a different story.
Choose your contract instrument deliberately. For a milestone-driven hardware program, the Other Transaction for Prototype has real advantages and DARPA describes them favorably. But it requires a completed Model OT with redlines in your proposal, which is legal work you need to start early. Firm-fixed-price is the low-friction path. Cost-plus requires a DCMA-approved accounting system you either have or do not.
Write to eliminate weaknesses. DARPA's selectability definition turns on whether accumulated weaknesses would require extensive negotiations or resubmission. Unlike a scored competition, a single unresolved weakness can move you from selectable to non-selectable. Have someone hostile read the proposal looking for gaps, not for polish.
Submit your technical questions by October 14 and read the FAQ. With no DSIP Q&A available, the DARPA FAQ is the only public clarification channel, and it is updated until a week before the deadline. Competitors' questions will surface requirements you had not considered.
DAF STTR DAF26TZ06-NV007: Automated Combat Assessment at the Edge
Deadline: October 21, 2026
Funding Award Size: $300k
Description: Complete guide to DAF STTR Phase I topic DAF26TZ06-NV007, automated combat assessment at the edge. Up to $300,000 over 6 months, STTR partner required. Closes October 21, 2026.
Quick Answer
DAF26TZ06-NV007 is a Department of the Air Force STTR Phase I topic under the DAF 2026 STTR Broad Agency Announcement, Release 6. The objective is a single sentence: develop a real time automated combat assessment at the edge system. The problem it solves is a timeline problem. Today, battle damage assessment arrives hours after mission execution as an after-action report built from multiple imagery collects, assembled through a sequence of independent manual tasks. The Air Force states plainly that this does not meet timeline requirements for prosecuting advanced enemy threat systems, because critical missions require battle damage assessment immediately post-strike to inform re-strike planning. Awards are up to $300,000 for a period of performance up to 6 months, with a 20 page technical volume limit. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
Two eligibility gates come first. This document states that small businesses majority-owned in part by multiple venture capital operating companies, hedge funds, or private equity funds are not eligible. And as an STTR topic it requires a research institution partner performing a minimum of 30 percent of the effort, with the small business conducting a minimum of 40 percent, and the DAF states it will not consider deviation requests.
Topic At a Glance
Topic number: DAF26TZ06-NV007
Title: Automated Combat Assessment at the Edge
Solicitation: Department of the Air Force 2026 Small Business Technology Transfer (STTR) Broad Agency Announcement (BAA), Release 6, Phase I Proposal Submission Instructions
Program: STTR, requiring a single partnering research institution
Program type: Phase I
Award maximum value: $300,000
Award maximum duration: 6 months
Technical volume page limit: 20 pages or slides
OUSD (R&E) Critical Technology Area: Applied Artificial Intelligence (AAI)
Component Technology Priority Area: Trusted AI and Autonomy
Projected CMMC level requirement: Level 2 (Self)
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic
Venture capital, hedge fund, private equity majority ownership: not eligible under this document
Work split: minimum 40 percent by the small business, minimum 30 percent by the single partnering research institution, for both Phase I and Phase II. Deviations will not be considered
Named standards: CJCSI 3162.02 for damage classification, Intelligence Community Directive ICD 302 for confidence intervals
Technical and Business Assistance: up to $6,500 per Phase I award, in addition to the per-topic total
Type size: no smaller than 10-point
Evaluation criteria: not restated in this document. See the FY26 STTR BAA
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026, at the time stated in the FY26 STTR BAA
Selection timeline: within approximately 90 calendar days of solicitation close, meaning on or about January 19, 2027
Submission portal: DSIP at dodsbirsttr.mil
Keywords: combat assessment, physical damage, functional damage, machine learning, edge processing, testing, evaluation
Note on the deadline time. The DAF instructions direct offerors to the FY26 STTR BAA for the proposal submission deadline date and time. DoW deadlines are ordinarily 12:00 p.m. Eastern on the close date. Confirm the exact time on the live DSIP posting rather than assuming end of day.
The Two Eligibility Gates
Venture capital ownership is disqualifying under this document
The instructions state that small business concerns owned in majority part by multiple venture capital operating companies, hedge funds, or private equity funds are not eligible to submit applications or receive awards for Department of Air Force Topics.
Because this provision is not uniform across the DAF Release 6 component instructions, here is where it actually stands across all seven documents.
Not eligible: the SBIR BAA Direct to Phase II instructions covering topic DV037; the STTR BAA Direct to Phase II instructions covering topic DV008; and this STTR BAA Phase I document covering topics NV006 and NV007.
Eligible: the SBIR BAA Phase I instructions covering topics NV034 through NV036; the SBIR CSO Direct to Phase II instructions; the SBIR CSO Phase I instructions; and the STTR CSO Phase I instructions.
Every Commercial Solutions Opening document permits venture-backed firms. Three of the four Broad Agency Announcement documents exclude them, the SBIR BAA Phase I instructions being the exception. The DAF has not explained the variation. What matters is that the exclusion appears in the document governing this topic, so treat it as operative, verify your status against 13 CFR 121.702, and raise it with the DAF SBIR/STTR One Help Desk during the open period rather than assuming it resolves in your favor.
The STTR work split, and no deviations
For both Phase I and Phase II, a minimum of 40 percent of each STTR award must be conducted by the awardee and a minimum of 30 percent of the effort must be performed by the single partnering research institution. Applicants and awardees may partner with multiple entities that separately meet the definition of a research institution as indicated in the STTR BAA. Applicants may use only one partnering research institution to meet STTR eligibility requirements. The DAF will not consider requests for deviations to these performance of work requirements.
The Subcontracts section restates the same figures and adds that deviations from these performance of work requirements are not permitted. The STTR funded work percentage calculation considers both direct and indirect costs after removal of the small business concern's proposed profit. The instructions also point to the DoW STTR FY26 BAA for information regarding the required Allocation of Rights Agreement.
On a $300,000 award, that means at least roughly $90,000 flowing to your research institution partner and at least roughly $120,000 staying with you. There is no deviation process available here, unlike several other DAF Release 6 documents, so the split has to work as designed.
This document is also the clearest statement of the STTR work requirement anywhere in the DAF Release 6 instructions. The STTR CSO Phase I document states a two-thirds in-house floor and the STTR BAA Direct to Phase II document states 50 percent plus a two-thirds figure labeled Phase I, neither of which matches the statutory STTR structure. If you are working a topic governed by one of those, this document is useful evidence of what the requirement actually is, and the discrepancy is worth a Help Desk question.
What the Air Force Is Actually Looking For
The timeline problem
Combat assessment, both physical and functional, provides tactically relevant information for re-strike and re-planning.
The state of practice for traditional targeting involves generating after-action reports including battle damage assessment hours after mission execution, relying on multiple imagery collects, specifically electro-optical. This analysis is performed as a sequence of independent tasks that are outdated, time consuming, and lack automation.
Then the consequence, stated bluntly. This does not meet timeline requirements for prosecution of advanced enemy threat systems, as critical missions require battle damage assessment immediately post-strike, to inform planning for re-strike and to drive potential changes to operations.
So the deliverable value is measured in time. Hours to minutes, or minutes to seconds, depending on the phase of the problem you attack. Your proposal should quantify the timeline improvement you expect, because that is the currency this topic trades in.
The three named barriers
Despite advancements in technology, challenges still remain for speeding processes.
The time-consuming and often incomplete nature of traditional damage assessment methods.
Limitations in remote sensing technologies, hindering data collection especially in remote or difficult-to-access areas.
The lack of real time pre- and post-event satellite imagery for model training.
That third barrier is worth pausing on, because it is a data problem rather than an algorithm problem, and it shapes what a six month Phase I can honestly claim. If real time pre- and post-event satellite imagery is not available for training, your proposal needs a credible answer about what you will train on. The topic's own invitation to propose novel or under-utilized data modalities is partly a response to this constraint.
The three operational realities the system must follow
The topic seeks the development of a system for the automation of damage assessment following three operational realities. Treat these as the three pillars of your technical approach.
Edge Processing Software. Software programs and data-sharing instructions designed to run directly on physical field equipment such as sensors and unmanned aerial vehicles. This covers multi-sensor, multi-modal data collection and processing, damage classification and assessment at the edge in near real time, and information sharing protocols, all under low size, weight, and power and contested environment constraints.
Note the compound constraint: multi-sensor and multi-modal, at the edge, in near real time, under low SWaP, in contested environments. Each of those individually is manageable. Together they are the hard part, and a proposal that addresses only the classification accuracy question has answered the easy third of the problem.
AI and ML Software Model Development. AI and ML software designed to automatically classify military damage. For damage classification, both physical and functional, aligned to CJCSI 3162.02 for DoD customers. Confidence intervals aligned to Intelligence Community Directive ICD 302. And performance metrics for accuracy, computation time, and sensors density and location.
Three specifics there. CJCSI 3162.02 is the Chairman of the Joint Chiefs of Staff Instruction on combat assessment methodology, and aligning your damage classification taxonomy to it is a stated requirement rather than a suggestion. ICD 302 governs confidence expression, which means your model outputs need calibrated, doctrinally expressed confidence rather than raw scores. And the three performance metrics are accuracy, computation time, and sensor density and location, which together mean the Air Force wants to know not just how well you classify but how fast and with how much sensing.
A note on the references. The topic body cites ICD 302 for confidence intervals, while the reference list cites ICD 203, the Analytic Standards Technical Amendment. ICD 203 is the directive most commonly associated with analytic confidence and probabilistic language standards. We cannot resolve which the author intended, so read both and address analytic confidence standards explicitly, and consider asking the DAF SBIR/STTR One Help Desk during the open period.
Robust Testing and Evaluation. To ensure the operational effectiveness, suitability, and survivability of the aforementioned technologies, testing and evaluation would be conducted in a phased approach. Beginning with laboratory-based testing to verify basic functionality and performance, followed by increasingly complex testing in simulated and field environments. The goal is to ensure that the technologies developed for combat assessment are operationally effective, suitable, and can be confidently deployed in operationally relevant environments.
Operational effectiveness, suitability, and survivability are formal test and evaluation terms. Their presence signals that the Air Force expects a real T&E plan, phased from laboratory through simulated to field, not an ad hoc validation approach.
Phase I
Phase I will focus on the analysis and design of an automated combat assessment at the edge system for re-strike, described as a particularly targeted asset, and re-attack recommendations, with preliminary laboratory model testing.
Awardees must demonstrate an innovative approach to physical and functional combat assessment, and novel or under-utilized data modalities.
The final report must detail feasibility study findings and a Phase II plan.
Four things to draw out. Phase I is analysis and design plus preliminary laboratory model testing, not a working system. The scope is re-strike and re-attack recommendations, which is narrower than combat assessment in general and gives you a bounded use case. Physical and functional assessment are both required, and they are genuinely different problems: physical damage is what the structure looks like, functional damage is whether the target can still do its job. And novel or under-utilized data modalities is an explicit requirement, which pairs with the stated barrier around imagery availability.
That last requirement is the clearest differentiation opportunity in the topic. The state of practice relies on electro-optical imagery. If your approach brings synthetic aperture radar, radio frequency signatures, acoustic sensing, infrared, multispectral, passive collection, or fused combinations, that is responsive to a stated requirement rather than a nice extra.
Phase II
Develop, implement, test, evaluate, and demonstrate a distributed automated damage assessment system at the edge that follows the criteria in the Description and Phase I sections of this proposal. The final report must detail all criteria implementations and a Phase III plan.
Phase II performance will focus on the development, integration, and operational evaluation of a functional prototype software system for distributed, edge-based automated combat assessment. Awardees will transition Phase I designs into working software models running on representative edge hardware, culminating in a demonstration within a simulated or operationally relevant environment.
Required deliverables may include three items. Functional prototype software and compiled AI and ML model weights. Comprehensive system architecture and interface control documentation. And a Phase II final report detailing test results, performance metrics, criteria implementations, and a commercialization and transition plan for Phase III production and deployment.
Note that compiled model weights are named as a deliverable. That has intellectual property implications worth thinking through in Phase I, particularly on an STTR award where a research institution partner may contribute to model development and where the Government obtains SBIR/STTR data rights for 20 years in what is developed under the contract.
Note also the words distributed and representative edge hardware. Phase II runs on real edge devices, not a workstation, and it is distributed rather than centralized. Your Phase I architecture should be designed for that from the start.
Phase III
The awardee will identify and pursue commercial applications, including urban surveillance, disaster and humanitarian response, and other use cases.
That is a short Phase III section but the two named applications are genuinely adjacent. Disaster and humanitarian response in particular is a strong dual-use fit: rapid automated damage assessment from multi-modal sensing at the edge is exactly what post-earthquake, post-hurricane, and post-flood assessment requires, and the customer set includes FEMA, international relief organizations, insurers, and infrastructure operators. Build the commercialization case there rather than straining for something more exotic.
The references
Three are cited, and all three are worth obtaining before you write.
"Analytic Standards, Intelligence Community Directive 203 Technical Amendment," Office of the Director of National Intelligence, 2023, available at dni.gov.
"Methodology for Combat Assessment," United States Joint Chiefs of Staff, 2021, which is CJCSI 3162.02A, available at jcs.mil.
Broad Agency Announcement title: Targeting Operations and Analytics Development, BAA FA8750-25-S-7002, available at sam.gov.
The third reference is the most interesting and the least obvious. FA8750 is the Air Force Research Laboratory Information Directorate contracting office prefix, and a live BAA on Targeting Operations and Analytics Development tells you there is an existing AFRL program area in this space. Reading that BAA tells you who the government is already working with and what language the program office uses. That is competitive intelligence available for free.
CJCSI 3162.02 is not optional reading. Your damage classification taxonomy must align to it, and the terms physical damage assessment, functional damage assessment, and target system assessment come from that doctrine.
Funding Allowance and Cost Structure
Award ceiling
Up to $300,000 across up to 6 months. The topic index states that proposals in excess of this amount will not be considered for evaluation or award, and proposals in excess of this duration will not be considered for evaluation or award.
The instructions add that any per-award or per-topic funding caps are budgetary estimates only and more or less funding may become available. Multiple procurements are planned and anticipated to be awarded as a result of the topic, each proposal is considered a separate procurement and will be evaluated on its own merit, and the Government may award all, some, or none of the proposals. Funding decisions are made with complete disregard to the other awards under the same topic.
With the mandatory 30 percent minimum to your research institution partner, roughly $90,000 or more leaves your firm. Six months of a small team plus an academic effort supports analysis, architecture, and preliminary laboratory model testing, which is what Phase I asks for. It does not support building a distributed edge system, and Phase I does not ask you to.
Contract type
The DAF primarily makes STTR Phase I and Phase II awards as firm-fixed-price contracts. Awardees are strongly urged to work toward a Defense Contract Audit Agency approved accounting system, since if the company intends to continue work with the Department of War an approved accounting system allows competition in a broader array of acquisition opportunities, including award of cost-reimbursement type contracts.
If no exceptions are taken to an offeror's proposal, the Government may award a contract without negotiations. Therefore the offeror's initial proposal should contain the offeror's best terms from a cost or price and technical standpoint. If there are questions regarding the award document, contact the Phase I Contracting Officer identified on the cover page. The Government reserves the right to reopen negotiations later if the Contracting Officer determines doing so to be necessary.
Technical and Business Assistance
The Small Business Innovation and Economic Security Act Section 7 mandates agencies to offer TABA. The DAF will provide up to $6,500 per Phase I award, and this total is in addition to the per-topic total identified in the DSIP Volume 3 Cost Proposal.
Awardees can only use TABA funding for the activities outlined in 15 U.S.C. 638(q)(1) and the purposes outlined in 15 U.S.C. 638(q)(1)(A) through (E). Eligible activities include access to a network of scientists and engineers engaged in a wide range of technologies, assistance with product sales, intellectual property protections, cybersecurity assistance, market research, market validation, development of regulations and manufacturing plans, and access to technical and business literature available through online databases. Those activities can be undertaken in furtherance of making better technical decisions concerning such projects, solving technical problems which arise during the conduct of such projects, minimizing technical risks associated with such projects, developing and commercializing new commercial products and processes resulting from such projects including intellectual property protections, and screening for potential foreign involvement in technology development or commercialization activities.
Given that compiled model weights are a Phase II deliverable and an Allocation of Rights Agreement with your research institution partner is required, intellectual property protections is the most directly useful eligible activity here.
Requests for TABA funding creditable to a TABA provider must include the following or will be subject to denial: the TABA providers; the providers' point of contact, email address, and phone number; an explanation of the provider's unique qualifications to provide the TABA service; the tasks that will be performed by the provider including the purpose and objective of the assistance, with the task milestone list tracking to the milestone payment schedule otherwise provided by the applicant; and total provider cost, number of hours, and labor rates, with average or blended rates acceptable.
If proposing TABA funding to hire new staff, augment staff, or direct staff to conduct or participate in training activities consistent with the purpose of TABA, the following must be included or may be subject to denial: names and positions, business need to be filled or training to be provided; number of employees to be hired, augmented, or directed to participate in training activities; qualifications of employees hired or augmented, or detailed need for training; tasks that will be performed including both a description of the activity and the purpose that it will serve; and total staff or training cost, number of hours, and labor rates.
A detailed request for TABA funding must be included in the Volume 5 Supporting Documents in DSIP. TABA requests that only specify a TABA request value in the Volume 3 Cost Proposal will not be considered.
Cost volume requirements
Cost information should be provided by completing the Cost Volume in DSIP and including the Cost Volume Itemized Listing. The Cost Volume detail must be adequate to enable Air Force personnel to determine the purpose, necessity, and reasonableness of each cost element. The DSIP Cost Volume and Itemized Cost Volume Information will not count against the specified page limit, and the itemized listing may also be submitted in Volume 5 under the "Other" dropdown option.
Direct cost materials. Justify costs for materials, parts, and supplies with an itemized list containing types, quantities, prices, and where appropriate purpose. On this topic, representative edge compute hardware for laboratory model testing belongs here.
Other direct costs. This category includes but is not limited to specialized services such as machining, milling, special testing or analysis, and costs incurred in temporarily using specialized equipment. Proposals including leased hardware must include an adequate lease versus purchase justification. Commercial satellite or aerial imagery purchases, and compute resources for model training, plausibly sit here and may be significant given the data barrier the topic names.
Direct labor. Identify key personnel by name, if possible, or by labor category, if not. Direct labor hours, labor overhead or fringe benefits, and actual hourly rates for each individual are necessary for the Contracting Officer to determine whether these hours, fringe rates, and hourly rates are fair and reasonable.
Travel. Travel costs must relate to project needs. Break out travel costs by trip, number of travelers, airfare, per diem, lodging, and similar. The number of trips required, as well as the destination and purpose of each, should be reflected. The instructions recommend budgeting at least one trip to the Air Force location managing the contract.
Subcontracts. Involvement of a consultant in the project's planning or research stages may be appropriate. If so, describe in detail and include information in the Cost Volume. A minimum of 40 percent of each STTR project must be conducted by the small business concern and a minimum of 30 percent of the effort performed by the single partnering research institution. Deviations from these performance of work requirements are not permitted. The STTR funded work percentage calculation considers both direct and indirect costs after removal of the small business concern's proposed profit. Support subcontract costs with copies of executed agreements that adequately describe the work to be performed, and at a minimum include a Statement of Work with a corresponding detailed Cost Volume for each planned subcontract. Additionally, see the DoW STTR FY26 BAA for more information regarding the required Allocation of Rights Agreement.
Special tooling, special test equipment, and material. The inclusion of equipment and materials will be carefully reviewed relative to need and appropriateness to the work proposed. Special tooling and special test equipment purchases must, in the Contracting Officer's opinion, be advantageous to the Government and relate directly to the effort, and should not be of a type that an offeror would otherwise possess in the normal course of business.
Consultants. Provide a separate agreement letter for each consultant, briefly stating what service or assistance will be provided, the number of hours required, and the hourly rate.
Where work must be performed
All R/R&D work must be performed in the United States. Based on a rare and unique circumstance, the DAF may approve a particular portion of the R/R&D work to be performed or obtained in a country outside the United States. The awarding Funding Agreement officer must approve each specific condition in writing. Applicants seeking this approval must make the request with their initial proposal submission, and the DAF will not consider these requests prior to proposal submission.
Personnel Disclosure
This topic does not carry a topic-level ITAR and EAR restriction paragraph. That said, targeting and combat assessment is sensitive subject matter, and your disclosure obligations apply in full.
Identify in the Technical Volume all key personnel who will be involved in this project, including information on directly related education, experience, and citizenship. A technical resume of the principal investigator, including a list of publications if any, must be included, and only one principal investigator or project manager can be designated to a proposal at any given time. Concise technical resumes for subcontractors and consultants are also useful.
Identify all U.S. permanent residents to be involved in the project as direct employees, subcontractors, or consultants. Identify all non-U.S. citizens expected to be involved in the project as direct employees, subcontractors, or consultants. For all non-U.S. citizens, in addition to technical resumes, provide countries of origin, the type of visa or work permit under which they are performing, and an explanation of their anticipated level of involvement on this project, as appropriate. Additional information may be requested during negotiations in order to verify the foreign citizen's eligibility to participate on a contract issued as a result of this announcement.
Do not upload information such as Permanent Resident Cards, birth certificates, Social Security Numbers, or other personally identifiable information to the DSIP system. Provide the categories of information requested, not the underlying identity documents.
On an STTR award this reaches your research institution partner's personnel, and computer vision and machine learning groups are among the most internationally staffed in engineering. Request a complete roster with citizenship and visa status from your partner early. Note also that foreign affiliation with a research institution in a country of concern is a mandatory award denial under the foreign risk provisions below.
Proposal Structure: The Seven Volumes
Formatting, and what does not count against the page limit
The Technical Volume should include all graphics and attachments but should not include the Cover Sheet, which is completed separately as Volume 1. Ensure that all graphics are distinguishable in black and white, which matters here since you will want to show imagery examples, classification outputs, and architecture diagrams.
The Technical Volume must be no smaller than 10-point on standard 8.5 by 11 inch paper with one-inch margins. This differs from the DAF Direct to Phase II instructions, which require 11-point.
The Phase I Technical Volume page and slide limits identified for the topics do not include the Cover Sheet, the Cost Volume, or the Cost Volume Itemized Listing. Only the Technical Volume and any enclosures or attachments count toward the page limit. The documents required for upload into Volume 5 under "Other" do not count toward the specified limits. In the interest of equity, pages or slides in excess of the stated limits will not be reviewed.
The limit for this topic is 20 pages or slides, and because it is expressed as pages or slides a slide-format volume is acceptable.
Note that this document, unlike the STTR CSO Phase I instructions in the same release, does not require a government proposal template.
Fraud, Waste and Abuse training must be completed prior to proposal submission, and DSIP will indicate completion of the Volume 6 requirement once the training is complete and certified.
Volume 1, Cover Sheet
Complete the proposal Cover Sheet in accordance with the instructions provided via DSIP. The technical abstract should include a brief description of the program objectives, a description of the effort, anticipated benefits and commercial applications of the proposed research, and a list of keywords and terms.
The technical abstract of each successful proposal will be submitted to the Office of the Secretary of War for publication and therefore must not contain proprietary or classified information. If selected for funding, the proposal's technical abstract and discussion of anticipated benefits will be publicly released. On a targeting-adjacent topic, be deliberate about abstract content.
Volume 2, Technical Volume
The Phase I technical volume shall contain the required elements below. These instructions supplement the FY26 STTR BAA, and in addition to the requirements found in the BAA, applicants are required to provide the following information in Volume 2.
Key personnel, as described above.
Phase I Statement of Work outline. The DAF uses the work plan outline as the initial draft of the Phase I Statement of Work. Therefore, do not include proprietary information in the work plan outline. To do so will necessitate a request for revision, if selected, and may delay award.
Include a work plan outline in the following format. Scope, listing the effort's major requirements and specifications. Task Outline, providing a brief outline of the work to be accomplished during the Phase I effort. Milestone Schedule. Deliverables. Progress reports. Final report with SF 298.
The SF 298 is the Report Documentation Page that accompanies a technical report. It is named in the required outline, so plan for it. Note that on this topic the final report is a substantive deliverable: it must detail feasibility study findings and a Phase II plan.
Volume 3, Cost Volume
Covered above.
Volume 4, Company Commercialization Report
Completion of the CCR as Volume 4 of the proposal submission in DSIP is required. Refer to the FY26 STTR BAA for full details. Information contained in the CCR will not be considered by the Air Force during proposal evaluations.
Volume 5, Supporting Documents
Three documents may be required if applicable to your proposal. DD Form 2345, for proposals submitted under export-controlled topics. Verification of Eligibility of Small Business Joint Ventures, Attachment 3 to the FY26 STTR BAA. Technical Data Rights Assertions, if asserting data rights restrictions.
Technical Data Rights Assertions matter here more than on most topics, because compiled AI and ML model weights are a named Phase II deliverable and because an STTR structure puts a research institution's contributions in the same codebase as yours. Pre-existing models, training pipelines, and architectures are background intellectual property, and this is the mechanism for saying so.
The detailed TABA request, if requesting TABA, also goes in Volume 5. Documents uploaded into Volume 5 under "Other" do not count toward the page limit, and the Cost Volume Itemized Listing may also be submitted there.
Volume 6, Fraud, Waste and Abuse Training
Material can be found in the Volume 6 section of the proposal submission module in DSIP and must be thoroughly reviewed once per year to proceed with proposal submission.
Volume 7, Disclosures of Foreign Affiliations or Relationships to Foreign Countries
Small business concerns must complete the webform in Volume 7 of the DSIP proposal submission. The disclosures will not be accepted as a PDF Supporting Document in Volume 5, and previous versions of this form must not be uploaded to Volume 5. For additional details, refer to the STTR Program BAA.
How Your Proposal Will Be Evaluated
The criteria live in the BAA
Proposals will be evaluated for overall merit in accordance with the criteria discussed in the FY26 STTR BAA. The DAF is seeking varying technical and scientific approaches and varying and new technologies that would be responsive to the problem statements and areas of interest in the topic.
This document does not reproduce the evaluation criteria, unlike the DAF CSO instructions which restate Criteria A, B, and C with their relative importance. Read the criteria in the FY26 STTR BAA before you begin writing.
Foreign risk evaluation
15 U.S.C. 638 and the Small Business Innovation and Economic Security Act of 2026 require the Department of War, in coordination with the Small Business Administration, to implement a due diligence program to assess security risks presented by small business concerns seeking a federally funded award.
The DAF will evaluate all small business concerns that submit proposals under this release on whether the concern presents a security risk for any reason. The measures include the due diligence process required under 15 U.S.C. 638(vv), disclosures required under 15 U.S.C. 638(g) and (o), and coordination with the intelligence community as defined in section 3 of the National Security Act of 1947 (50 U.S.C. 3003), federal law enforcement, and other counterintelligence capabilities of the United States Government.
The DAF will assess using a risk-based approach as appropriate: the cybersecurity practices; patent analysis; employee analysis; foreign ownership of a small business concern seeking an award, including the financial ties and obligations, which shall include surety, equity, and debt obligations, of the concern and employees of the concern to a foreign country, foreign person, or foreign entity; foreign affiliations of a covered individual, owner, or other key personnel of a concern with an entity in a foreign country of concern; investment relationships of a concern with an individual or entity in a foreign country of concern; technology licensing agreements or joint ventures, including joint venture like agreements, with an individual or entity in a foreign country of concern; and business relationships between a covered individual, owner, or other key personnel and an individual or entity in a foreign country of concern.
The DAF will also assess proposals utilizing open-source analysis and analytical tools for the nondisclosures of the information set forth in 15 U.S.C. 638(g)(13) or 638(o)(17), and examine any relationship of a concern seeking an award to any entity or individual included on the lists described in 15 U.S.C. 638(g)(16)(D) and 638(o)(20)(D).
If the DAF assesses that a concern has security risks, the DAF will review the proposal, the evaluation, and the security risks and may choose either to create a plan to mitigate the risks or to not select the proposal for award based upon a totality of the review.
Applicants will be required to disclose, under penalty of perjury, the representations, attestations, and certifications required under 15 U.S.C. 638(g)(13) and 638(o)(17), fulfilled by completing Volume 7. In addition, a written statement with any substantial changes to a foreign disclosure form must be provided to the awarding agency within 30 days of any changes while on a project for the DAF.
The DAF will not make an award if it determines the concern has an owner or covered individual that is party to a malign foreign talent recruitment program; has a business entity, parent company, or subsidiary located in the People's Republic of China or another foreign country of concern; has an owner or covered individual that has a foreign affiliation with a research institution located in the PRC or another foreign country of concern; or has a security risk connecting the concern to an entity, including any affiliates of the entity, or individual on any of the following: the UFLPA Entity List maintained by the Department of Homeland Security; the Non-SDN Chinese Military-Industrial Complex Companies List of the Office of Foreign Assets Control maintained by the Department of the Treasury; the Section 889 Prohibition List established under section 889 of the John S. McCain National Defense Authorization Act for Fiscal Year 2019 and maintained by the Department of War; the list of Chinese Military Companies required under section 1260H of the William M. Thornberry National Defense Authorization Act for Fiscal Year 2021 and maintained by the Department of War; the Military End User List maintained by the Bureau of Industry and Security of the Department of Commerce; the Entity List maintained by the Bureau of Industry and Security; the List of Equipment and Services maintained by the Federal Communications Commission; or the Withhold Release Orders and Findings List maintained by U.S. Customs and Border Protection.
The DAF will also not make an award if it determines the concern has a security risk with a primary source that is classified or has a security risk the DAF determines warrants a denial.
If an award is denied on these grounds, the DAF will, as appropriate pursuant to its discretion and in a manner that does not compromise security, provide a notification advising the small business of the determination, the basis for the determination, and a statement that denial of award does not prohibit the concern from being eligible for an award in a subsequent award cycle. The same applies to denials under 15 U.S.C. 638(vv).
Two items deserve extra attention on this topic. Employee analysis and foreign affiliations, because computer vision research is globally distributed and academic appointments abroad are common. And the eight denial lists, because edge compute hardware sourcing can touch them.
Support contractors, status, feedback, and protests
Restrictive notices notwithstanding, proposals may be handled for administrative purposes only by support contractors, which may include but are not limited to TEC Solutions, Inc., APEX, Oasis Systems, Riverside Research, Peerless Technologies, HPC-COM, Mile Two, Montech, Wright Brothers Institute, and MacB (an Alion Company). In addition, only Government employees and technical personnel from the FFRDCs MITRE and Aerospace Corporations working under contract to provide technical support to AF Life Cycle Management Center and Space and Missiles Centers may evaluate proposals. All support contractors are bound by appropriate non-disclosure agreements. Contact the DAF SBIR/STTR Contracting Officer with concerns regarding the use of support contractors.
The Principal Investigator and Corporate Official indicated on the Proposal Cover Sheet will be notified by email regarding proposal selection or non-selection. Small businesses will receive a notification for each proposal submitted, so read each notification carefully and note the Proposal Number and Topic Number referenced.
Automated feedback will be provided for Phase I proposals designated Not Selected. Additional feedback may be provided at the sole discretion of the DAF.
Proposals submitted to the DAF are received and evaluated by different organizations, handled by topic. Each organization operates within its own schedule for proposal evaluation and selection, so updates and notification timeframes will vary.
The Air Force anticipates that all proposals will be evaluated and selections finalized within approximately 90 calendar days of solicitation close. Refrain from contacting the BAA Contracting Officer for proposal status before that time.
Refer to the FY26 STTR BAA for procedures to protest the Announcement. As further prescribed in FAR 33.106(b) and FAR 52.233-3, protests after award should be submitted to Air Force SBIR/STTR Contracting Officer Daniel J. Brewer.
All final reports will be submitted to the awarding DAF organization in accordance with contract instructions. Companies will not submit final reports directly to the Defense Technical Information Center.
How Phase II works
DAF organizations may request Phase II proposals while Phase I technical performance is ongoing or at any time after the conclusion of the period of performance. This decision will be based on the awardee's technical progress, as determined by a DAF Technical Point of Contact review using the Phase II review criteria.
Phase II is the demonstration of the technology found feasible in Phase I. Only Phase I awardees are eligible to submit a Phase II proposal. All Phase I awardees will be sent a notification with the Phase II proposal submittal date and detailed preparation instructions. If the physical or email addresses or firm points of contact have changed since submission of the Phase I proposal, correct information shall be sent to the AF SBIR/STTR One Help Desk. Phase II dollar values, performance periods, and proposal content will be specified in the Phase II request for proposal. Only one Phase II proposal may be submitted for each Phase I award.
The DAF reserves the right to modify the Phase II submission requirements, and should the requirements change, all Phase I awardees will be notified.
Because the Phase II request can arrive while Phase I is still running and is based on demonstrated technical progress, getting preliminary laboratory model testing results early rather than at the end of the six months is worth planning for.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
Proposal deadline: October 21, 2026, at the time specified in the FY26 STTR BAA, ordinarily 12:00 p.m. Eastern. Confirm on DSIP.
Selections finalized: within approximately 90 calendar days of close, on or about January 19, 2027
Period of performance: up to 6 months from award
A working backward plan
Before September 23. Settle eligibility: venture-backed majority ownership excludes you under this document. Contact your research institution partner's sponsored programs office, get their internal timeline and their formal budget with federally negotiated rates, and confirm the 40/30 split closes with no deviation available. Begin the Allocation of Rights Agreement conversation. Read the evaluation criteria in the FY26 STTR BAA. Obtain and read CJCSI 3162.02 on combat assessment methodology, ICD 203 on analytic standards, and the AFRL Targeting Operations and Analytics Development BAA FA8750-25-S-7002 on sam.gov. Settle your novel or under-utilized data modality and your training data answer, given the topic's stated barrier around real time pre- and post-event imagery. Select representative edge hardware for laboratory testing. Request a personnel roster with citizenship and visa status from your partner. Complete Volume 6 training and verify SAM and DSIP alignment.
September 23 through October 3. Draft the technical volume against the 20 page limit. Structure it around the three operational realities: edge processing software under low SWaP in contested environments, AI and ML model development aligned to CJCSI 3162.02 with ICD-compliant confidence expression and the three named performance metrics, and a phased test and evaluation plan from laboratory through simulated to field. Make the timeline improvement quantitative. Write the work plan outline separately with no proprietary content, since it becomes the Statement of Work.
October 4 through October 12. Build the cost volume and itemized listing, neither of which counts against the page limit. Verify the 40/30 split in the numbers. Itemize edge hardware, imagery purchases, and compute with quotes. Get executed agreements with statements of work and detailed cost volumes for your partner and every consultant. Include the recommended trip to the Air Force location managing the contract. Write the detailed TABA request into Volume 5 if requesting TABA.
October 13 through October 16. Assemble Volume 5 with Technical Data Rights Assertions, complete the Volume 7 webform, finish the Volume 4 CCR, and run compliance: 20 page limit, 10-point minimum type, one-inch margins, graphics distinguishable in black and white, no PII uploaded to DSIP, no proprietary content in the technical abstract or the work plan outline.
October 17 through October 19. Submit and certify in DSIP.
October 20 through October 21. Buffer only.
Frequently Asked Questions
What is DAF STTR topic DAF26TZ06-NV007?
DAF26TZ06-NV007 is a Department of the Air Force STTR Phase I topic titled "Automated Combat Assessment at the Edge," released under the DAF 2026 STTR Broad Agency Announcement, Release 6. The objective is to develop a real time automated combat assessment at the edge system, replacing a current practice in which battle damage assessment arrives hours after mission execution.
How much funding is available under DAF26TZ06-NV007?
Up to $300,000 for a period of performance up to 6 months. Proposals in excess of that amount or duration will not be considered for evaluation or award. Up to $6,500 in Technical and Business Assistance is available per Phase I award, in addition to the per-topic total.
When is the proposal deadline?
The topic closes October 21, 2026. The DAF instructions direct offerors to the FY26 STTR BAA for the exact submission time, ordinarily 12:00 p.m. Eastern on the close date. Confirm on the live DSIP posting.
When does this topic open?
September 23, 2026, giving a 29 day submission window.
Are venture capital backed companies eligible?
No, under this document. The instructions state that firms owned in majority part by multiple venture capital operating companies, hedge funds, or private equity funds are not eligible for Department of Air Force topics. The provision is not uniform across DAF Release 6: all three Commercial Solutions Opening documents permit such firms, while three of the four Broad Agency Announcement documents exclude them, the SBIR BAA Phase I instructions being the exception. The exclusion appears in the document governing this topic, so treat it as operative.
Do I need a research institution partner?
Yes. For both Phase I and Phase II, a minimum of 40 percent of each STTR award must be conducted by the awardee and a minimum of 30 percent of the effort must be performed by the single partnering research institution. On a $300,000 award that means roughly $90,000 or more to your partner.
Can I get a waiver on the work split?
No. The instructions state that the DAF will not consider requests for deviations to these performance of work requirements, and the Subcontracts section states deviations are not permitted.
What is the problem this topic solves?
A timeline problem. The state of practice generates after-action reports including battle damage assessment hours after mission execution, relying on multiple electro-optical imagery collects, performed as a sequence of independent tasks that are outdated, time consuming, and lack automation. The Air Force states this does not meet timeline requirements for prosecution of advanced enemy threat systems, because critical missions require battle damage assessment immediately post-strike to inform re-strike planning and drive potential changes to operations.
What are the three named barriers?
The time-consuming and often incomplete nature of traditional damage assessment methods; limitations in remote sensing technologies hindering data collection especially in remote or difficult-to-access areas; and the lack of real time pre- and post-event satellite imagery for model training.
What are the three operational realities the system must follow?
Edge Processing Software, meaning software and data-sharing instructions running directly on field equipment such as sensors and UAVs, covering multi-sensor multi-modal data collection and processing, damage classification and assessment at the edge in near real time, and information sharing protocols, all under low SWaP and contested environment constraints. AI and ML Software Model Development, meaning software that automatically classifies military damage, physical and functional, aligned to CJCSI 3162.02, with confidence intervals aligned to Intelligence Community Directive ICD 302, and performance metrics for accuracy, computation time, and sensors density and location. And Robust Testing and Evaluation in a phased approach from laboratory-based testing through increasingly complex simulated and field environments.
What standards must my damage classification align to?
CJCSI 3162.02, the Chairman of the Joint Chiefs of Staff Instruction on methodology for combat assessment, for damage classification covering both physical and functional damage. Read it before writing, since the terminology comes from that doctrine.
Is it ICD 302 or ICD 203 for confidence intervals?
The topic body cites ICD 302 for confidence intervals, while the reference list cites ICD 203, the Analytic Standards Technical Amendment. ICD 203 is the directive most commonly associated with analytic confidence and probabilistic language standards. We cannot resolve which the author intended, so read both, address analytic confidence standards explicitly, and consider asking the DAF SBIR/STTR One Help Desk during the open period.
What performance metrics does the topic name?
Accuracy, computation time, and sensors density and location. Note that two of the three are not about classification quality: the Air Force wants to know how fast your model runs and how much sensing it needs, which are edge deployment questions.
What does Phase I deliver?
Analysis and design of an automated combat assessment at the edge system for re-strike and re-attack recommendations, with preliminary laboratory model testing. Awardees must demonstrate an innovative approach to physical and functional combat assessment, and novel or under-utilized data modalities. The final report must detail feasibility study findings and a Phase II plan.
What counts as a novel or under-utilized data modality?
The topic does not define it, but the state of practice it describes relies on electro-optical imagery. Synthetic aperture radar, radio frequency signatures, acoustic sensing, infrared, multispectral, passive collection, and fused combinations are all plausible answers. This is an explicit requirement, so name your modality and explain why it is under-utilized and what it adds.
What is the difference between physical and functional damage assessment?
Physical damage assessment concerns the observable structural effect on the target. Functional damage assessment concerns whether the target can still perform its mission. Both are required by this topic, and they are genuinely different inference problems. CJCSI 3162.02 defines the terms.
What happens in Phase II?
Develop, implement, test, evaluate, and demonstrate a distributed automated damage assessment system at the edge following the criteria in the Description and Phase I sections. Awardees transition Phase I designs into working software models running on representative edge hardware, culminating in a demonstration within a simulated or operationally relevant environment. Required deliverables may include functional prototype software and compiled AI and ML model weights, comprehensive system architecture and interface control documentation, and a Phase II final report detailing test results, performance metrics, criteria implementations, and a commercialization and transition plan for Phase III.
Compiled model weights are a deliverable. What are the IP implications?
Significant, and worth planning in Phase I. The Government obtains SBIR/STTR data rights in data developed or generated under the contract for 20 years, and on an STTR award your research institution partner may contribute to model development. Use Technical Data Rights Assertions in Volume 5 to identify pre-existing models, training pipelines, and architectures as background intellectual property, and settle the Allocation of Rights Agreement with your partner early.
What are the commercial applications?
The topic names urban surveillance and disaster and humanitarian response, plus other use cases. Disaster response is the strongest dual-use fit, since rapid automated damage assessment from multi-modal sensing at the edge is exactly what post-earthquake, post-hurricane, and post-flood assessment requires, with customers including emergency management agencies, relief organizations, insurers, and infrastructure operators.
How long can my technical volume be?
20 pages or slides, per the topic index. The Technical Volume must be no smaller than 10-point on standard 8.5 by 11 inch paper with one-inch margins. The Cover Sheet, Cost Volume, and Cost Volume Itemized Listing do not count toward the limit; only the Technical Volume and any enclosures or attachments do. Documents uploaded into Volume 5 under "Other" do not count either.
Is there a required proposal template?
No. Unlike the DAF STTR CSO Phase I instructions in this same release, this BAA document does not require a government template.
Is this topic export controlled?
This topic does not carry a topic-level ITAR and EAR restriction paragraph. That said, targeting and combat assessment is sensitive subject matter, and your disclosure obligations apply in full, including identifying all U.S. permanent residents and all non-U.S. citizens involved with countries of origin, visa or work permit type, and anticipated level of involvement.
How will my proposal be evaluated?
This document does not restate the evaluation criteria. It says only that proposals will be evaluated for overall merit in accordance with the criteria discussed in the FY26 STTR BAA. Read those criteria before writing.
Does the Company Commercialization Report affect my score?
No. Completing the CCR as Volume 4 in DSIP is required, but the instructions state information contained in the CCR will not be considered by the Air Force during proposal evaluations.
What references should I read?
Three are cited and all three are worth obtaining. ICD 203 Analytic Standards Technical Amendment, ODNI, 2023. "Methodology for Combat Assessment," Joint Chiefs of Staff, 2021, which is CJCSI 3162.02A. And BAA FA8750-25-S-7002, Targeting Operations and Analytics Development, on sam.gov. That third reference is the least obvious and the most useful competitively: FA8750 is the AFRL Information Directorate contracting prefix, and reading that BAA shows you the existing program area and the language the program office uses.
What is TABA and how do I request it?
Up to $6,500 per Phase I award, in addition to the per-topic total, identified in the Volume 3 Cost Proposal. The detailed request must be in Volume 5 with provider name, point of contact with email and phone, an explanation of unique qualifications, the tasks with purpose and objective, and total cost with hours and labor rates, with the task milestone list tracking to your milestone payment schedule. Requests specifying only a value in Volume 3 will not be considered. Given the model weights deliverable and the required Allocation of Rights Agreement, intellectual property protections is the most useful eligible activity.
How do I get to Phase II?
DAF organizations may request Phase II proposals while Phase I technical performance is ongoing or at any time after the period of performance concludes, based on the awardee's technical progress as determined by a DAF Technical Point of Contact review. Only Phase I awardees are eligible, and only one Phase II proposal may be submitted for each Phase I award.
When will I hear about selection?
The Air Force anticipates all proposals will be evaluated and selections finalized within approximately 90 calendar days of solicitation close, meaning on or about January 19, 2027. The Principal Investigator and Corporate Official on the cover sheet are notified by email.
Who do I contact with questions?
The DAF SBIR/STTR One Help Desk at usaf.team@afsbirsttr.us for general program and proposal preparation questions, and the DAF encourages requesting clarifying information as early as possible because delays constrain its ability to respond. For DSIP submission system issues, dodsbirsupport@reisystems.com. For technical questions about the topics during the pre-announcement and open period, reference the FY26 STTR BAA. The Air Force SBIR/STTR Contracting Officer is Mr. Daniel J. Brewer at Daniel.Brewer.13@us.af.mil. Address or point of contact changes after submission go to the One Help Desk with the subject line "FY26 STTR BAA Address Change."
Positioning Advice for Companies Considering This Topic
Clear the two gates first. Venture-backed majority ownership excludes you under this document, and the 40/30 STTR split is not waivable. Your research institution partner has to do at least 30 percent of the work, so pick one whose contribution is real.
Make the timeline claim quantitative. The entire premise is that hours is too slow. State what your approach delivers and when: seconds, minutes, or the first actionable assessment within a stated window. A proposal that describes better classification without addressing latency has missed the point of the topic.
Read CJCSI 3162.02 and use its vocabulary. Physical damage assessment, functional damage assessment, and target system assessment are doctrinal terms with specific meanings. Aligning your classification taxonomy to that instruction is a stated requirement, and getting the language right signals you understand the customer's process rather than just the computer vision problem.
Answer the confidence expression requirement properly. Whether the governing directive is ICD 302 or ICD 203, the point is that model outputs must carry calibrated, doctrinally expressed confidence rather than raw softmax scores. Analysts and targeteers work in probabilistic language standards, and a model that cannot express uncertainty in their terms is not usable in their workflow.
Take the compound edge constraint seriously. Multi-sensor, multi-modal, near real time, low SWaP, contested environment. Any one of those is tractable. Together they force real architecture decisions about what runs on the sensor, what runs on the platform, what gets transmitted, and what happens when the link degrades. Show those decisions.
Bring a real answer on training data. The topic names the lack of real time pre- and post-event satellite imagery as a barrier. Synthetic data generation, transfer learning from civil damage datasets, few-shot approaches, physics-based simulation, or commercial imagery partnerships are all plausible. Silence on this looks like you have not thought about it.
Use the novel modality requirement to differentiate. It is an explicit Phase I requirement and most competitors will default to electro-optical imagery with better models. If you bring synthetic aperture radar, radio frequency, acoustic, or fused multi-modal sensing, lead with it.
Write a real test and evaluation plan. Operational effectiveness, suitability, and survivability are formal terms, and the topic specifies a phased approach from laboratory through simulated to field. A credible phased plan with entry and exit criteria per stage distinguishes a serious proposal.
Read the AFRL Targeting Operations and Analytics BAA. FA8750-25-S-7002 is cited in the references and it is on sam.gov. It tells you the existing program area, the office behind it, and the terminology in use. Very few SBIR topics hand you a live related BAA to study.
Plan the intellectual property boundary now. Compiled model weights as a deliverable, 20 year Government data rights, and a research institution co-developing the models is a combination that needs a plan before award, not after. The Allocation of Rights Agreement is required, and TABA can fund IP support.
Build the commercial case on disaster response. Urban surveillance and disaster and humanitarian response are the two named applications, and the second is a genuinely large market with the same technical shape: rapid, automated, multi-modal damage assessment at the edge where connectivity is poor.
DAF STTR DAF26TZ06-NV006: Artificial Intelligence Performance Evaluation Tool
Deadline: October 21, 2026
Funding Award Size: $300k
Description: Complete guide to DAF STTR Phase I topic DAF26TZ06-NV006, AI Performance Evaluation Tool for Air Force Test Center. Up to $300,000 over 6 months. Closes October 21, 2026.
Quick Answer
DAF26TZ06-NV006 is a Department of the Air Force STTR Phase I topic under the DAF 2026 STTR Broad Agency Announcement, Release 6. The Air Force Test Center has a problem it states plainly: current AI evaluation processes lack standardization and reproducibility across varying operational environments, which makes it difficult for AFTC to confidently assess AI models' trustworthiness, reliability, and operational readiness. This topic funds the design of a tool that fixes that, capable of evaluating AI-enabled sensor suites and aligning with the simulation environments AFTC already uses. Awards are up to $300,000 for a period of performance up to 6 months, with a 20 page technical volume limit. The topic opens September 23, 2026 and closes October 21, 2026 through the Defense SBIR/STTR Innovation Portal.
Two eligibility facts to settle before you invest effort. This document states that small businesses majority-owned in part by multiple venture capital operating companies, hedge funds, or private equity funds are not eligible. And because this is STTR, you need a research institution partner, with a work split this document states precisely and refuses to waive: a minimum of 40 percent of each STTR award conducted by the awardee and a minimum of 30 percent performed by the single partnering research institution, with the DAF stating it will not consider deviation requests.
Topic At a Glance
Topic number: DAF26TZ06-NV006
Title: Artificial Intelligence Performance Evaluation Tool
Solicitation: Department of the Air Force 2026 Small Business Technology Transfer (STTR) Broad Agency Announcement (BAA), Release 6, Phase I Proposal Submission Instructions
Program: STTR, requiring a single partnering research institution
Program type: Phase I
Award maximum value: $300,000
Award maximum duration: 6 months
Technical volume page limit: 20 pages or slides
OUSD (R&E) Critical Technology Area: Applied Artificial Intelligence (AAI)
Component Technology Priority Area: Trusted AI and Autonomy
Projected CMMC level requirement: Level 2 (Self)
Export control status: no topic-level ITAR or EAR restriction paragraph appears on this topic
Venture capital, hedge fund, private equity majority ownership: not eligible under this document
Work split: minimum 40 percent by the small business, minimum 30 percent by the single partnering research institution, for both Phase I and Phase II. Deviations will not be considered
Technology readiness: Phase I initial TRL 2, target TRL 4. Phase II initial TRL 4, target TRL 6
Named customer: Air Force Test Center, with the 412th Test Wing test environment and the 416th Test Squadron at Edwards AFB as the Phase III transition partner
Named simulation environments: AFSIM, Ansys Test and Evaluation Tool Kit (TETK), or Joint Simulation Environment (JSE)
Technical and Business Assistance: up to $6,500 per Phase I award, in addition to the per-topic total
Type size: no smaller than 10-point
Evaluation criteria: not restated in this document. See the FY26 STTR BAA
Topic open date: September 23, 2026
Proposal deadline: October 21, 2026, at the time stated in the FY26 STTR BAA
Selection timeline: within approximately 90 calendar days of solicitation close, meaning on or about January 19, 2027
Submission portal: DSIP at dodsbirsttr.mil
Keywords: artificial intelligence, AI performance evaluation, trusted AI, autonomy, explainability, AI metrics, AI validation, AI systems, sensor technologies, information systems, operational testing, ABMS, 412TW, AFTC, DoD AI standards, test and evaluation
Note on the deadline time. The DAF instructions direct offerors to the FY26 STTR BAA for the proposal submission deadline date and time. DoW deadlines are ordinarily 12:00 p.m. Eastern on the close date. Confirm the exact time on the live DSIP posting rather than assuming end of day.
The Two Eligibility Gates
Venture capital ownership is disqualifying under this document
The instructions state that small business concerns owned in majority part by multiple venture capital operating companies, hedge funds, or private equity funds are not eligible to submit applications or receive awards for Department of Air Force Topics.
This provision is not uniform across the DAF Release 6 component instructions, so here is the actual state of it across all seven documents.
Not eligible: the SBIR BAA Direct to Phase II instructions covering topic DV037; the STTR BAA Direct to Phase II instructions covering topic DV008; and this STTR BAA Phase I document covering topics NV006 and NV007.
Eligible: the SBIR BAA Phase I instructions covering topics NV034 through NV036; the SBIR CSO Direct to Phase II instructions; the SBIR CSO Phase I instructions; and the STTR CSO Phase I instructions.
Read as a pattern, every Commercial Solutions Opening document permits venture-backed firms, and three of the four Broad Agency Announcement documents exclude them, with the SBIR BAA Phase I instructions the lone exception. Whether that is deliberate or a drafting inconsistency, the DAF has not said. What matters practically is that the exclusion appears in the document governing this topic, so treat it as operative. Verify your status against 13 CFR 121.702 and the definition in the instructions, and raise it with the DAF SBIR/STTR One Help Desk during the open period rather than assuming it resolves in your favor.
The STTR work split, stated precisely and not waivable
This document is the clearest of the DAF Release 6 instructions on the STTR partnership requirement, and it is worth quoting closely because two other DAF STTR documents in the same release handle it less clearly.
For both Phase I and Phase II, a minimum of 40 percent of each STTR award must be conducted by the awardee and a minimum of 30 percent of the effort must be performed by the single partnering research institution. Applicants and awardees may partner with multiple entities that separately meet the definition of a research institution as indicated in the STTR BAA. Applicants may use only one partnering research institution to meet STTR eligibility requirements. The DAF will not consider requests for deviations to these performance of work requirements.
The Subcontracts section restates it: a minimum of 40 percent of each STTR project must be conducted by the small business concern and a minimum of 30 percent of the effort performed by the single partnering research institution, and deviations from these performance of work requirements are not permitted. The STTR funded work percentage calculation considers both direct and indirect costs after removal of the small business concern's proposed profit.
Four practical points. The 40/30 split is the statutory STTR structure, correctly stated here, and it applies to both phases. You may have multiple research institution partners, but only one of them counts toward satisfying STTR eligibility, so designate which one deliberately. Deviations are not available, so unlike several other DAF Release 6 documents there is no deviation process to fall back on. And the percentage calculation uses direct and indirect costs after removing your proposed profit, which is the same convention used elsewhere in the release.
The instructions also point to the DoW STTR FY26 BAA for more information regarding the required Allocation of Rights Agreement. That agreement governs intellectual property between you and your research institution partner, and it is a real document your partner's technology transfer office will need to negotiate. Start it early.
Worth noting for anyone reading across our other DAF Release 6 pages: the STTR CSO Phase I instructions state a two-thirds in-house floor and the STTR BAA Direct to Phase II instructions state 50 percent plus a two-thirds figure labeled Phase I, neither of which matches the STTR statutory split. This document gets it right at 40/30. If you are working a topic governed by one of those other documents, that discrepancy is worth a Help Desk question, and this document is useful evidence of what the STTR requirement actually is.
What the Air Force Is Actually Looking For
The problem, in AFTC's terms
The proposed research effort aims to address critical gaps in the evaluation and validation of artificial intelligence systems within the Air Force by developing an Artificial Intelligence Performance Evaluation Tool. The tool will provide a standardized, adaptive, and robust framework capable of assessing the performance, reliability, and explainability of AI models, with a targeted application in sensor and information systems used in test and evaluation missions under the 412th Test Wing of the Air Force Test Center. This effort aligns with the Department of Defense's thrust toward advancing Trusted AI and Autonomy and operationally focused Advanced Battle Management System initiatives.
The unmet need is stated directly. Current AI evaluation processes lack standardization and reproducibility across varying operational environments, making it difficult for the AFTC to confidently assess AI models' trustworthiness, reliability, and operational readiness. Furthermore, the escalation in complexity of AI-enabled systems has underscored the need for technologies that can quantitatively measure AI performance while adhering to DoD Responsible AI principles. These challenges prevent the Air Force from leveraging AI advancements at scale, impacting readiness and modernization goals.
That last sentence is the value argument. The bottleneck on fielding AI is not model capability, it is the inability to evaluate models to a standard someone will sign off on. Frame your proposal around removing that bottleneck.
The desired outcome
By creating a modular AI evaluation tool, this effort provides an opportunity to establish a scalable and adaptable solution that aligns with emerging DoD AI standards. The desired outcome is a measurable enhancement in the fidelity, speed, and consistency of AI performance assessments, reducing decision-making cycles and increasing mission readiness. The tool will enable practical implementation of AI validation in secure environments, ensuring compliance with applicable standards and operational conditions. Successful deployment will support the broader modernization priorities tied to mission-critical AI deployment and advanced autonomy.
Fidelity, speed, and consistency are the three named improvement axes. Quantify your expected gains against all three where you can.
The hard requirements
Three requirements in the objective paragraph function as gates, and a proposal that does not address them is not responsive.
Sensor suite evaluation. To demonstrate early-stage feasibility, the proposed tool must be capable of evaluating AI-enabled sensor suites, with electro-optical and infrared, radar, and radio frequency systems given as the examples. This is not a generic model evaluation framework. It has to handle AI operating on sensor data.
Environment alignment. The tool must align with AFTC's primary simulation and digital engineering environments, specifically the Advanced Framework for Simulation, Integration, and Modeling (AFSIM), the Ansys Test and Evaluation Tool Kit (TETK), or the Joint Simulation Environment (JSE). Note the "or," meaning you may pick one, but you must pick one and show the alignment. If you have never worked with any of the three, that is the gap to close before writing.
Standards compliance. Ensuring compliance with evolving Department of Defense AI standards, such as those delineated in the DoD Responsible AI Strategy and Implementation Pathway.
The three critical gaps the tool must address
By addressing critical gaps in assessing AI robustness under dynamic and contested conditions, the effort aims to establish early-stage feasibility for an adaptive framework capable of identifying key performance indicators, assessing trustworthiness through explainability metrics, and ensuring compliance with evolving DoD AI standards.
The three named robustness gaps are real-time inference latency, adversarial resilience, and data drift detection. Those three recur in the Phase I research task, so treat them as your core metric set.
Phase I, and a note on the document's structure
The topic contains two Phase I descriptions and two Phase II descriptions, which do not fully align with each other. The first pair is written around the AI evaluation framework and sits under an Approach heading with explicit TRL targets. The second pair appears after a Conclusion paragraph and is written around flight test analysis workflows, hardware and software architectural design, Python-based extensibility, and Time Space Position Information processing accuracy.
The second pair reads as though it may have been carried in from a related but distinct topic, since a hardware architectural design and TSPI processing accuracy are not natural fits for a software AI evaluation framework. We have not resolved which governs, and neither have the instructions. The prudent approach is to write primarily to the first, more detailed pair, while addressing the elements of the second pair that are compatible: the analysis of current flight test analysis workflows to identify bottlenecks, constraints, and user requirements is genuinely useful, and modularity, security, and Python-based extensibility are reasonable architectural commitments. Ask the DAF SBIR/STTR One Help Desk during the open period which description governs.
Here is what the primary Phase I description requires. Phase I is Feasibility and Concept Development, initial TRL 2, target TRL 4.
Research. Conduct an initial landscape analysis of AI performance metrics, methodologies, and tools with a focus on gaps and challenges specific to AFTC's mission space. Identify and assess algorithms for measuring real-time inference latency, adversarial resilience, and data drift specifically for different sensor models. This research must establish and utilize open-source or synthetically generated surrogate sensor datasets that are representative of AFTC flight test scenarios.
That surrogate data requirement is the most consequential sentence in the Phase I description. You will not be given real AFTC flight test sensor data. You must establish and use open-source or synthetic surrogate datasets that are representative of AFTC flight test scenarios, and the deliverables list names a Surrogate Data Strategy and Baselining as a possible component. Your proposal should name your datasets or your synthesis approach and argue their representativeness.
Framework Design. Develop a software architecture for an adaptable AI performance evaluation framework designed to address reliability, robustness, and explainability metrics. Awardees will need to base their design on widely accepted DoD digital engineering and telemetry standards. The architecture must explicitly define the assumed data ingestion pipelines for surrogate sensor data and detail the computational methods that will be used to calculate reliability, robustness, and explainability metrics.
Note two explicit requirements there: data ingestion pipelines defined, and computational methods for the metrics detailed. Not described in general terms, but defined and detailed.
Simulation. Perform initial simulation tests to explore early-stage integration of explainability metrics and KPI identification methodologies. This will consist of running dry-runs within a localized software sandbox. The objective of these tests is to demonstrate that the proposed mathematical models can successfully process these inputs, identify key performance indicators such as data drift, and output quantifiable explainability scores.
Deliverables. A feasibility report describing how the tool can practically operate within the AFTC environment, a detailed framework design document, and initial simulation results demonstrating concept validity. This could include Surrogate Data Strategy and Baselining, algorithmic and mathematical validation, architectural alignment and computational overhead, and Phase II scalability and transition pathway.
Computational overhead appears in that list, which is a useful signal. A framework that produces excellent metrics at prohibitive computational cost will not fit a test environment, and the Air Force is asking you to characterize the cost.
The secondary Phase I description adds: a comprehensive study of current flight test analysis workflows to identify specific bottlenecks, constraints, and user requirements; a detailed hardware and software architectural design emphasizing modularity, security, and Python-based extensibility; development and simulation of proof-of-concept algorithms for automated metric generation and data processing; and a final report summarizing findings, the complete system design, simulation results, and a detailed plan for Phase II prototype development.
Phase II
Phase II is Prototype Development and Validation, initial TRL 4, target TRL 6.
Prototype Development. Build a functional prototype of the performance evaluation tool with a secure Python-based backend and a user-friendly graphical user interface.
Integration. Test and integrate the tool with existing sensor and information systems in a controlled 412th Test Wing test environment to assess real-world performance.
AI Metrics Library. Develop an expandable library of explainability metrics aligned with DoD Trusted AI standards to support diverse mission applications.
Validation. Conduct comprehensive testing in operationally relevant scenarios to ensure reliability, scalability, and compliance with DoD standards.
Deliverables. A fully functional prototype, test reports validating system performance, and a roadmap for field-level deployment.
The secondary Phase II description adds performance validation against a set of metrics including Time Space Position Information processing accuracy, comparative analysis against legacy methods, integrity of automated report generation, and system security and resilience against simulated threat and victim scenarios, plus lab-based testing in a secure closed-loop environment and an end-to-end demonstration.
Upon completion of Phase II, the AI Performance Evaluation Tool is expected to reach TRL 6, with a clearly defined transition pathway toward operational implementation across the 412th Test Wing and other DoD components.
Phase III and the named transition partner
The 416th Test Squadron at Edwards Air Force Base, which specializes in F-16 flight testing and has a history of integrating advanced systems, will serve as the dedicated transition partner for this effort.
That is unusually specific for a Phase I topic and it is the most valuable line in the Phase III section. A named squadron with a stated history of integrating advanced systems is a transition pathway you can reference and build toward.
The commercialization strategy is described as multifaceted. Productization, refining the successful Phase II prototype into a robust, user-friendly, and cost-effective commercial product. User-centric feedback, conducting extensive operational testing with the 416th Test Squadron and other target users to gather feedback for iterative product improvements. Dual-use applications, where the primary customer is the Department of War but commercial aerospace, avionics testing, and other sectors requiring complex data analysis will be explored. And infrastructure and support, establishing comprehensive manufacturing, customer support, and training infrastructure to ensure a smooth transition from a prototype to a fully supported operational tool.
The reference list, which is a reading assignment
Eleven references, and they map the standards landscape you are expected to work within.
Department of Defense Responsible Artificial Intelligence Strategy and Implementation Pathway, at ai.mil. National Institute of Standards and Technology Artificial Intelligence Risk Management Framework, at nist.gov/ai-risk-management. Air Force Test Center Organizational Overview, at aftc.af.mil. John M. McQuade, et al., "Metrics for Trustworthy AI: Evaluating Reliability, Robustness, and Explainability in Defense Applications," DoD Artificial Intelligence Symposium, 2023. Executive Order 13960, Promoting the Use of Trustworthy Artificial Intelligence in the Federal Government. Advanced Battle Management System Fact Sheet, U.S. Air Force. Defense Innovation Unit, "Trusted AI Framework: Operationalizing AI Ethics and Reliability," March 2022. Test and Evaluation Framework for Autonomy, presented at T&E Working Group, AFTC, 2022. "Explainable AI (XAI) Principles for DoD Applications," DARPA XAI Program Summary, 2021. Office of the Under Secretary of Defense for Research and Engineering, "Key Performance Indicators for AI and Autonomy Testing," Report No. OUSD(A&S)-21-0045, 2023. And an April 15, 2026 AFTC news article, "From weeks to minutes: How AI is accelerating the flight test process."
Three observations. The McQuade paper title maps almost exactly onto the topic's three metric categories of reliability, robustness, and explainability, so it is likely the conceptual backbone. The OUSD KPI report on AI and autonomy testing is directly on point for the KPI identification requirement. And the April 2026 AFTC news article is recent and specific to this customer, which suggests the topic was written by people close to a live effort. Read it.
Funding Allowance and Cost Structure
Award ceiling
Up to $300,000 across up to 6 months. The topic index states that proposals in excess of this amount will not be considered for evaluation or award, and proposals in excess of this duration will not be considered for evaluation or award.
The instructions add that any per-award or per-topic funding caps are budgetary estimates only and more or less funding may become available. Multiple procurements are planned and anticipated to be awarded as a result of the topic, each proposal is considered a separate procurement and will be evaluated on its own merit, and the Government may award all, some, or none of the proposals. Funding decisions are made with complete disregard to the other awards under the same topic.
At $300,000 with a mandatory 30 percent minimum to your research institution partner, roughly $90,000 or more flows to the partner and your own share is at least $120,000. Six months of a small team plus an academic effort is enough for a rigorous landscape analysis, an architecture, a surrogate data strategy, and sandbox simulation, which is what the topic asks for. It is not enough for a working tool, and Phase I does not ask for one.
Contract type
The DAF primarily makes STTR Phase I and Phase II awards as firm-fixed-price contracts. Awardees are strongly urged to work toward a Defense Contract Audit Agency approved accounting system, since if the company intends to continue work with the Department of War an approved accounting system allows competition in a broader array of acquisition opportunities, including award of cost-reimbursement type contracts.
If no exceptions are taken to an offeror's proposal, the Government may award a contract without negotiations. Therefore the offeror's initial proposal should contain the offeror's best terms from a cost or price and technical standpoint. If there are questions regarding the award document, contact the Phase I Contracting Officer identified on the cover page. The Government reserves the right to reopen negotiations later if the Contracting Officer determines doing so to be necessary.
Technical and Business Assistance
The Small Business Innovation and Economic Security Act Section 7 mandates agencies to offer TABA. The DAF will provide up to $6,500 per Phase I award, and this total is in addition to the per-topic total identified in the DSIP Volume 3 Cost Proposal.
Note that this STTR BAA document carries the full detailed TABA request requirements, unlike the DAF CSO Phase I documents which state the amount in a single sentence.
Awardees can only use TABA funding for the activities outlined in 15 U.S.C. 638(q)(1) and the purposes outlined in 15 U.S.C. 638(q)(1)(A) through (E). Eligible activities include access to a network of scientists and engineers engaged in a wide range of technologies, assistance with product sales, intellectual property protections, cybersecurity assistance, market research, market validation, development of regulations and manufacturing plans, and access to technical and business literature available through online databases. Those activities can be undertaken in furtherance of making better technical decisions concerning such projects, solving technical problems which arise during the conduct of such projects, minimizing technical risks associated with such projects, developing and commercializing new commercial products and processes resulting from such projects including intellectual property protections, and screening for potential foreign involvement in technology development or commercialization activities.
On an STTR award with a required Allocation of Rights Agreement, intellectual property protections is the standout eligible activity.
Requests for TABA funding creditable to a TABA provider must include the following or will be subject to denial: the TABA providers; the providers' point of contact, email address, and phone number; an explanation of the provider's unique qualifications to provide the TABA service; the tasks that will be performed by the provider including the purpose and objective of the assistance, with the task milestone list tracking to the milestone payment schedule otherwise provided by the applicant; and total provider cost, number of hours, and labor rates, with average or blended rates acceptable.
If proposing TABA funding to hire new staff, augment staff, or direct staff to conduct or participate in training activities consistent with the purpose of TABA, the following must be included or may be subject to denial: names and positions, business need to be filled or training to be provided; number of employees to be hired, augmented, or directed to participate in training activities; qualifications of employees hired or augmented, or detailed need for training; tasks that will be performed including both a description of the activity and the purpose that it will serve; and total staff or training cost, number of hours, and labor rates.
A detailed request for TABA funding must be included in the Volume 5 Supporting Documents in DSIP. TABA requests that only specify a TABA request value in the Volume 3 Cost Proposal will not be considered.
Cost volume requirements
Cost information should be provided by completing the Cost Volume in DSIP and including the Cost Volume Itemized Listing. The Cost Volume detail must be adequate to enable Air Force personnel to determine the purpose, necessity, and reasonableness of each cost element. The DSIP Cost Volume and Itemized Cost Volume Information will not count against the specified page limit, and the itemized listing may also be submitted in Volume 5 under the "Other" dropdown option.
Direct cost materials. Justify costs for materials, parts, and supplies with an itemized list containing types, quantities, prices, and where appropriate purpose.
Other direct costs. This category includes but is not limited to specialized services such as machining, milling, special testing or analysis, and costs incurred in temporarily using specialized equipment. Proposals including leased hardware must include an adequate lease versus purchase justification. Compute resources for model training and simulation, and any licensing for AFSIM, TETK, or JSE access, plausibly sit here.
Direct labor. Identify key personnel by name, if possible, or by labor category, if not. Direct labor hours, labor overhead or fringe benefits, and actual hourly rates for each individual are necessary for the Contracting Officer to determine whether these hours, fringe rates, and hourly rates are fair and reasonable. On a study and design Phase I this is the dominant element.
Travel. Travel costs must relate to project needs. Break out travel costs by trip, number of travelers, airfare, per diem, lodging, and similar. The number of trips required, as well as the destination and purpose of each, should be reflected. The instructions recommend budgeting at least one trip to the Air Force location managing the contract. On a topic with a named customer at Edwards AFB, that trip has substantive value.
Subcontracts. Involvement of a consultant in the project's planning or research stages may be appropriate. If so, describe in detail and include information in the Cost Volume. A minimum of 40 percent of each STTR project must be conducted by the small business concern and a minimum of 30 percent of the effort performed by the single partnering research institution. Deviations from these performance of work requirements are not permitted. The STTR funded work percentage calculation considers both direct and indirect costs after removal of the small business concern's proposed profit. Support subcontract costs with copies of executed agreements that adequately describe the work to be performed, and at a minimum include a Statement of Work with a corresponding detailed Cost Volume for each planned subcontract. Additionally, see the DoW STTR FY26 BAA for more information regarding the required Allocation of Rights Agreement.
Special tooling, special test equipment, and material. The inclusion of equipment and materials will be carefully reviewed relative to need and appropriateness to the work proposed. Special tooling and special test equipment purchases must, in the Contracting Officer's opinion, be advantageous to the Government and relate directly to the effort, and should not be of a type that an offeror would otherwise possess in the normal course of business.
Consultants. Provide a separate agreement letter for each consultant, briefly stating what service or assistance will be provided, the number of hours required, and the hourly rate.
Where work must be performed
All R/R&D work must be performed in the United States. Based on a rare and unique circumstance, the DAF may approve a particular portion of the R/R&D work to be performed or obtained in a country outside the United States. The awarding Funding Agreement officer must approve each specific condition in writing. Applicants seeking this approval must make the request with their initial proposal submission, and the DAF will not consider these requests prior to proposal submission.
Personnel Disclosure
This topic does not carry a topic-level ITAR and EAR restriction paragraph. Your disclosure obligations still apply in full.
Identify in the Technical Volume all key personnel who will be involved in this project, including information on directly related education, experience, and citizenship. A technical resume of the principal investigator, including a list of publications if any, must be included, and only one principal investigator or project manager can be designated to a proposal at any given time. Concise technical resumes for subcontractors and consultants are also useful.
Identify all U.S. permanent residents to be involved in the project as direct employees, subcontractors, or consultants. Identify all non-U.S. citizens expected to be involved in the project as direct employees, subcontractors, or consultants. For all non-U.S. citizens, in addition to technical resumes, provide countries of origin, the type of visa or work permit under which they are performing, and an explanation of their anticipated level of involvement on this project, as appropriate. Additional information may be requested during negotiations in order to verify the foreign citizen's eligibility to participate on a contract issued as a result of this announcement.
Do not upload information such as Permanent Resident Cards, birth certificates, Social Security Numbers, or other personally identifiable information to the DSIP system. Provide the categories of information requested, not the underlying identity documents.
On an STTR award, this reaches your research institution partner's personnel. University AI and machine learning groups are internationally staffed, so request a complete roster with citizenship and visa status from your partner early. Note also that the foreign risk evaluation below treats foreign affiliation with a research institution in a country of concern as a mandatory award denial.
Proposal Structure: The Seven Volumes
Formatting, and what does not count against the page limit
The Technical Volume should include all graphics and attachments but should not include the Cover Sheet, which is completed separately as Volume 1. Ensure that all graphics are distinguishable in black and white.
The Technical Volume must be no smaller than 10-point on standard 8.5 by 11 inch paper with one-inch margins. Note this differs from the DAF Direct to Phase II instructions, which require 11-point.
The Phase I Technical Volume page and slide limits identified for the topics do not include the Cover Sheet, the Cost Volume, or the Cost Volume Itemized Listing. Only the Technical Volume and any enclosures or attachments count toward the page limit. The documents required for upload into Volume 5 under "Other" do not count toward the specified limits. In the interest of equity, pages or slides in excess of the stated limits will not be reviewed.
The limit for this topic is 20 pages or slides. Because it is expressed as pages or slides, a slide-format volume is acceptable.
Note that this document, unlike the STTR CSO Phase I instructions in the same release, does not require a government proposal template. You build the technical volume yourself, subject to the required elements below.
Fraud, Waste and Abuse training must be completed prior to proposal submission, and DSIP will indicate completion of the Volume 6 requirement once the training is complete and certified.
Volume 1, Cover Sheet
Complete the proposal Cover Sheet in accordance with the instructions provided via DSIP. The technical abstract should include a brief description of the program objectives, a description of the effort, anticipated benefits and commercial applications of the proposed research, and a list of keywords and terms.
The technical abstract of each successful proposal will be submitted to the Office of the Secretary of War for publication and therefore must not contain proprietary or classified information. If selected for funding, the proposal's technical abstract and discussion of anticipated benefits will be publicly released.
Volume 2, Technical Volume
The Phase I technical volume shall contain the required elements below. These instructions supplement the FY26 STTR BAA, and in addition to the requirements found in the BAA, applicants are required to provide the following information in Volume 2.
Key personnel, as described above.
Phase I Statement of Work outline. The DAF uses the work plan outline as the initial draft of the Phase I Statement of Work. Therefore, do not include proprietary information in the work plan outline. To do so will necessitate a request for revision, if selected, and may delay award.
Include a work plan outline in the following format. Scope, listing the effort's major requirements and specifications. Task Outline, providing a brief outline of the work to be accomplished during the Phase I effort. Milestone Schedule. Deliverables. Progress reports. Final report with SF 298.
The SF 298 is the Report Documentation Page that accompanies a technical report. It is named in the required outline, so plan for it. Note also that the topic's secondary Phase I description makes the final report substantive: it must summarize findings, the complete system design, simulation results, and a detailed plan for Phase II prototype development.
Volume 3, Cost Volume
Covered above.
Volume 4, Company Commercialization Report
Completion of the CCR as Volume 4 of the proposal submission in DSIP is required. Refer to the FY26 STTR BAA for full details. Information contained in the CCR will not be considered by the Air Force during proposal evaluations.
Volume 5, Supporting Documents
Three documents may be required if applicable to your proposal. DD Form 2345, for proposals submitted under export-controlled topics. Verification of Eligibility of Small Business Joint Ventures, Attachment 3 to the FY26 STTR BAA. Technical Data Rights Assertions, if asserting data rights restrictions.
Technical Data Rights Assertions deserve attention on an STTR AI topic. Your existing models, evaluation code, and metric implementations are likely background intellectual property, and your research institution partner will have its own. The Allocation of Rights Agreement and these assertions together are how the boundaries get drawn.
The detailed TABA request, if requesting TABA, also goes in Volume 5. Documents uploaded into Volume 5 under "Other" do not count toward the page limit, and the Cost Volume Itemized Listing may also be submitted there.
Volume 6, Fraud, Waste and Abuse Training
Material can be found in the Volume 6 section of the proposal submission module in DSIP and must be thoroughly reviewed once per year to proceed with proposal submission.
Volume 7, Disclosures of Foreign Affiliations or Relationships to Foreign Countries
Small business concerns must complete the webform in Volume 7 of the DSIP proposal submission. The disclosures will not be accepted as a PDF Supporting Document in Volume 5, and previous versions of this form must not be uploaded to Volume 5. For additional details, refer to the STTR Program BAA.
How Your Proposal Will Be Evaluated
The criteria live in the BAA
Proposals will be evaluated for overall merit in accordance with the criteria discussed in the FY26 STTR BAA. The DAF is seeking varying technical and scientific approaches and varying and new technologies that would be responsive to the problem statements and areas of interest in the topic.
This document does not reproduce the evaluation criteria, unlike the DAF CSO instructions which restate Criteria A, B, and C with their relative importance. Read the criteria in the FY26 STTR BAA before you begin writing, and do not assume the CSO criteria carry over.
Foreign risk evaluation
15 U.S.C. 638 and the Small Business Innovation and Economic Security Act of 2026 require the Department of War, in coordination with the Small Business Administration, to implement a due diligence program to assess security risks presented by small business concerns seeking a federally funded award.
The DAF will evaluate all small business concerns that submit proposals under this release on whether the concern presents a security risk for any reason. The measures include the due diligence process required under 15 U.S.C. 638(vv), disclosures required under 15 U.S.C. 638(g) and (o), and coordination with the intelligence community as defined in section 3 of the National Security Act of 1947 (50 U.S.C. 3003), federal law enforcement, and other counterintelligence capabilities of the United States Government.
The DAF will assess using a risk-based approach as appropriate: the cybersecurity practices; patent analysis; employee analysis; foreign ownership of a small business concern seeking an award, including the financial ties and obligations, which shall include surety, equity, and debt obligations, of the concern and employees of the concern to a foreign country, foreign person, or foreign entity; foreign affiliations of a covered individual, owner, or other key personnel of a concern with an entity in a foreign country of concern; investment relationships of a concern with an individual or entity in a foreign country of concern; technology licensing agreements or joint ventures, including joint venture like agreements, with an individual or entity in a foreign country of concern; and business relationships between a covered individual, owner, or other key personnel and an individual or entity in a foreign country of concern.
The DAF will also assess proposals utilizing open-source analysis and analytical tools for the nondisclosures of the information set forth in 15 U.S.C. 638(g)(13) or 638(o)(17), and examine any relationship of a concern seeking an award to any entity or individual included on the lists described in 15 U.S.C. 638(g)(16)(D) and 638(o)(20)(D).
If the DAF assesses that a concern has security risks, the DAF will review the proposal, the evaluation, and the security risks and may choose either to create a plan to mitigate the risks or to not select the proposal for award based upon a totality of the review.
Applicants will be required to disclose, under penalty of perjury, the representations, attestations, and certifications required under 15 U.S.C. 638(g)(13) and 638(o)(17), fulfilled by completing Volume 7. In addition, a written statement with any substantial changes to a foreign disclosure form must be provided to the awarding agency within 30 days of any changes while on a project for the DAF.
The DAF will not make an award if it determines the concern has an owner or covered individual that is party to a malign foreign talent recruitment program; has a business entity, parent company, or subsidiary located in the People's Republic of China or another foreign country of concern; has an owner or covered individual that has a foreign affiliation with a research institution located in the PRC or another foreign country of concern; or has a security risk connecting the concern to an entity, including any affiliates of the entity, or individual on any of the following: the UFLPA Entity List maintained by the Department of Homeland Security; the Non-SDN Chinese Military-Industrial Complex Companies List of the Office of Foreign Assets Control maintained by the Department of the Treasury; the Section 889 Prohibition List established under section 889 of the John S. McCain National Defense Authorization Act for Fiscal Year 2019 and maintained by the Department of War; the list of Chinese Military Companies required under section 1260H of the William M. Thornberry National Defense Authorization Act for Fiscal Year 2021 and maintained by the Department of War; the Military End User List maintained by the Bureau of Industry and Security of the Department of Commerce; the Entity List maintained by the Bureau of Industry and Security; the List of Equipment and Services maintained by the Federal Communications Commission; or the Withhold Release Orders and Findings List maintained by U.S. Customs and Border Protection.
The DAF will also not make an award if it determines the concern has a security risk with a primary source that is classified or has a security risk the DAF determines warrants a denial.
If an award is denied on these grounds, the DAF will, as appropriate pursuant to its discretion and in a manner that does not compromise security, provide a notification advising the small business of the determination, the basis for the determination, and a statement that denial of award does not prohibit the concern from being eligible for an award in a subsequent award cycle. The same applies to denials under 15 U.S.C. 638(vv).
The mandatory denial for foreign affiliation with a research institution in a country of concern is a first-order diligence item on an STTR award with a required academic partner. Work through your partner's personnel affiliations and international collaborations before you propose.
Support contractors, status, feedback, and protests
Restrictive notices notwithstanding, proposals may be handled for administrative purposes only by support contractors, which may include but are not limited to TEC Solutions, Inc., APEX, Oasis Systems, Riverside Research, Peerless Technologies, HPC-COM, Mile Two, Montech, Wright Brothers Institute, and MacB (an Alion Company). In addition, only Government employees and technical personnel from the FFRDCs MITRE and Aerospace Corporations working under contract to provide technical support to AF Life Cycle Management Center and Space and Missiles Centers may evaluate proposals. All support contractors are bound by appropriate non-disclosure agreements. Contact the DAF SBIR/STTR Contracting Officer with concerns regarding the use of support contractors.
The Principal Investigator and Corporate Official indicated on the Proposal Cover Sheet will be notified by email regarding proposal selection or non-selection. Small businesses will receive a notification for each proposal submitted, so read each notification carefully and note the Proposal Number and Topic Number referenced.
Automated feedback will be provided for Phase I proposals designated Not Selected. Additional feedback may be provided at the sole discretion of the DAF.
Proposals submitted to the DAF are received and evaluated by different organizations, handled by topic. Each organization operates within its own schedule for proposal evaluation and selection, so updates and notification timeframes will vary. If contacted regarding a proposal submission, it is not necessary to request information regarding additional submissions.
The Air Force anticipates that all proposals will be evaluated and selections finalized within approximately 90 calendar days of solicitation close. Refrain from contacting the BAA Contracting Officer for proposal status before that time.
Refer to the FY26 STTR BAA for procedures to protest the Announcement. As further prescribed in FAR 33.106(b) and FAR 52.233-3, protests after award should be submitted to Air Force SBIR/STTR Contracting Officer Daniel J. Brewer.
All final reports will be submitted to the awarding DAF organization in accordance with contract instructions. Companies will not submit final reports directly to the Defense Technical Information Center.
How Phase II works
DAF organizations may request Phase II proposals while Phase I technical performance is ongoing or at any time after the conclusion of the period of performance. This decision will be based on the awardee's technical progress, as determined by a DAF Technical Point of Contact review using the Phase II review criteria.
Phase II is the demonstration of the technology found feasible in Phase I. Only Phase I awardees are eligible to submit a Phase II proposal. All Phase I awardees will be sent a notification with the Phase II proposal submittal date and detailed preparation instructions. If the physical or email addresses or firm points of contact have changed since submission of the Phase I proposal, correct information shall be sent to the AF SBIR/STTR One Help Desk. Phase II dollar values, performance periods, and proposal content will be specified in the Phase II request for proposal. Only one Phase II proposal may be submitted for each Phase I award.
The DAF reserves the right to modify the Phase II submission requirements, and should the requirements change, all Phase I awardees will be notified. The DAF also reserves the right to change any administrative procedures that will improve management of the DAF SBIR/STTR Program at any time.
Because the Phase II request can arrive while Phase I is still running and is based on demonstrated technical progress, producing real simulation results early rather than saving them for the final report is worth planning for on a six month award.
Timeline and What to Do When
The dates
Topic opens: September 23, 2026
Proposal deadline: October 21, 2026, at the time specified in the FY26 STTR BAA, ordinarily 12:00 p.m. Eastern. Confirm on DSIP.
Selections finalized: within approximately 90 calendar days of close, on or about January 19, 2027
Period of performance: up to 6 months from award
A working backward plan
Before September 23. Settle eligibility: if venture capital, hedge fund, or private equity funds hold majority ownership in part, you are excluded under this document, so resolve your status first. Contact your research institution partner's sponsored programs office and get their internal review timeline plus their formal budget with federally negotiated rates, and confirm the 40/30 split closes with no deviation available. Begin the Allocation of Rights Agreement conversation. Read the evaluation criteria in the FY26 STTR BAA. Work through the eleven references, especially the McQuade metrics paper, the OUSD KPI report, and the April 2026 AFTC article. Choose your simulation environment from AFSIM, TETK, or JSE and establish what access requires. Identify your surrogate sensor datasets, open-source or synthetic, and your representativeness argument. Ask the One Help Desk which of the two Phase I descriptions governs. Request a personnel roster with citizenship and visa status from your partner. Complete Volume 6 training and verify SAM and DSIP alignment.
September 23 through October 3. Draft the technical volume against the 20 page limit. Lead with the AFTC gap, then your framework architecture with data ingestion pipelines explicitly defined and computational methods for reliability, robustness, and explainability metrics detailed, then the surrogate data strategy, then the sandbox simulation plan. Address real-time inference latency, adversarial resilience, and data drift detection as your named metric set. Write the work plan outline separately with no proprietary content, since it becomes the Statement of Work.
October 4 through October 12. Build the cost volume and itemized listing, neither of which counts against the page limit. Verify the 40/30 split in the numbers. Get executed agreements with statements of work and detailed cost volumes for your partner and every consultant. Include the recommended trip to the Air Force location managing the contract. Write the detailed TABA request into Volume 5 if requesting TABA, and consider intellectual property protections as the activity.
October 13 through October 16. Assemble Volume 5 with Technical Data Rights Assertions, complete the Volume 7 webform, finish the Volume 4 CCR, and run compliance: 20 page limit, 10-point minimum type, one-inch margins, graphics distinguishable in black and white, no PII uploaded to DSIP, no proprietary content in the technical abstract or the work plan outline.
October 17 through October 19. Submit and certify in DSIP.
October 20 through October 21. Buffer only.
Frequently Asked Questions
What is DAF STTR topic DAF26TZ06-NV006?
DAF26TZ06-NV006 is a Department of the Air Force STTR Phase I topic titled "Artificial Intelligence Performance Evaluation Tool," released under the DAF 2026 STTR Broad Agency Announcement, Release 6. It seeks to design, develop, and demonstrate a tool that streamlines, standardizes, and enhances the evaluation of AI-enabled systems across various operational conditions, with a targeted application in sensor and information systems used in test and evaluation missions under the 412th Test Wing of the Air Force Test Center.
How much funding is available under DAF26TZ06-NV006?
Up to $300,000 for a period of performance up to 6 months. Proposals in excess of that amount or duration will not be considered for evaluation or award. Up to $6,500 in Technical and Business Assistance is available per Phase I award, in addition to the per-topic total.
When is the proposal deadline?
The topic closes October 21, 2026. The DAF instructions direct offerors to the FY26 STTR BAA for the exact submission time, ordinarily 12:00 p.m. Eastern on the close date. Confirm on the live DSIP posting.
When does this topic open?
September 23, 2026, giving a 29 day submission window.
Are venture capital backed companies eligible?
No, under this document. The instructions state that firms owned in majority part by multiple venture capital operating companies, hedge funds, or private equity funds are not eligible for Department of Air Force topics. This provision is not uniform across DAF Release 6: all three Commercial Solutions Opening documents permit such firms, while three of the four Broad Agency Announcement documents exclude them, the SBIR BAA Phase I instructions being the exception. The exclusion appears in the document governing this topic, so treat it as operative.
Do I need a research institution partner?
Yes. This is an STTR topic. For both Phase I and Phase II, a minimum of 40 percent of each STTR award must be conducted by the awardee and a minimum of 30 percent of the effort must be performed by the single partnering research institution.
Can I use more than one research institution?
You may partner with multiple entities that separately meet the definition of a research institution as indicated in the STTR BAA, but you may use only one partnering research institution to meet STTR eligibility requirements. Designate which one deliberately.
Can I get a waiver on the work split?
No. The instructions state that the DAF will not consider requests for deviations to these performance of work requirements, and the Subcontracts section states that deviations are not permitted. This differs from several other DAF Release 6 documents, which provide a deviation process.
How is the work percentage calculated?
The STTR funded work percentage calculation considers both direct and indirect costs after removal of the small business concern's proposed profit.
What is the Allocation of Rights Agreement?
A required agreement governing intellectual property between the small business and the research institution. The instructions direct you to the DoW STTR FY26 BAA for more information. Your partner's technology transfer office will need to negotiate it, so start early.
What must the tool be capable of evaluating?
AI-enabled sensor suites, with electro-optical and infrared, radar, and radio frequency systems given as examples. This is not a generic model evaluation framework; it must handle AI operating on sensor data.
Which simulation environments must I align with?
The tool must align with AFTC's primary simulation and digital engineering environments, specifically the Advanced Framework for Simulation, Integration, and Modeling (AFSIM), the Ansys Test and Evaluation Tool Kit (TETK), or the Joint Simulation Environment (JSE). You may choose one, but you must choose one and demonstrate the alignment.
What are the three critical robustness gaps?
Real-time inference latency, adversarial resilience, and data drift detection. Those recur in the Phase I research task, so treat them as your core metric set.
Will the Air Force give me flight test data?
No. The Phase I research task requires you to establish and utilize open-source or synthetically generated surrogate sensor datasets that are representative of AFTC flight test scenarios, and the deliverables list names a Surrogate Data Strategy and Baselining as a possible component. Name your datasets or synthesis approach and argue their representativeness.
What technology readiness levels apply?
Phase I is initial TRL 2 with a target of TRL 4. Phase II is initial TRL 4 with a target of TRL 6. Upon completion of Phase II the tool is expected to reach TRL 6 with a clearly defined transition pathway toward operational implementation across the 412th Test Wing and other DoD components.
What are the Phase I deliverables?
A feasibility report describing how the tool can practically operate within the AFTC environment, a detailed framework design document, and initial simulation results demonstrating concept validity. This could include Surrogate Data Strategy and Baselining, algorithmic and mathematical validation, architectural alignment and computational overhead, and Phase II scalability and transition pathway.
Why does the topic seem to describe Phase I twice?
The topic contains two Phase I descriptions and two Phase II descriptions that do not fully align. The first pair is written around the AI evaluation framework with explicit TRL targets. The second pair, appearing after a Conclusion paragraph, is written around flight test analysis workflows, hardware and software architectural design, Python-based extensibility, and Time Space Position Information processing accuracy, which reads as though it may have come from a related but distinct topic. Write primarily to the first, more detailed pair while addressing compatible elements of the second, and ask the DAF SBIR/STTR One Help Desk which governs.
What does the architecture requirement specifically demand?
Develop a software architecture for an adaptable AI performance evaluation framework addressing reliability, robustness, and explainability metrics, based on widely accepted DoD digital engineering and telemetry standards. The architecture must explicitly define the assumed data ingestion pipelines for surrogate sensor data and detail the computational methods that will be used to calculate reliability, robustness, and explainability metrics. Note that both the pipelines and the computational methods must be defined and detailed, not described generally.
Who is the Phase III transition partner?
The 416th Test Squadron at Edwards Air Force Base, which specializes in F-16 flight testing and has a history of integrating advanced systems, will serve as the dedicated transition partner for this effort. That is an unusually specific transition pathway for a Phase I topic and worth referencing throughout your proposal.
How long can my technical volume be?
20 pages or slides, per the topic index. The Technical Volume must be no smaller than 10-point on standard 8.5 by 11 inch paper with one-inch margins. The Cover Sheet, Cost Volume, and Cost Volume Itemized Listing do not count toward the limit; only the Technical Volume and any enclosures or attachments do. Documents uploaded into Volume 5 under "Other" do not count either.
Is there a required proposal template?
No. Unlike the DAF STTR CSO Phase I instructions in this same release, which require a government template, this BAA document does not. You build the technical volume yourself, subject to the required key personnel and work plan outline elements.
Is this topic export controlled?
This topic does not carry a topic-level ITAR and EAR restriction paragraph. Your general disclosure obligations still apply, including identifying all U.S. permanent residents and all non-U.S. citizens involved with countries of origin, visa or work permit type, and anticipated level of involvement.
How will my proposal be evaluated?
This document does not restate the evaluation criteria. It says only that proposals will be evaluated for overall merit in accordance with the criteria discussed in the FY26 STTR BAA. Read those criteria before writing, and do not assume the Criteria A, B, and C structure from the DAF CSO instructions carries over.
Does the Company Commercialization Report affect my score?
No. Completing the CCR as Volume 4 in DSIP is required, but the instructions state information contained in the CCR will not be considered by the Air Force during proposal evaluations.
Why does the work plan outline matter so much?
Because the DAF uses it as the initial draft of the Phase I Statement of Work. The instructions direct you not to include proprietary information in it, warning that doing so will necessitate a request for revision if selected and may delay award.
What is TABA and how do I request it?
Up to $6,500 per Phase I award, in addition to the per-topic total, identified in the Volume 3 Cost Proposal. This BAA document carries the full detailed request requirements: provider name, point of contact with email and phone, an explanation of unique qualifications, the tasks with purpose and objective, and total cost with hours and labor rates, with the task milestone list tracking to your milestone payment schedule. The detailed request must be in Volume 5, and requests specifying only a value in Volume 3 will not be considered.
What references should I read?
Eleven are cited, and they map the standards landscape. The most directly useful are the McQuade et al. paper on metrics for trustworthy AI evaluating reliability, robustness, and explainability, which maps onto the topic's three metric categories; the OUSD report on Key Performance Indicators for AI and Autonomy Testing; the DoD Responsible AI Strategy and Implementation Pathway; the NIST AI Risk Management Framework; and an April 15, 2026 AFTC news article, "From weeks to minutes: How AI is accelerating the flight test process," which is recent and specific to this customer.
How do I get to Phase II?
DAF organizations may request Phase II proposals while Phase I technical performance is ongoing or at any time after the period of performance concludes, based on the awardee's technical progress as determined by a DAF Technical Point of Contact review. Only Phase I awardees are eligible, and only one Phase II proposal may be submitted for each Phase I award.
When will I hear about selection?
The Air Force anticipates all proposals will be evaluated and selections finalized within approximately 90 calendar days of solicitation close, meaning on or about January 19, 2027. The Principal Investigator and Corporate Official on the cover sheet are notified by email.
Who do I contact with questions?
The DAF SBIR/STTR One Help Desk at usaf.team@afsbirsttr.us for general program and proposal preparation questions, and the DAF encourages requesting clarifying information as early as possible because delays constrain its ability to respond. For DSIP submission system issues, dodsbirsupport@reisystems.com. For technical questions about the topics during the pre-announcement and open period, reference the FY26 STTR BAA. The Air Force SBIR/STTR Contracting Officer is Mr. Daniel J. Brewer at Daniel.Brewer.13@us.af.mil. Address or point of contact changes after submission go to the One Help Desk with the subject line "FY26 STTR BAA Address Change."
Positioning Advice for Companies Considering This Topic
Clear the two gates before anything else. Venture-backed majority ownership disqualifies you under this document, and the 40/30 STTR split is not waivable. Both are threshold questions, and the second one means your research institution partner has to be real and has to be doing at least 30 percent of the work.
Pick your simulation environment and prove you know it. AFSIM, TETK, or JSE. Alignment with one of the three is a stated requirement, and it is the fastest way to separate a proposal written by someone who understands AFTC from one written from the outside. If your team lacks that experience, your research institution partner may be where you get it.
Make the surrogate data strategy a centerpiece. You will not get real flight test data, and the topic requires representative open-source or synthetic surrogates. Naming specific datasets, describing your synthesis approach, and arguing representativeness against AFTC flight test scenarios is where a credible proposal distinguishes itself from a hand-wave.
Detail the computational methods, because the topic asks you to. The architecture must explicitly define data ingestion pipelines and detail the computational methods for reliability, robustness, and explainability metrics. Vague references to standard techniques will not satisfy that. Show the math.
Report computational overhead honestly. It appears in the deliverables list, and a framework that produces beautiful metrics at prohibitive cost is not deployable in a test environment. Characterizing overhead is a credibility signal.
Frame the value as unblocking AI fielding. The topic says these challenges prevent the Air Force from leveraging AI advancements at scale, impacting readiness and modernization goals. That is the argument: evaluation is the bottleneck, and your tool removes it. Fidelity, speed, and consistency of assessments are the three named improvement axes.
Use the named customer chain. The 412th Test Wing test environment, the Air Force Test Center, the 416th Test Squadron at Edwards as the dedicated transition partner, and ABMS alignment are all in the topic. Very few Phase I topics hand you a specific squadron. Build your transition narrative on it.
Take explainability seriously, not decoratively. The reference list includes DARPA XAI, the DIU Trusted AI Framework, and the McQuade metrics paper. The topic asks for quantifiable explainability scores from the sandbox simulation. That is a measurable output, not a discussion section.
Ask about the duplicated phase descriptions. The two Phase I and two Phase II descriptions in this topic do not align, and one pair mentions hardware architecture and TSPI accuracy that fit a different kind of effort. A quick Help Desk question in the first week could save you from writing to the wrong specification.
Start the Allocation of Rights Agreement now. It is required, your partner's technology transfer office has to be involved, and TABA can fund intellectual property support. On a six month award, an IP dispute discovered at contract negotiation is expensive.