OUSD (R&E) CRITICAL TECHNOLOGY AREA(S): Trusted AI and Autonomy; Advanced Computing and Software The technology within this topic is restricted under the International Traffic in Arms Regulation (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 the Export Administration Regulation (EAR), 15 CFR Parts 730-774, which controls dual use items. Offerors must disclose any proposed use of foreign nationals (FNs), their country(ies) of origin, the type of visa or work permit possessed, and the statement of work (SOW) tasks intended for accomplishment by the FN(s) 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. OBJECTIVE: The objective is to develop, analyze, and deploy artificial intelligence (AI) or machine learning (ML) based techniques for the detection and geolocation of weak/obscure RF emitters from a single Collaborative Combat Aircraft (CCA) platform in the presence of strong emitters for improved battlespace awareness. DESCRIPTION: Traditional geolocation is often accomplished using direction of arrival (DOA) estimation in the single platform case, or time difference of arrival (TDOA)/frequency difference of arrival (FDOA) in the multiple platform case. Multiple platform geolocation has improved geolocation accuracy compared to single platform geolocation but suffers from issues such as slow computation speed and the requirement of precise time synchronization between platforms [1]. These drawbacks make single platform geolocation preferred in a time critical environment such as a battlespace. Emitter detection sensitivity is an important factor for detecting weak emitters and is often increased by deploying large, high-powered antennas on large unmanned aerial vehicles (UAV). With the previously mentioned time constraints, it is much more feasible to deploy small UAVs with reasonable size, weight, and power (SWaP) requirements. Unfortunately, conventional single platform geolocation techniques suffer from poor resolution and are insufficient for the detection and geolocation of weak emitters in the presence of strong emitters when the antenna array and platform is small [2]. AI/ML advancements have dominated the DOA estimation literature in recent years [3-6]. These techniques have been shown to significantly improve the DOA estimation performance compared to conventional methods, and in many cases reach the theoretical minimum achievable error dictated by the Cramer-Rao bound (CRB) [7]. Despite these advancements with AI/ML techniques, there is a lack of research that progresses from the DOA estimation case to single platform geolocation. Single platform geolocation requires a method of combining multiple DOA estimates accurately and efficiently to achieve geolocation estimates, which is absent from current AI/ML DOA literature. This topic therefore aims to investigate successful AI/ML techniques for DOA estimation that can be expanded for our geolocation scenario. Investment End State: A software prototype for AI/ML based detection and geolocation of weak emitters in the presence of strong emitters from a single CCA platform. PHASE I: This topic is intended for technology proven ready to move directly into Phase II. Therefore, a Phase I award is not required. The offeror is required to provide detail and documentation in the Direct to Phase II proposal which demonstrates accomplishment of a Phase I-like effort. This includes a detailed technical approach to achieving the goal of improved detection and geolocation of weak RF emitters in the presence of strong emitters from a single CCA platform. For this topic, we are classifying weak emitters as having a signal-to-noise ratio (SNR) at least 20 dB below the SNR of the strong emitters. Offeror's previous success with AI/ML based DOA estimation and a clear plan to expand from DOA estimation to single platform geolocation is preferred. The proposed geolocation approach should provide solutions and any constraints for the application. PHASE II: Eligibility for a Direct to Phase Two (D2P2) is predicated on the offeror having performed a Phase I-like effort predominantly separate from the SBIR/STTR Programs. Offerors implement the algorithmic approach to geolocation outlined in PHASE I. Offerors are expected to create a proof of concept for their algorithm with simulated data and compare their approach with other conventional geolocation techniques and theoretical bounds such as the CRB. Extensive Monte Carlo simulations with metrics such as detection success rate and geolocation accuracy should be carried out to demonstrate algorithm performance. Offerors should demonstrate computational efficiency of algorithm compared to conventional approaches. PHASE III DUAL USE APPLICATIONS: The geolocation algorithm will be implemented on a CCA platform and tested in scenarios of interest. Phase III shall provide a business plan and address the ability to transition technology and system concepts to commercial applications. The adapted non-Defense commercial solutions shall provide expanded mission capability for a broad range of potential Governmental and civilian users and alternate mission applications. Integration and other technical support to operational users may be required. REFERENCES: 1. S. Management, Comparison of time-difference-of-arrival and angle-of-arrival methods of signal geolocation, tech. rep., ITU-R, 2018. 2. Li, F.; Liu, H.; Vaccaro, R.J. Performance analysis for DOA estimation algorithms: Unification, simplification, and observations. IEEE Trans. Aerosp. Electron. Syst. 1993, 29, 1170 1184. 3. H. Huang, J. Yang, H. Huang, Y. Song and G. Gui, "Deep Learning for Super-Resolution Channel Estimation and DOA Estimation Based Massive MIMO System," in IEEE Transactions on Vehicular Technology, vol. 67, no. 9, pp. 8549-8560, Sept. 2018. 4. Z. -M. Liu, C. Zhang and P. S. Yu, "Direction-of-Arrival Estimation Based on Deep Neural Networks with Robustness to Array Imperfections," in IEEE Transactions on Antennas and Propagation, vol. 66, no. 12, pp. 7315-7327, Dec. 2018. 5. A. M. Elbir, "DeepMUSIC: Multiple Signal Classification via Deep Learning," in IEEE Sensors Letters, vol. 4, no. 4, pp. 1-4, April 2020. 6. S. Feintuch, J. Tabrikian, I. Bilik and H. Permuter, "Neural-Network-Based DOA Estimation in the Presence of Non-Gaussian Interference," in IEEE Transactions on Aerospace and Electronic Systems, vol. 60, no. 1, pp. 119-132, Feb. 2024. 7. P. Stoica and A. Nehorai, Music, maximum likelihood, and cramer-rao bound, IEEE Transactions on Acoustics, Speech, and Signal Processing, vol. 37, no. 5, pp. 720 741, 1989. KEYWORDS: Geolocation; RF Emitters; Artificial Intelligence; Machine Learning; Weak/Obscure Emitter Detection, Single Platform; Air Platforms
