You are here

Machine Learning Waveform Agnostic Electronic Warfare Countermeasures for Army Tactical Radios


TECHNOLOGY AREA(S): Electronics, 

OBJECTIVE: The objective of this research will be to develop and demonstrate countermeasures to electronic warfare (EW) attacks on communication’s systems thereby producing communications systems that have increased EW resilience. 

DESCRIPTION: Historically, US Army tactical radios use hard-coded TRANSEC techniques to counter specific simple EW threats against very specific communications waveforms. With the advent of Software Defined Radios new electronic attack (EA) techniques are increasing in sophistication and being developed and deployed at a rate outpacing waveform development. Artificial Intelligence (AI), and especially Machine Learning techniques, have been demonstrated as applied research (TRL 3-5), that can learn performance of friendly communications waveforms over time, recognize anomalies in behavior, and adapt to overcome EW threats. The US Air Force and Navy have found some success in employing AI against Radar threats. Largely land bases, the US Army is primarily concerned with communications threats, yet much of the same AI might apply for both RADAR and comms. 

PHASE I: The Phase I effort should demonstrate, in a laboratory environment, the ability of the AI algorithms to, based on performance observed over time, learn to recognize when SINCGARS, MUOS, and/or TSM waveforms are experiencing an EW attack (brought on by skilled EA operators using sophisticated modern attack techniques) and respond within five minutes to adapt and counter the attack without a-priori information regarding the nature of the attack. 

PHASE II: Following the precedent of previous Product Manager Electronic Warfare Integration (PdM EWI) Radio Interference Mitigation (RIM) efforts, the Phase II effort should demonstrate (TRL-6/7), within a relevant environment (e.g., Yuma, EPG, etc.) embodiment of this AI solution in a form factor with a clear path to a fieldable tactical solutions that runs independent to any particular radio acquisition effort. 

PHASE III: Phase III should advance the technology maturity to embodiment of this AI solution in a form factor demonstrated an operational environment at operational vehicle speeds and at operational range separations with anticipated range of red EA systems. Commercial applications include robust resilient communications for private security industry, air traffic control, first responders counter terrorism EA against private sector targets. May also likely counter non-malevolent interference events in congested RF environments. 


1: Waveform Agnostic Communications via Deep Learning, DeepSig,, Tim OShea

KEYWORDS: Electronic Warfare, EW, Resilient, Radio, Radio Frequency, Communications, Artificial Intelligence, AI, Waveform Agnostic, Antijam, AJ, Machine Learning, ML, 

US Flag An Official Website of the United States Government