The Defense Advanced Research Projects Agency plans to develop a machine learning system that incorporates abilities to read, discern, automatically adjust and learn through radio frequencies as part of a new program.
DARPA said Friday Paul Tilghman, manager for the Radio Frequency Machine Learning Systems program within the Microsystems Technology Office, believes there is a need to integrate radio-frequency capabilities to machine learning so that artificial intelligence may be used to monitor radio signals that could present a threat.
“We want to be able to understand and trust what is happening in the Internet of Things and to stand up an RF forensics capability to identify unique and peculiar signals among the proverbial cocktail party of signals out there,” he said.
The program consists of four technical components that focus on signal feature learning, detection of the important visual and auditory stimuli, automatic signal reception adjustment and waveform synthesis for any device, DARPA added.
The agency said in a FedBizOpps notice posted Friday it will hold a Proposers Day on Aug. 31 to discuss the RFMLS program, which will have three technical areas covering the development of algorithms for RF forensics and situational awareness as well as an RF system integrator.
DARPA intends to award up to $4 million annually per award for Technical Areas 1 and 2 and up to $1.5 million annually per award for Technical Area 3.
The notice stated that DARPA will accept proposals through Oct. 10 for work that is scheduled to commence April 2018.