Researchers at the Naval Surface Warfare Center, Crane Division (NSWC Crane) have released Crane’s first open source project to provide warfighters with more advanced systems. The open source software project was developed to support the Defense Advanced Research Projects Agency (DARPA) Radio Frequency Machine Learning Systems (RFMLS) program.
“The DARPA RFMLS program is developing the foundations for applying neural networks to RF signal processing problems,” said Paul Tilghman, the former RFMLS Program Manager in the Microsystems Technology Office at DARPA. “Creating a strong community that builds on each other’s work requires open source tools and standards. RF, unlike many other sensor modalities, doesn’t have existing datasets and because of its unique characteristics, requires a special standard for that data. The contributions of the NSWC Crane team to the SIG-MF standard are an important aspect of achieving this goal.”
Applying machine learning (ML) to Radio Frequency (RF) data is an emerging field of research. The RF spectrum is becoming increasingly crowded with phones, appliances, drones, traffic lights, security systems, environmental sensors, and other radio-connected devices – and the use of ML will aid in identifying signals in order to gain situational awareness.
“Our software contributions reflect Crane’s participation in this exciting field of research,” says Zachary Davis, an engineer at NSWC Crane. “An open source software project is one that is freely available to the public and able to be modified and improved upon by the community. We have used the ‘master’ software for RFMLS purposes, modifying and expanding its capabilities in the process. By integrating our modifications back into the ‘master’ version of the software, we are allowing the RF machine learning community at large, comprised of academia, government, and industry, to leverage the work that we have done in order to keep pushing this area of research forward.”