Machine Learning Enhanced – RADAR Enhancements and RF Anomaly Classification

The technical objectives represent the follow-on implementation of DeepSig’s Phase I study and further advancements to the system design that enable anomalies to be pulled out from large RF data sets, with a focus on RADAR signal classes. In addition, enhancements to provide for training at the edge, or tactical training will be developed with the requirement that little to no of the original RF dataset will be required. Finally, this effort will explore the ability to classify extremely large swaths of RF spectrum, up to 40 GHz, using only frequency domain data.