Dignitas Technologies is proud to announce that we have been awarded Small Business Innovation Research (SBIR) Phase I funding to support research addressing the challenges with radio frequency (RF) signal classification. This work supports the Army's artificial intelligence accelerator objectives under Project Linchpin. Our solution uses an approach for Ensemble Automatic Modulation Classification (EAMC) that considers both temporal and spatial features of RF signals.  

​

Our EAMC approach combines the spatial feature advantages of Convolutional Neural Networks (CNNs) with the support of temporal tracking from Long Short-Term Memory networks (LSTM) for optimal modulation classification. We fuse complementary heterogeneous signal representations to alleviate the inherent challenges with RF signal classification. The advances made with this research and development will support communication and intelligence operations, enhance situational awareness, and contribute to safer decision-making in dynamic environments.