Jamming and classification of drones using full-duplex radios and deep learning

Abstract

The emerging full-duplex (FD) radio concept is set to double the spectral efficiency of commercial wireless networks, but it also has potential applications in the defense and security domains. In the form of multifunction military full-duplex radios (MFDRs), the FD capability could enable armed forces to conduct simultaneous electronic attacks, electronic support measures, and tactical communications. This paper demonstrates the feasibility of simultaneous jamming and reconnaissance of drones' remote control (RC) systems using a prototype MFDR. Alongside, we apply deep learning in the form of a convolutional neural network (CNN) for classifying the RC signals and analyze the effect of FD operation on the classification performance.