Research on Pulse Radar Jamming Recognition based on DE-SVDD

With the rapid development of modern electronic warfare, new jamming based on digital radio frequency memory forwarding emerge in an endless stream. In order to solve the problem that only known jamming can be distinguished, this paper proposes a pulsed radar jamming recognition based on DE-SVDD. The MobileNetV3 is used as the backbone feature extraction network. Then the features are sent to the DE-SVDD module based on the differential evolution algorithm to accept available classes and reject unknown classes, realizing the effective recognition of new class jamming under small samples. The simulation results show that the proposed method has an overall recognition rate of up to 98.3% for seven jamming, including three unknown jamming within the SNR of 0~10dB, and the recognition rate of known types reaches 99.79% under 8dB. The jamming recognition rate reached 91.33%, realizing the effective classification and recognition of unknown modulation methods.