Radar signal recognition algorithm based on entropy theory

Abstract

With the increasingly complex electromagnetic environment of communication, as well as the gradually increased radar signal types, how to effectively identify the types of radar signals at low SNR becomes a hot topic. A radar signal recognition algorithm based on entropy features, which describes the distribution characteristics for different types of radar signals by extracting Shannon entropy, Singular spectrum Shannon entropy and Singular spectrum index entropy features, was proposed to achieve the purpose of signal identification. Simulation results show that, the algorithm based on entropies has good anti-noise performance, and it can still describe the characteristics of signals well even at low SNR, which can achieve the purpose of identification and classification for different radar signals.