Radar Emitter Signal Recognition Based on Fusion Entropy Features
Zhibin Yu · Modern Radar · 2010
Aiming at the problem of low degree of recognition in closed radars emitter signal recognition,a novel approach is proposed.In this approach,the fusion feature of wavelet packet reconstruction coefficient(WPRC)including characteristics of radar signal is extracted with the principal component analysis(PCA),and the fusion energy entropy(FEnEn)of the fusion feature and the fusion probability entropy(FPrEn)of the fusion feature are used to construct a feature vector,and the support vector machine is used to identify closed radar emitter signals(ARES)automatically.This approach can achieve very satisfying accurate recognition when signal-to-noise rate(SNR)varies in a large range.Even for SNR=5 dB,the accurate recognition rate of the approximate LFM is 91%.The validity of the approach is demonstrated by experiments.