Abstract
Aiming at the problems of the radar emitter signal (RES) recognition based on intra-pulse feature, a novel entropy feature extraction approach is proposed. In this method the sample entropy (SampEn) and fuzzy entropy (FzzyEn) are presented to extract features from RES. The SampEn can measure the complexity of RES from a short signal data, and the FzzyEn is used as a measure of the uncertainty. Feature vectors abstracted from 6 typical RES are used as the input of support vector machine (SVM) classifier to perform the signal recognition. Experimental result indicates that in a large range of SNR the introduced method achieves a good accuracy recognition rate. Simulation verifies the method to be feasible.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
