Research on Radar Emitters Classification with Fuzzy Support Vector Machines

In this paper, a novel method based on Kernel Principle Component Analysis is proposed to extract features of radar emitter signals image of Choi-Williams distribution. Then these discriminative and low dimensional features obtained were fed to the classifier designed for different radar LFM signals which is based on fuzzy Support Vector Machines (FSVMs). In simulation experiments, the classifier attains over 90% overall average correct classification rate. Experimental results show that the proposed FSVM classifier is efficient for different complex radar signals detection and classification.