Radar target recognition based on fuzzy clustering

Abstract

A new method of recognizing the aircraft number of a radar target from a narrowband IF signal of non-coherent radar is presented. According to the received narrowband IF echo signal, its autocorrelation matrix is computed. The feature vector is the eigenvalue of the autocorrelation matrix, and the orthogonal transformation is accomplished to remove the unnecessary information in the feature. The Karhunen-Loeve (K-L) transformation is the optimum orthogonal transformation in LMSE. For a stationary Markov process, if p/spl rarr/1, the K-L transformation matrix is the same with the DCT and can be accomplished by the FFT. Based on the K-nearest neighbor classification rule, a new fuzzy K-NN rule is presented. Satisfactory results have been obtained using the above methods.