Frequency-swept target recognition by kernel principal component analysis on radar cross-section

In this article, frequency-swept target recognition is given by kernel principal component analysis on radar cross-section. The goal of frequency-swept technology in this study is to reduce the number of measuring locations, which may be difficult to obtain in practical applications. Kernel principal component analysis has been widely applied to different fields of signal processing. It utilizes the kernel technique so that one does not need to know the explicit form of nonlinear mapping for data projection. This article combines frequency-swept radar cross-section and kernel principal component analysis techniques to implement target recognition. Numerical examples show that the proposed recognition scheme is very efficient and accurate and can well tolerate random fluctuations of radar cross-section. Of great importance, the number of measuring locations is greatly reduced due to the use of frequency-swept technologies, and one can still obtain accurate target recognition results.