Detection and estimation of moving targets based on fractional Fourier transform

Abstract

The key part of classical MTD is the filter bank based on FFT. Generally, a moving target moves in the way of segment constant acceleration. Thus, its echo can be approximately regarded as a chirp signal, which is a typical nonstationary signal. The traditional Fourier transform cannot handle chirp signals. Therefore, time frequency analysis is a more appropriate method than FFT in such an application. Furthermore, WVD have the best time frequency centralization performance among all the time frequency distributions. Instead of FFT, we can employ WVD to analyze the echo of moving targets. WVD for multiple chirp signals, however, produces inevitably some cross terms, which lead to the appearance of false targets. A method of radar moving target detection and estimation based on the fractional Fourier transform is proposed in this paper. Compared with the MTD based on WVD, the MTD based on FRFT need not consider the cross-term interference in the case of multi moving targets, and therefore reduces the computation burden, simplifies the processing procedure and improves the performance of detection. The experiment results verify validity of the method.