Applications of time-frequency and time-scale transforms to ultra-wideband radar transient signal detection

摘要

Abstract: In this paper we use a non-stationary approach andanalyze ultra-wideband (UWB) radar data usingtime-frequency and time-scale transformations. Thetime-frequency transformations considered are theShort-Time Fourier Transform (STFT), the Wigner-VilleDistribution (WD), the Instantaneous Power Spectrum(IPS), and the ZAM transform. Two discreteimplementations of the Wavelet Transform (DWT) are alsoinvestigated: the decimated A- trous algorithm proposedby Holschneider et al, which uses non-orthogonalwavelets; and the Mallat algorithm, which employsorthogonal wavelets. The transients under study are UWBradar returns from a boat (with and without cornerreflector) in the presence of sea clutter, multipath,and radio frequency interferences (RFI). Results showthat all time-frequency and time-scale transformsclearly detect the transient radar returnscorresponding to the boat with a corner reflector.However, as the radar cross section of the targetdecreases (boat without a corner reflector), resultschange drastically as the RFI component dominates thesignal. Simulations show that the Instantaneous PowerSpectrum may be better adapted for localizing thetransient among the time-frequency techniques studied.The decimated A-trous algorithm has the best timeresolution of the techniques studied as the returnappears better localized in the scalogram. !12