Signal analysis by means of time-frequency (Wigner-type) distributions-applications to sonar and radar echoes

Abstract

Time series data have been traditionally analyzed in either the time or the frequency domains. For signals with a time-varying frequency content, the combined time-frequency (TF) representations, based on the Cohen class of (generalized) Wigner distributions (WD's) offer a powerful analysis tool. Using them, it is possible to: (1) trace the time-evolution of the resonance features usually present in a standard sonar cross section (SCS), or in a radar cross section (RCS) and (2) extract target information that may be difficult to even notice in an ordinary SCS or RCS. After a brief review of the fundamental properties of the WD, we discuss ways to reduce or suppress the cross term interference that appears in the WD of multicomponent systems. These points are illustrated with a variety of three-dimensional (3-D) plots of Wigner and pseudo-Wigner distributions (PWD). The plots are all obtained from an extensive analysis we have made over the years of the resonance acoustic echoes backscattered by a variety of elastic shells submerged in water, when they are excited by various types of incident pressure waves, including the short pulses generated by explosive charges. We also review studies we have made of the echoes returned by conducting or dielectric targets in the atmosphere, when they are illuminated by broadband radar pings. These short incident pulses are used to analytically model the performance of ultrawide band (UWB) radars, often called impulse radars. A TF domain analysis of these impulse radar returns demonstrates their superior information content.