CUMULANT PROCESSING OF ULTRA WIDEBAND RADAR SIGNALS

Abstract. A method for the estimation of the information parameters of ultra wideband ra-dar signals based on the higher order statistics (HOS) is presented. It is shown that the appli-cation of the fourth order cumulants for parameter estimation of the resonant model allow to reduces the signal-to-noise ratio till the –12 dB. Ultra wideband (UWB) radars include methods and systems for detection, parameters ’ meas-urement and targets ’ identification at application of signals, which spectrum has a ultra wide relative band. In accordance with Baum’s singularity expansion method [1] the electromagnetic radar target response can be decomposed into a finite sum of damped sinusoids and so the natural electromagnetic radiation of objects can be described by the resonant model. The model consists of parameters of two types: dependent on an energizing signal (polarization, form, direction of arrival) and the natural com-plex resonant frequencies of targets [2]. These frequencies are determined by basic geometrical sizes and shape of objects, practically do not depend on a aspect angle and can be used for identification of the radar targets. The natural portion of the response can be used for the aspect-independent active and passive radar target discrimination