Nonlinear Estimation for Ultra-Wideband Radar Based on Bayesian Particle Filtering Detector

Abstract

In order to combat the nonlinearity of the radio component in UWB radar systems, in this paper we present a promising blind estimation algorithm based on the particle filtering (PF). Based on the conception of Bayesian approximation and sequential importance sampling, this appealing Monte-Carlo method can deal with many complicated statistic estimation problems. In sharp contrast to the classical linear equalization problem, nevertheless, in this considered problem the PF based method may become valid due to the nonlinearity and the resulting non-analytic importance function. Thus, a novel PF framework based on the linearization technique is suggested, and we show in particular how to linearize the involved nonlinearity transform. The merit of this method is that it can deal with discrete time dynamic models that are typically nonlinear and non-Gaussian. Experimental simulations demonstrate the superior performance of the presented PF scheme, which may be properly applied to UWB radar systems.