Robust Algebraic Approach for Radar Signal Processing: Noise Filtering, Time-Derivative Estimation and Perturbation Estimation

Abstract

In this paper, we propose a robust algebraic approach to deal with noisy signal from radar in the context of Stop-and-Go scenarios which need inter-distance measurements for the control law. In general, commercial radars for cars are of such low quality that the measured signal is very noisy and the performance is not good enough to be used directly in the control law. In addition to this, a constants bias and a distance proportional perturbation factor, which varies during the lifetime of radar, can corrupt the signal too. When the signal is noisy, its time derivative is specially corrupted. Hence, an algebraic nonlinear estimation technique is proposed to deal with filtering, estimating time derivatives and perturbation factors. It is important to point out that these filters, differentiators and estimators are not of asymptotic nature, and do not require any statistical knowledge of the corrupting noises.