Statistical modelling of measured automotive radar reflections

Abstract

A statistical analysis was performed on measured radar reflections from a broad range of personal vehicle classes. The outcome of this study is two-fold: 1.) An improved understanding of the radar scattering components of automobiles, which informs the design of surrogate test targets for evaluating automotive pre-collision system (PCS) radars, and 2.) statistical models for evaluating surrogate targets and characterizing target models for PCS radar system designs. We examined the validity of two-parameter distribution models applied to measurements of subject vehicle's radar cross-section (RCS) and found the Weibull distribution to be the best fit. In evaluating the goodness-of-fit of the Weibull distribution model, using the Kolmogorov-Smirnov test, we deem an acceptable fit between the model and the measured RCS data for our intended project outcome.