Ranging success probability of PPP distributed automotive radar in presence of generalized fading

TL;DR

This paper derives a mathematical expression for the probability of successful object detection by automotive radars, considering realistic fading effects and interference modeled by Poisson point processes, enhancing understanding of radar performance in complex environments.

Contribution

It introduces a novel analysis of ranging success probability incorporating generalized fading models for interference and Rayleigh fading for the desired signal in automotive radar systems.

Findings

Derived closed-form expression for ranging success probability with generalized fading.

Analyzed impact of shadowing and clustering on success probability.

Showed how LOS component influences radar detection in interference-limited scenarios.

Abstract

In automotive radar applications, multiple radars are used in all vehicles for improving the imaging quality. However this causes radar-to-radar interference from neighbouring vehicles, thus reducing the imaging quality. One metric to measure the imaging quality is ranging success probability. The ranging success probability is the probability that a multiple radar system successfully detects an object at a given range, under certain operating conditions. In state-of-the-art literature, closed form expressions for ranging success probability have been derived assuming no fading in desired signal component. Similarly in literature, though distribution of fading in interferers is assumed to be arbitrary, closed form expression is derived only for no-fading assumption in interferers. As fading is always present in a wireless channel, we have derived ranging success probability assuming desired channel experiences the popular Rayleigh fading. And we have assumed generalized $\kappa$-$\mu$ shadowed fading for interfering channels that generalizes many popular fading models such as Rayleigh, Rician, Nakagami-$m$, $\kappa$-$\mu$ etc. The interferers are assumed to be located on points drawn from a Poisson point process distribution. We have also studied how the relationship between shadowing component and number of clusters can affect the impact of LOS component on ranging success probability.