Sparse Reconstruction of Chirplets for Automotive FMCW Radar Interference Mitigation

TL;DR

This paper introduces a novel sparse reconstruction method using chirplet basis and OMP algorithm to effectively identify and mitigate FMCW radar interference in automotive scenarios, enhancing signal clarity.

Contribution

It presents a new blind interference mitigation technique based on sparse reconstruction with chirplet basis and OMP, specifically designed for automotive FMCW radar interference.

Findings

Significant reduction of noise and interference in simulated data.

Effective interference removal in real automotive sensor measurements.

Method preserves target signal integrity with minimal information loss.

Abstract

Mutual interference in automotive radar scenarios is going to become a major concern as the density of vehicles with radar sensors in the roads increases. The present work tackles the problem of frequency modulated continuous wave (FMCW) to FMCW and continuous wave interference. In this context, we propose a signal processing technique to blindly identify and remove interference by using the fast Orthogonal Matching Pursuit (OMP) algorithm to project the interference signals in a reduced chirplet basis, and separate it from the target signal with minimal loss of information. Significant reduction of the noise-plus-interference levels are observed in both simulated and measured data, the later acquired with state of the art automotive sensors.