Combining MIMO DBF With Automotive Synthetic Aperture Radar Imaging and Phase Error Correction

Abstract

Automotive radar systems face challenges in generating high-resolution images that are essential for advancing autonomous driving technology. One promising solution to improve the angle resolution of radar images is the synthetic aperture radar (SAR) technique. However, achieving satisfactory SAR images involves overcoming difficulties such as high computational burden and accurate platform location determination. To address these challenges, we propose an innovative approach that integrates SAR imaging with digital beamforming (DBF) and multiple input multiple output (MIMO) techniques. The proposed approach significantly reduces the computational time required for SAR image formation and demonstrates superior phase error suppression compared to conventional methods. Our implemented algorithm reduces the number of radar samples and imaging complexity by up to a factor of 10 without compromising resolution and image quality. Furthermore, our proposed angle variant phase correction method can be used in challenging automotive scenarios to efficiently mitigate the effects of platform position inaccuracies and undesirable motions. Through simulations and practical experiments, we present promising results to highlight the advantages of combining real and synthetic apertures for radar imaging and phase error correction.

Real-Time Synthetic Aperture Radar Imaging with Random Sampling Employing Scattered Power Mapping

“…To this end, a few image-reconstruction algorithms based on the back-projection (or back-propagation) algorithm (BPA) have been proposed [ 24 , 27 , 29 , 47 , 48 ]. The BPAs, when applied in real space, can inherently handle any sampling position; thus, they are preferred in the case of nonuniform and random sampling.…”