High-Resolution Radar Imaging Jointly Exploiting Multiple Beams with Sidelobes and Grating Lobes

Millimeter-wave (mmWave) radar in beamforming mode offers high-resolution sensing with an enhanced signal-to-noise ratio (SNR). However, in sparse antenna arrays, the effectiveness of beamforming is limited by sidelobe and grating lobe interference, which degrades angular resolution and imaging performance. Traditional algorithms often discard sidelobe and grating lobe echoes as noise or interference, missing valuable information for improved resolution. In this paper, we propose a novel high-resolution imaging algorithm that integrates sidelobe and grating lobe effects into the data processing. By developing an enhanced frequency-modulated continuous wave (FMCW) radar beamforming signal model, we jointly process signals from all beam directions, including both the main lobe and non-main lobe reflections, using a least-squares (LS) approach. This method improves angular resolution, target discrimination, and noise suppression. Simulation results of range-angle map, chamfer distance, and success rate of distinguishing adjacent points demonstrate that our approach outperforms conventional beamforming (CBF) methods and canonical polyadic decomposition (CPD) methods, improving resolution and reducing sidelobe and grating lobe artifacts, offering a promising solution for high-resolution mmWave radar imaging in sparse array configurations.