Low-Sidelobe Imaging Algorithm Based on Joint Low-Rank and Sparse Priors for Small-Array UWB Radar

In order to enhance portability and reduce system complexity, small-array ultra-wideband (UWB) radar is employed for target detection. However, the radar images can be distorted due to high sidelobes. Traditional coherent weighting methods have limited effect on the suppression of sidelobes. Compressed sensing (CS) methods are accompanied by high computational effort which are difficult to utilize in practice. Additionally, noise interference results in defocusing of radar images under low signal-to-noise ratio (SNR) conditions. To address these challenges, we propose a low-sidelobe imaging algorithm based on joint low-rank and sparse priors (JLRSP). By simultaneously considering the norm constraints on target characteristics and noise level, the solution model is constructed. According to the alternating direction method of multipliers (ADMM), iterative solutions for each variable are derived, which successfully reconstruct low-sidelobe images under strong interference. Furthermore, the proposed algorithm effectively avoids multiple high-dimensional matrix multiplications and inversion operations, significantly reducing computational complexity. The image entropy is used as an evaluation metric to ensure the appropriate parameter selection and iteration termination. Both simulations and experiments validate the fast low-sidelobe imaging capability of the proposed algorithm and its strong robustness to noise.