Joint Beamforming for NLOS Target Detection with RIS-Assisted Radar in Multipath Environments

Reconfigurable intelligent surfaces (RIS) have gained prominence in communications and are now extending their applications to radar, offering innovative solutions for non-line-of-sight (NLOS) target detection. This paper introduces a new joint transmit and RIS beamforming approach for NLOS target detection using RIS-assisted radar systems operating in multipath environments. Leveraging manifold-based optimization, the proposed method jointly optimizes the radar's active beamformer and the RIS phase shifts to enhance detection robustness by managing the interference and signal degradation caused by multipath propagation. We provide a detailed analysis of detection performance, including the probability of detection and probability of false alarm, to assess the effectiveness of the approach. The proposed method is further compared with a conventional approach that focuses solely on the line-ofsight (LOS) component of the multipath signals. While the conventional method performs comparably in scenarios with a strong dominant LOS path, the proposed approach demonstrates notably superior performance across more general multipath environments. Simulation results underscore the applicability of the proposed method, particularly in environments with high angular spread, highlighting RIS's potential for improved detection in complex, multipath-dense settings where traditional approaches may be limited.