RIS Assisted Radar-Communication Coexistence System With Discrete Reflection Coefficients

The advent of reconfigurable intelligent surfaces (RIS) introduces a novel paradigm to address mutual interference issues within radar-communication coexistence (RCC) systems. However, the predominantly discrete reflection coefficients (DRC) of most real-world RISs pose new challenges for the design of interference-mitigating reflection coefficients. This paper investigates a RIS-assisted RCC system with DRC. First, a radar SINR maximization problem under the constraints of communication signal-to-interference plus noise ratio (SINR) requirements is formulated to obtain feasible designs for the DRC and radar beamforming vectors. Then, an alternating optimization approach is applied, decomposing the original problem into two sub-problems. For the DRC optimization sub-problem, a design approach based on semidefinite relaxation (SDR) and Gaussian randomization is proposed. For the radar beamforming sub-problem, the closed-form expressions for the optimal radar beamforming vectors are obtained by the generalized Rayleigh quotient and the Karush-Kuhn-Tucker (KKT) conditions. Finally, it is demonstrated that within a communication-centric scenario, the RIS needs to maintain orthogonality between the interference channel and the radar-target channel in order to achieve the upper bound of the radar SINR. Based on this conclusion, a low-complexity approach is proposed to optimize the radar SINR in a communication-centric scenario.