BD-RIS-Aided Multi-Target ISAC for Perception-for-Communication

Abstract

Indoor integrated sensing and communications systems suffer from severe link blockage and quadruple path-loss in monostatic reconfigurable intelligent surface (RIS)-assisted sensing setups, while conventional designs typically regard sensing as a communication burden. To address these challenges, this letter proposes a beyond-diagonal reconfigurable intelligent surface (BD-RIS) aided perception-for-communication (P4C) framework, where target-scattered signals are exploited as additional communication multipath components through joint beamforming and BD-RIS optimization. A communication-benefit-aware adaptive sensing allocation strategy is developed to allocate sensing resources according to the communication contribution of different targets. Based on the monotonic relationship between the posterior Cramér–Rao bound for position estimation and effective sensing power, the sensing constraints are transformed into tractable power allocation requirements. An efficient joint beamforming and BD-RIS optimization algorithm is then developed using alternating optimization, weighted minimum mean-square error-based successive convex approximation, and Riemannian manifold optimization. Simulation results demonstrate that the proposed BD-RIS-aided P4C framework achieves a higher communication sum-rate than conventional diagonal RIS architectures while satisfying the prescribed sensing constraints.