RIS-Aided Communication Network Deployment Strategy Based on BP Model: From Single-Hop to Multi-Hop

Reconfigurable Intelligent Surfaces (RISs) represent a promising technological advancement capable of establishing a smart radio environment through the utilization of reconfigurable passive elements. This paper examines an RIS-assisted multiuser multiple-input single-output (MU-MISO) downlink communication system and introduces a blockage probability (BP) model aimed at optimizing the deployment strategy of RISs. Initially, we develop a block network model and subsequently align it with the actual wireless communication environment by adjusting the probabilities of random events and the blockage threshold. Upon determining the optimal deployment positions for the RISs, we formulate an optimization problem designed to maximize the system’s sum-rate by jointly designing the active beamforming at the base station (BS) and the passive beamforming of the RIS elements. The resulting non-convex problem is addressed using an alternating optimization (AO) method, which incorporates the weighted minimum mean square error (WMMSE) algorithm and the Riemannian conjugate gradient (RCG) algorithm. In addition, we also analyze the system robustness with bounded CSI error model. Simulation results indicate that, in comparison to other deployment strategies, our proposed model facilitates greater control freedom and enhances the system’s sum-rate.