System Level Simulation Method and Performance of RIS-Assisted Wireless Networks

Reconfigurable Intelligent Surface (RIS) is a promising technique in the next generation of wireless communication systems. Many researches have discussed the signal quality promotion introduced by RIS via link-level simulations or field tests, but few are done in the system level. In this paper we conduct a System-Level Simulation (SLS) of the RIS-assisted network based on the Geometry-based Stochastic Channel Model (GSCM), and use the results to answer two questions: 1) How much coverage gain is brought by RIS, and how many users benefit from RIS? 2) How to avoid prohibitively high complexity in the RIS SLS due to the enormous increase of RIS-related links and equivalent channels? For the first question, we derive an closed-form Reference Signal Received Power (RSRP) formula for RIS equivalent channel considering the impact of spatial directional beamforming. From the formula, we further analyze the typical propagation conditions that a user may benefit from RIS for coverage enhancement. For the second question, the RSRP distributions are derived for both the desired signal from associated BS and RIS and the adverse signal from interfering BS or RIS, from which we prove that the cross-sector RIS interference are weak enough to be neglected in the SLS, reducing the simulation complexity significantly.