Optimal Placement of Reconfigurable Intelligent Surfaces with Random Obstacle Distribution

Reconfigurable intelligent surfaces (RISs) have been a promising technology to maintain connection performance for millimeter wave (mmWave) communication in non-line-of-sight (NLoS) case by providing an indirect link between access point and user. In this paper, we explore the advantage of multi-RIS deployment to improve connection probability in a scenario with randomly distributed obstacles by solving a modified thinnest covering problem. Optimal RIS deployment in 3D scenario up to six RISs and selection of RIS number based on room size are investigated analytically. A heuristic optimization method of RIS size and orientation is also proposed to guarantee adequate received signal strength. The proposed deployment strategy is validated by simulation that connection probability is significantly improved with only very few RISs.