Ray-tracing Based Investigations on the Deployment of RISs in Indoor Scenarios

Sammanfattning: Reconfigurable intelligence surface (RIS) is a promising candidate technology for future 6G wireless communication systems. In existing communication systems, the network operators are unable to control the propagation environment, which causes significant limitations on communication performance. RIS aims to create favorable propagation conditions via programmable phase shifts. Employing RIS can improve power/spectrum efficiency, which improves the quality of communication and reduces costs. Understanding the trade-offs between the number of base stations (BSs) and the number of RISs is highly desired by both industry and academia. To this end, a ray-tracing-based deterministic indoor channel model is developed. Based on the ray tracing simulator, this thesis aims at providing trade-offs between the number of BSs and the number of RISs regarding the received power of user equipment (UE). Ray tracing simulations are made in the Wireless InSite™ software, and data processing in MATLAB™. To investigate the trade-offs, we evaluate the communication performance with different numbers of activated RISs and BSs. Here, our approach is to investigate a large number of tentative deployed BS/RIS positions and constrain the number of activated BS/RIS and the minimum acceptable power threshold at UE. Given the constraints, we optimize over the BS/RIS placements. In total, four optimization problems are solved in this thesis. We will present our mathematical modeling of optimization problems and how we solve them using MATLAB™. After the completion of the thesis, we conclude that deploying a small number of RISs can reduce the number of BSs in indoor scenarios without any performance loss regarding the received power at UE. Moreover, RIS is more applicable in environments with abundant shadowing objects. Allowing a non-zero outage probability, i.e., dropping some UE positions with bad communication conditions, is verified to improve power efficiency greatly. Lastly, we also show the necessity to optimize the phase shifts of RIS to enhance the RIS performance in a wideband system