Performance of RIS Assisted Networks in Noise Limited and Interference Limited Scenarios

This paper investigates the performance of Reconfigurable Intelligent Surface (RIS) assisted networks. We consider noise-limited and interference-limited scenarios. We introduce the concept of capture area; the RIS elements are distributed in the capture area of a transmit-receive pair to improve the performance of communication. We show the existence of a knee region inside the capture area; the distribution of RIS within the knee area will be more beneficial in terms of the capacity returns. This paper also characterizes various inter-node distances in RIS-assisted networks, which are essential to analyze the performance. Our results also suggest that even though RIS can improve the link-level performance in noise-limited scenarios, the accumulated reflected signal may hurt the performance in interference-limited systems.