PERFORMANCE ANALYSIS OF INTELLIGENT AND SURFACE-ENABLED WIRELESS NETWORKS

Abstract

One of the technologies that will allow 6G to function is reconfigurable intelligent surface (RIS), which was the focus of this master's thesis. The article provided a thorough analysis of RIS, including its design, operation, and features. The RIS-aided system's performance was assessed using two system models, namely cascaded RISs-aided system and RIS partitioning system. The current state of the art in research was outlined and research gaps were identified for this system models. For both models, closed-form probability distribution functions (PDFs) were derived and validated by using Monte-Carlo simulations. For the cascaded RIS-aided systems, analytical approximation errors were validated by the Kolmogorov-Smirnov goodness-of-fit-test. Based on the results of the system assessment, it was found that the outage probability (OP) metric drops as the number of RISs in series grows. This is because more RISs mean more path-loss in the system, which means more power is needed. An improvement in the OP metric is seen when other system parameters, such as diversity path parameter (m), average received power parameter (Ω), and number of reflecting elements per RIS (N), are increased. A longer route for the signal to go to the user, and thus greater power received at the destination, is represented by an increase in the m parameter. A higher value for the received average power, as measured by the average power parameter, indicates a better operation metric. An interesting finding is that the received power doubles by 8 dBm for every doubling increase of N, which is the number of reflecting components. Analytical and simulational data are provided for each of these findings. All of the results had their approximation errors computed, and the KS goodness-of-fit test confirmed them. The RIS partitioning network is the following system model. Analytical and experimental studies were conducted on this system. We obtained closed-form PDF for the analytical findings. The Monte-Carlo simulation confirmed the closed-form equation. The results of the experiments show that RIS partitioning is effective and offers several benefits. There is a 6 dBm improvement in received power for near-field communication when both non-orthogonal-multiple-access (NOMA) users are in the same location. Since each element of the RIS is dedicated to serving just one user, the received power remains same even when NOMA users are located in various locations. When comparing far-field communication employing all RIS components to half-RIS, the difference in received power was just -5 dBm.