Fairness driven TDMA with joint phase and beamforming optimization for ND-IRS assisted MU-MISO communication systems

Abstract

This paper investigates the optimization of rate performance in downlink intelligent reflecting surface (IRS)-assisted multi-user multiple-input single-output (MU-MISO) systems with a focus on fairness. The IRS operates in either diagonal (D) or non-diagonal (ND) configurations, with time allocation designed to maximize the minimum signal-to-noise ratio (SNR), effectively optimizing data rates for all users. The D-IRS configuration is commonly used due to its simplicity in implementation and control link load. However, ND-IRS allows signal reflection from different elements with adjustable phase shifts, enhancing system performance. The study examines how different IRS configurations and time allocation strategies influence data rates in a fairness-driven MU-MISO, time division multiple access (TDMA) transmission system. Using alternating optimization (AO) for the joint phase of the IRS and beamforming at the base station (BS), combined with fairness-based time allocation, the research explores the impact of transmit power, IRS elements, and user count on performance. Results show that ND-IRS with fairness-based time allocation consistently delivers the highest data rates across various scenarios. ND-IRS outperforms D-IRS, and random phase IRS setups, especially as transmit power and the number of IRS elements increase. Our results demonstrate that ND-IRS with fairness-based time allocation consistently achieves the highest data rates across various scenarios. For example, ND-IRS with fairness based time allocation performs better as compared to the equal time allocation scheme. Further, ND-IRS outperforms D-IRS with optimized phase as well as random phase IRS setups. Computer simulations confirm these findings.