Energy Efficiency and Fairness in Large Scale Systems Using RSMA

Abstract

The performance of Space Division Multiple Access (SDMA) in Multiuser Multiple Input Single Output (MU-MISO) systems degrades significantly under imperfect Channel State Information at the Transmitter (CSIT). To address this, Rate-Splitting Multiple Access (RSMA) has been shown to outperform SDMA in scenarios where CSIT is imperfect, particularly in underloaded scenarios, where the number of potential users is lower than the number of antennas. This paper investigates the use of RSMA in a large-scale, overloaded system, where the number of users exceeds the number of antennas at the base station. We propose a novel approximation of the RSMA sum rate under limited feedback conditions and develop an optimal power allocation strategy that dynamically switches between RSMA and SDMA to maximize energy efficiency and system performance. Additionally, a robust Minimum Mean Square Error (MMSE) based precoding method is introduced to mitigate the effects of imperfect CSIT in private streams of RSMA. Numerical simulations validate our analytical derivations and show that RSMA offers superior performance over SDMA in large user regimes with low feedback loads, providing significant performance gains in realistic network conditions. These findings offer new insights into the design of energy-efficient and scalable downlink communication systems for future wireless networks.