A Rate Splitting Approach For Multi-user MISO Communication Systems Using Genetic Algorithm

Recent research efforts in the field of 6G communication systems shows that the Rate splitting (RS) technique is a promising strategy for non-orthogonal transmission and interference management. RS consists of dividing a user message into two parts, i.e., common and private messages, such that each of those parts can be decoded at one or multiple receivers. In this paper, we introduce a competent RS approach based on the Genetic Algorithm (GA) to maximize the sum rate of Multi-Input Single-Output Broadcast Channel (MISO BC) communication systems. Existing solutions for the sum rate maximization problem are often based on approximation algorithms that lead to poor solutions. We consider two scenarios to divide the power between the common and private messages. The objective of the first scenario is to evaluate the performance of the GA-based approach compared to the existing RS benchmarks. In this case, the power of private massages is assumed to be equal, and we deal with a single-variable optimization problem. In order to generalize the proposed method, the second scenario considers a case in which the private messages have different power coefficients. The simulation results indicate that the proposed algorithm achieves a better Ergodic Sum Rate (ESR) than the existing approximation algorithms.