Multi‐objective resource allocation in multiuser orthogonal frequency division multiplexing system

Abstract

This study presents a new technique for resource allocation in multiuser orthogonal frequency division multiplexing systems. The goal is to maximise the minimum data rate available to any user while minimising the total transmitted power. The strength Pareto evolutionary algorithm (SPEA-2) is used to achieve this goal. The SPEA-2 algorithm solves the contradicting multiple objectives by evaluating individual's fitness value based on the number of external non-dominated individuals that dominate it and then searching the solution space to minimise this fitness value. Most of the existing multi-objective solutions, for the problem under consideration, have used binary coded chromosomes which restricted the number of users to be in power of two only. This limitation is overcome in the proposed scheme by using an integer coded chromosome. The population density information is also incorporated into the fitness function to refine the search. Simulation results indicate that the proposed algorithm achieves higher data rates as compared with previous algorithms. Furthermore, the proposed scheme allocates both subcarriers and bits jointly, without being computationally expensive. The faster convergence of the algorithm to near-optimal value, as compared with previous algorithms is indicative of its reduced complexity, which is attributed to the modification in the power objective.