Enhancing energy efficiency in downlink NOMA systems using FCM clustering and PRSO power optimization

Comparing Non-Orthogonal Multiple Access (NOMA) systems for the 5 G and beyond to traditional Orthogonal Multiple Access (OMA) schemes, the former could offer multiple connections that significantly improve overall energy efficiency and the bandwidth of the spectrum. As a consequence, researchers have started their studies on the NOMA system. In addition, the resource allocation (RA) problem in the NOMA system has received substantial research from an energy-efficient perspective but needs adequate evaluation. Furthermore, energy consumption is still a major issue that requires efficient resource allocation. Particle Rat Swarm Optimisation (PRSO) based RA is proposed for a downlink NOMA system. A system model of the downlink NOMA system consists of a gNodeB and the users are allocated to sub-channels. Users are grouped using Fuzzy C-means (FCM) clustering and a PRSO which is a combination of Particle Swarm Optimisation (PSO) and Rat Swarm Optimizer (RSO) is used for the allocation of power coefficients to the users in the group. The proposed FCM and PRSO-based NOMA system attains a sum rate of 732.4 Mbps and an energy efficiency of 11.3 Mbits/joule and thus, PRSO outperforms other algorithms such as PSO, RSO and Greedy for power allocation in the NOMA system.