Development of NOMA‐CR System Using Machine Learning Algorithm

ABSTRACT New mobile users and the growing need for real‐time applications cause a sharp rise in network traffic and complexity. The main issues being faced by 5G networks are spectrum restrictions, heterogeneity, and network capacity. These requirements of network traffic are not met by conventional multiple access methods like orthogonal multiple access (OMA). These needs could be met by the non‐orthogonal multiple access (NOMA) approach in conjunction with the cognitive radio (CR) network which can serve several users at a time. Therefore, in order to meet the demands of the 5G network, the main focus is on developing a unique technique for the NOMA‐CR system employing hierarchical clustering. Here, a unique algorithm that combines channel codification, fake detection alerts, and linear optimization techniques is developed using NOMA CR user pairing. Then, based on sum rate, spectrum efficiency, power efficiency, and throughput, the suggested system is compared with OMA, TDMA, k‐NN with equal power allocation, water filling algorithm, and RL algorithm. In comparison to these systems, the novel technique yields a sum rate of 5.8 bps/Hz, throughput of 58 Mbps, and spectrum efficiency of 95%. The results obtained after implementation of the newly developed system demonstrate that the suggested system provides the best overall performance. As a result, this novel solution will be the best option for overcoming the difficulties experienced by 5G networks.