Deep Learning Technique for Power Domain Non-Orthogonal Multiple Access Using Optimised LSTM in Cooperative Networks

Abstract

: Non-orthogonal Multiple Access (NOMA) is the technique proposed for multiple accesses in the fifth-generation (5G) cellular network. In NOMA, different users are allocated different power levels and are served using the same time/frequency Resource Blocks (RBs). The main challenges in existing NOMA systems are the limited channel feedback and the difficulty of merging them with advanced adaptive coding and modulation schemes. The 5G system in NOMA aims to access low latency, efficiency in superior spectra, and balanced user fairness. NOMA allows multiple users with different power levels to share resources in radio frequency time. The existing Orthogonal Multiple Access (OMA) system produces high latency, high computational complexity, and throughput complexity in modifying wireless channels. To overcome these issues, this paper proposed optimising deep learning-based power domain NOMA of Long Short-Term Memory (LSTM) with particles Swarm optimisation (PSO) technique. This proposed work (LSTM-PSO) is deployed with a Cooperative network model. The advantage of LSTM-PSO in Cooperative Non-orthogonal Multiple Access (CNOMA) is that it provides high performance, better utilisation of downlink, efficiency in sharing of resources, enhancing the activity of users, capacity of the base station and improving quality of service, estimation of channel condition. LSTM-PSO got a higher accuracy rate of 92.05%, LSTM got 86.45%, PSO got 88.13%, and the accuracy rate of ANN and DNN was 83.76% and 84.70%.