Data-driven Based Technique for Joint Power Allocation and User Pairing in NOMA System

Abstract

As one of the proposed multiple access techniques of the fifth-generation and beyond systems, NOMA, which stands for non-orthogonal multiple access. NOMA is a power domain access technique, which ensures better spectrum efficiency and system capacity. However, the representation of NOMA is challenging because of inter-user interference and the complexity of successive interference cancellation (SIC) in the receiver. We suggest a low-complexity power allocation (PA) and user pairing (UP) scheme using a model-free technique that depends on the real-time input and output calculations, data-driven (DD), where the near-optimal solution has been found. So, data-driven can be utilized as an optimization tool for finding a near-optimum solution of the original NP-hard problem of joint PA and UP. Analytical and simulation results referred that the proposed algorithm can enhance spectrum efficiency compared with the existing algorithms.