Adaptive Resource Allocation Techniques for Nonorthogonal Multiple Access (NOMA) Systems

NOMA is a promising multiple-access technology that enables multiple users to access the same frequency resources simultaneously, achieving good spectral efficiency. However, to realize the full potential of NOMA, suitable resource allocation strategies for resource assignment to diverse users are required. This paper reviews the currently available resource allocation techniques for NOMA systems and proposes a new adaptive resource allocation method. In this approach, machine learning algorithms and dynamic resource allocation are used based on the current channel conditions, requirements for power, and additional metrics across the network. Thus, this method provides better spectrum utilization and efficient user fairness and minimizes serialized users. The resource allocation techniques of NOMA and the challenges and their solutions in real systems are deduced. Consequently, based on this analysis, we derive an optimization-based joint user scheduling and resource allocation scheme and, theoretically and via simulations, show that its performance in system throughput, fairness and outage probability are superior to that of the existing solution. Results show the robustness of adaptive techniques in addressing resource allocation optimization of the NOMA systems and highlight their promise to mitigate the struggles of future prospective wireless networks.