Deep Learning-Based Optimization of Power Allocation Schemes in Noma-Enabled Wireless Networks

Abstract

With the rapid development of 5G and the transition toward 6G, wireless networks face increasing demands for high-speed data, energy efficiency, and massive user connectivity. Non-Orthogonal Multiple Access (NOMA) has emerged as a promising technique to meet these demands by allowing multiple users to share the same spectrum resources through power-domain multiplexing. However, one of the main challenges in NOMA systems is the efficient allocation of power among users with different channel conditions. Traditional power allocation techniques such as convex optimization, Fractional Transmit Power Allocation (FTPA), and heuristic algorithms like Particle Swarm Optimization (PSO) often require high computational effort and fail to adapt in real-time to changing network conditions. This paper proposes a deep learning-based framework for optimizing power allocation in downlink NOMA-enabled wireless networks. A deep neural network (DNN) model is trained using simulated channel data to predict optimal power distribution among users while satisfying Quality of Service (QoS) constraints and maximizing energy efficiency. The model significantly reduces the time required for decision-making and adapts efficiently to dynamic environments. Simulation results show that the proposed DNN-based approach outperforms traditional methods in terms of energy efficiency, user fairness, and outage performance. This research demonstrates that integrating artificial intelligence techniques like deep learning into NOMA systems can greatly enhance the performance of future wireless communication networks.