A Deep Q-Learning Bisection Approach for Power Allocation in Downlink NOMA Systems

• Published in: IEEE communications letters Vol. 26; no. 2; pp. 316 - 320
• Main Authors: Youssef, Marie-Josepha, Nour, Charbel Abdel, Lagrange, Xavier, Douillard, Catherine
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.02.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE); Institute of Electrical and Electronics Engineers
• Subjects: Artificial Intelligence; Complexity theory; Computer Science; Deep learning; deep reinforcement learning; Downlink; Downlinking; Genetic algorithms; Information Theory; Machine learning; Networking and Internet Architecture; NOMA; Non-orthogonal multiple access; Nonorthogonal multiple access; Optimization; Resource management; Silicon carbide; successive interference cancellation stability; Upper bound; weighted sum-rate maximization; Deep reinforcement learning DRL; Non-orthogonal multiple access NOMA; successive interference cancellation stability; weighted sum-rate maximization
• ISSN: 1089-7798; 1558-2558
• DOI: 10.1109/LCOMM.2021.3130102

In this work, we study the weighted sum-rate maximization problem for a downlink non-orthogonal multiple access (NOMA) system. With power and data-rate constraints, this problem is generally non-convex. Therefore, a novel solution based on the deep reinforcement learning (DRL) framework is proposed for the power allocation problem. While previous work based on DRL restrict the solution to a limited set of possible power levels, the proposed DRL framework is specifically designed to find a solution with a much larger granularity, emulating a continuous power allocation. Simulation results show that the proposed power allocation method outperforms two baseline algorithms. Moreover, it achieves almost 85% of the weighted sum-rate obtained by a far more complex genetic algorithm that approaches exhaustive search in terms of performance.