Resource Allocation Mechanism for NOMA-enabled Underwater Visible Light Communication Systems

We jointly optimize channel allocation and power control to maximize the sum proportional bit rate to balance spectrum efficiency and user fairness in a non-orthogonal multiple access (NOMA)-enabled underwater visible light communication system. To address the mixed-integer non-convex optimization problem, we decompose it into a power control subproblem and a channel allocation subproblem. We then propose two low-complexity algorithms to solve these subproblems by leveraging convex optimization and fractional programming theories. These two algorithms are executed alternately until convergence, yielding a locally optimal solution to the original problem. Simulation results demonstrate that the proposed mechanism improves performance by 3.9% and 3.0% compared to baselines based on swap-matching theory and the K-means clustering algorithm. It substantially reduces execution time by 99.2%, with only a minor performance loss of 1.6% compared to the genetic algorithm-based baseline.