OTFS-NOMA System Resource Allocation Scheme Based on SADEBOA Algorithm

This paper investigates the power allocation problem in the downlink of an Orthogonal Time Frequency Space-Non-Orthogonal Multiple Access (OTFS-NOMA) system under high-speed conditions, with the objective of maximizing system capacity and rate. A novel Butterfly Optimization Algorithm based on Differential Evolution (DE) and Simulated Annealing (SA), referred to as SADEBOA, is proposed. In the global search phase, SADEBOA incorporates DE to enhance the global search capability of the Butterfly Optimization Algorithm (BOA), mutating the optimal individual and guiding the population towards the optimal individual's location. In the local search phase, SADEBOA combines SA to accept a local optimal solution as a new solution with a certain probability, introducing a certain bias that is more likely to approach the optimal solution. Comparative simulations are conducted with BOA, MSBOA, DA, and RSO algorithms, as well as traditional power allocation schemes. The simulation results demonstrate that, compared to the baseline BOA algorithm, MSBOA, DA, and RSO algorithms, SADEBOA achieves performance improvements of approximately 15%, 5.3%, 21.2%, and 28.4%, in enhancing OTFS-NOMA system capacity.