Multi-Objective Optimization of Resource Allocation in 5G Networks Using a Reinforcement Learning-Based Genetic Algorithm
Abstract
This paper presents a hybrid Reinforcement Learning-Genetic Algorithm (RL-GA) for multi-objective resource allocation in 5G networks. Traditional optimization methods struggle with simultaneously optimizing throughput, latency, and energy consumption. The proposed method employs an RL agent that dynamically adjusts Genetic Algorithm (GA) parameters (crossover and mutation rates) based on population state, inspired by dopamine-mediated reward prediction mechanisms in cognitive science. Experimental results demonstrate statistically significant improvements over standard GA, adaptive GA, and PSO: 21.9% throughput increase, 35.9% latency reduction, and 16.7% energy efficiency improvement (p<0.001). The approach achieves faster convergence and superior Pareto front quality, addressing critical challenges in dynamic 5G environments.
Keywords:
Multi-objective optimization, Genetic algorithm, Reinforcement learning, 5G networks, Resource allocation, 5G-advancedReferences
- [1] Kamal, M. A., Raza, H. W., Alam, M. M., Su’ud, M. M., & Sajak, A. B. (2021). Resource allocation schemes for 5G network: a systematic review. Sensors, 21(19), 6588. https://doi.org/10.3390/s21196588
- [2] Moreira, C. L., Kamienski, C. A., & Bianchi, R. A. C. (2024). 5G and edge: A reinforcement learning approach for virtual network embedding with cost optimization and improved acceptance rate. Computer networks, 247, 110434. https://doi.org/10.1016/j.comnet.2024.110434
- [3] Mnih, V., Kavukcuoglu, K., Silver, D., Rusu, A. A., Veness, J., Bellemare, M. G., … & Hassabis, D. (2015). Human-level control through deep reinforcement learning. Nature, 518(7540), 529–533. https://doi.org/10.1038/nature14236
- [4] Kawachi, M., & Ando, N. (1992). Genetic algorithms in search, optimization & machine learning. Artificial intelligence, 7(1), 168-168. (In Chinese). https://doi.org/10.11517/jjsai.7.1_168
- [5] Deb, K., Pratap, A., Agarwal, S., & Meyarivan, T. (2002). A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE transactions on evolutionary computation, 6(2), 182–197. https://doi.org/10.1109/4235.996017
- [6] Jin, Y., & Chen, Z. (2023). A fast resource allocation algorithm based on reinforcement learning in edge computing networks considering user cooperation. Electronics, 12(6), 1459. https://doi.org/10.3390/electronics12061459
- [7] Watkins, C. J. C. H., & Dayan, P. (1992). Q-learning. Machine learning, 8(3), 279–292. https://doi.org/10.1007/BF00992698
- [8] Amaya, F., & Bahia, K. (2024). The state of 5G 2024. https://www.gsmaintelligence.com/research/the-state-of-5g-in-2024
- [9] Guo, J., Yao, H., Mai, T., Ouyang, T., & Wang, F. (2024). Reinforcement learning-based genetic algorithm for differentiated traffic scheduling in industrial TSN-5G networks. https://doi.org/10.1109/IWCMC61514.2024.10592554.
- [10] Nguyen, V. (2025). The state of 5G: Growth, challenges, and opportunities in 2025. https://www.5gamericas.org/the-state-of-5g-growth-challenges-and-opportunities-in-2025/
Issue 1