A Federated Learning Approach for Predicting Resource Allocation in Multi-Access Edge Computing
DOI:
https://doi.org/10.63282/3050-9246.IJETCSIT-V6I4P106Keywords:
Federated Learning, Multi-Access Edge Computing, Resource Allocation, 5G/6G Networks, Non-IID Data, Distributed Machine Learning, Energy Efficiency, Model AggregationAbstract
In modern networks, where stronger ultra-low latency and data throughput are needed, Multi-Access Edge Computing (MEC) becomes a necessary architecture for 5G/6G networks that support real-time applications. Nevertheless, a dynamic edge ecosystem, diverse device properties, and privacy preservation needs interfere with MEC resource management. This paper proposes a new Federated Learning (FL) framework to predict resource allocation in MEC that removes such barriers by enabling decentralized model training to be performed directly at the network edge. In contradiction to conventional centralized strategies, our approach significantly reduces communication costs by up to 90% while providing competitive performance due to the efficient use of non-IID data at edge locations. Feeding lightweight CNNs and reducing the whole energy demand is achieved by the balanced computational requirements in the design aggregation through FedOpt aggregation. Based on the results of outcome analysis on MNIST and Fashion-MNIST, we observe accelerated convergence, increased energy savings and performance scalability, where energy consumption per training round is 29% lower than in centralized systems. This approach shows impressive results in processing non-IID data due to reliable performance on different edge devices. Such discoveries show that FL has a high potential to transform MEC resource allocation and thus contribute to more adaptive, protected, and efficient edge computing architecture
Downloads
References
[1] Sarah, A., Nencioni, G., & Khan, M. M. I. (2023). Resource allocation in multi-access edge computing for 5G-and-beyond networks. Computer Networks, 227, 109720.
[2] Wang, Y., Chen, X., Chen, Y., & Du, S. (2021, October). Resource allocation algorithm for MEC based on Deep Reinforcement Learning. In 2021 IEEE International Performance, Computing, and Communications Conference (IPCCC) (pp. 1-6). IEEE.
[3] Choudhury, A., Ghose, M., & Islam, A. (2024). Machine learning-based computation offloading in multi-access edge computing: A survey. Journal of Systems Architecture, 148, 103090.
[4] Mahimalur, R. K. (2025). Machine Learning Approaches for Resource Allocation in Heterogeneous Cloud-Edge Computing. International Journal of Scientific Research in Computer Science, Engineering and Information Technology, 11(2), 10-32628.
[5] Vahabi, M., & Fotouhi, H. (2025). Federated learning at the edge in Industrial Internet of Things: A review. Sustainable Computing: Informatics and Systems, 101087.
[6] Li, H., Pan, Y., Zhu, H., Gong, P., & Wang, J. (2023). Resource Management for MEC Assisted Multi-Layer Federated Learning Framework. IEEE Transactions on Wireless Communications, 23(6), 5680-5693.
[7] Quan, P. K., Kundroo, M., & Kim, T. (2023). Experimental evaluation and analysis of federated learning in edge computing environments. IEEE Access, 11, 33628-33639.
[8] Kotecha, P., Dhoka, T., Bhatia, J., Kumhar, M., Gupta, R., Tanwar, S., & Jadav, N. K. (2024). Performance Evaluation of Federated Learning in Edge Computing Environment. Procedia Computer Science, 235, 2955-2964.
[9] Woisetschläger, H., Erben, A., Mayer, R., Wang, S., & Jacobsen, H. A. (2024, December). FLEdge: Benchmarking Federated Learning Applications in Edge Computing Systems. In Proceedings of the 25th International Middleware Conference (pp. 88-102).
[10] Samafou, F., Adoum, B. A., Fidel, F. M., Ari, A. A. A., Moungache, A., Armi, N., & Gueroui, A. M. (2024). A new approach to joint resource management in MEC-IoT based federated meta-learning. Bulletin of Electrical Engineering and Informatics, 13(5), 3196-3217.
[11] Mughal, F. R., He, J., Das, B., Dharejo, F. A., Zhu, N., Khan, S. B., & Alzahrani, S. (2024). Adaptive federated learning for resource-constrained IoT devices through edge intelligence and multi-edge clustering. Scientific Reports, 14(1), 28746.
[12] Schwanck, F. M., Leipnitz, M. T., Carbonera, J. L., & Wickboldt, J. A. (2025). A Framework for testing Federated Learning algorithms using an edge-like environment. Future Generation Computer Systems, 166, 107626.
[13] Federated Learning for Privacy-Preserving Edge Computing, dialzara, 2024. online. https://dialzara.com/blog/federated-learning-for-privacy-preserving-edge-computing/
[14] Zarandi, S. (2021). Resource Allocation in Multi-access Edge Computing (MEC) Systems: Optimization and Machine Learning Algorithms.
[15] Li, J., Gao, H., Lv, T., & Lu, Y. (2018, April). Deep reinforcement learning-based computation offloading and resource allocation for MEC. In 2018 IEEE wireless communications and networking conference (WCNC) (pp. 1-6). IEEE.
[16] Jing, Y., Wang, J., Jiang, C., & Zhan, Y. (2022). Satellite MEC with federated learning: Architectures, technologies and challenges. IEEE Network, 36(5), 106-112.
[17] He, Y., Yang, M., He, Z., & Guizani, M. (2023). Computation offloading and resource allocation based on DT-MEC-assisted federated learning framework. IEEE Transactions on Cognitive Communications and Networking, 9(6), 1707-1720.
[18] Hao, X., Yeoh, P. L., She, C., Vucetic, B., & Li, Y. (2023). Secure deep reinforcement learning for dynamic resource allocation in wireless MEC networks. IEEE Transactions on Communications, 72(3), 1414-1427.
[19] Feng, C., Zhao, Z., Wang, Y., Quek, T. Q., & Peng, M. (2021). On the design of federated learning in the mobile edge computing systems. IEEE Transactions on Communications, 69(9), 5902-5916.
[20] Wu, W., He, L., Lin, W., & Mao, R. (2020). Accelerating federated learning over reliability-agnostic clients in mobile edge computing systems. IEEE Transactions on Parallel and Distributed Systems, 32(7), 1539-1551.
