Federated Learning Framework for Edge Devices with Reducing Communication Network Costs and Enhancing Performance

Abstract

In this study, the effectiveness of the proposed federation learning method is evaluated through some experiments conducted on the MNIST dataset. The proposed method aims to significantly reduce the communication costs while increasing the model accuracy thanks to a hierarchical approach and adaptive weighting mechanism. The experimental results show that the proposed method significantly shortens the training time and reduces the communication cost. Especially, in scenarios where edge devices with different computational power are present in the network environments, the presented method showed better performance. It was also observed that the proposed method increased the model accuracy and provided better generalization ability to the models. The findings obtained in this study show that the proposed federation learning model is an effective and efficient solution for model training in edge computing systems. This method is considered to have great potential especially for applications where communication bandwidth is limited and privacy is important. In future studies, it is planned to evaluate the performance of this method on different datasets and more complex model architectures.

Keywords

• Federated learning
• edge computing
• communication networks
• convolutional neural networks

İletişim Ağı Maliyetlerini Azaltan ve Performansı Geliştiren Kenar Aygıtları için Federasyonlu Öğrenme Çerçevesi

Abstract

Bu çalışmada, önerilen federasyon öğrenme yönteminin etkinliği MNIST veri kümesi üzerinde yapılan bazı deneylerle değerlendirilmiştir. Önerilen yöntem, hiyerarşik bir yaklaşım ve uyarlanabilir ağırlıklandırma mekanizması sayesinde model doğruluğunu artırırken iletişim maliyetlerini önemli ölçüde azaltmayı amaçlamaktadır. Deneysel sonuçlar, önerilen yöntemin eğitim süresini önemli ölçüde kısalttığını ve iletişim maliyetini azalttığını göstermektedir. Özellikle, ağ ortamlarında farklı hesaplama gücüne sahip uç cihazların bulunduğu senaryolarda, sunulan yöntem daha iyi performans göstermiştir. Ayrıca, önerilen yöntemin model doğruluğunu artırdığı ve modellere daha iyi genelleme yeteneği sağladığı görülmüştür. Bu çalışmada elde edilen bulgular, önerilen federasyon öğrenme modelinin uç bilişim sistemlerinde model eğitimi için etkili ve verimli bir çözüm olduğunu göstermektedir. Bu yöntemin, özellikle iletişim bant genişliğinin sınırlı olduğu ve gizliliğin önemli olduğu uygulamalar için büyük bir potansiyele sahip olduğu düşünülmektedir. Gelecekteki çalışmalarda, bu yöntemin performansının farklı veri kümeleri ve daha karmaşık model mimarileri üzerinde değerlendirilmesi planlanmaktadır.

Keywords

• Federasyonlu öğrenme
• uç bilişim
• iletişim ağları
• evrişimli sinir ağları

References

1. Abdellatif, A.A., Mhaisen, N., Mohamed, A., Erbad, A., Guizani, M., Dawy, Z., Nasreddine, W. (2022). Communication-efficient hierarchical federated learning for IoT heterogeneous systems with imbalanced data, Future Generation Computer Systems, 128, 406-419.
2. Akhtarshenas, A., Vahedifar, M. A., Ayoobi, N., Maham, B., Alizadeh, T., Ebrahimi, S., López-Pérez, D. (2024). Federated learning: A cutting-edge survey of the latest advancements and applications, Computer Communications, 228, 107964.
3. Al-Shedivat, M., Gillenwater, J., Xing, E., Rostamizadeh, A. (2020). Federated learning viaposterior averaging: A new perspective and practical algorithms, arXivpreprintarXiv:2010.05273.
4. Beltrán, E.T.M., Pérez, M.Q., Sánchez, P.M.S., Bernal, S.L., Bovet, G., Pérez, M.G., Pérez, G.M., Celdrán, A.H. (2023). Decentralized federated learning: Fundamentals, state of the art, frameworks, trends, and challenges, IEEE Commun. Surv. Tutor.
5. Du, H., Chen, Y., Feng, X., Xiang, Q., Liu, H. (2023). An efficient federated learning framework for multi-channeled mobile edge network with layered gradient compression, Computer Networks, 221, 109517.
6. Herabad, M. G. (2023). Communication-efficient semi-synchronous hierarchical federated learning with balanced training in heterogeneous IoT edge environments, Internet of Things, 21, 100642.
7. Grafberger, A., Chadha, M., Jindal, A., Gu, J., Gerndt, M. (2021). FedLess: Secure andscalable federated learning using serverless computing, in: 2021 IEEE Interna-tional Conference on Big Data, Big Data, 164–173.
8. Guo, J., Zhou, H., Liu, X., Zhao, L., Leung, V. (2024). A stackelberg game-based wirelesspowered federated learning, in: 2024 27th International Conference on ComputerSupported Cooperative Work in Design, CSCWD, 2024, pp. 278–283.

9. Khatua, S., Mukherjee, A., De, D. (2024). FedGen: Federated learning-based green edge computing for optimal route selection using genetic algorithm in Internet of Vehicular Things, Vehicular Communications, 49, 100812.
10. Li, B., Schmidt, M.N., Alstrø, T.S., Stich, S.U. (2023). On the effectiveness of partial variance reduction in federated learning with heterogeneous data, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 3964–3973.
11. Li, H., Yang, Y., Liu, Y., Pei, W. (2024). Federated dueling DQN based microgrid energy management strategy in edge-cloud computing environment, Sustainable Energy, Grids and Networks, 38, 101329.
12. Liu, Y., Dong, Y., Wang, H., Jiang, H., Xu, Q. (2022). Distributed fog computing and federated-learning-enabled secure aggregation for IoT devices, IEEE Internet Things J. 9 (21), 21025–21037.
13. Rahmani, A. M., Alsubai, S., Alanazi, A., Alqahtani, A., Zaidi, M.M., Hosseinzadeh, M. (2024). The role of mobile edge computing in advancing federated learning algorithms and techniques: A systematic review of applications, challenges, and future directions, Computers and Electrical Engineering, 120, Part C, 109812.
14. Rajagopal, S.M., Supriya, M., Buyya, R. (2025). Leveraging blockchain and federated learning in Edge-Fog-Cloud computing environments for intelligent decision-making with ECG data in IoT, Journal of Network and Computer Applications, 233, 104037.
15. Qian, Z., Li, G., Qi, T., Dai, C. (2025). Federated deep reinforcement learning-based cost-efficient proactive video caching in energy-constrained mobile edge networks, Computer Networks, 258, 111062.
16. Sabah, F., Chen, Y., Yang, Z., Raheem, A., Azam, M., Ahmad, N., Sarwar, R. (2025). Communication optimization techniques in Personalized Federated Learning: Applications, challenges and future directions, Information Fusion, 117, 102834.
17. Sattler, F. (2021). Concepts for Efficient, Adaptive and Robust Deep Learning from Distributed Data, Technische Universitaet Berlin, Germany.
18. 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.
19. Singh, B., Adhikari, M. (2025). PopFL: A scalable Federated Learning model in serverless edge computing integrating with dynamic pop-up network, Ad Hoc Networks, 169. 103728.
20. Wang, B., Tian, Z., Ma, J., Zhang, W., She, W., Liu, W. (2025). A decentralized asynchronous federated learning framework for edge devices, Future Generation Computer Systems, 166, 107683.
21. Zhang, J., Wu, Q., Fan, P., Fan, Q.(2024). A Comprehensive Survey on Joint Resource Allocation Strategies in Federated Edge Learning, Computers, Materials and Continua, 81(2), 1953-1998.
22. Zhao, X., Wu, Y., Zhao, T., Wang, F., Li, M. (2024). Federated deep reinforcement learning for task offloading and resource allocation in mobile edge computing-assisted vehicular networks, Journal of Network and Computer Applications, 229, 103941.