Optimization of Federated Learning in Edge Computing Using Adaptive Client Selection

Arkan Zayd; Farzana Farzana; Thariq Thariq

Penulis

Arkan Zayd Information & Computing Technology Division, College of Science and Engineering, Hamad Bin Khalifa University, Doha, Qatar
Farzana Farzana Department of Electrical Engineering, Information Technology University, Lahore, Pakistan,
Thariq Thariq Department of Electrical Engineering, Information Technology University, Lahore, Pakistan

Kata Kunci:

Federated Learning, Edge Computing, Adaptive Client Selection, Resource Heterogeneity, Communication Efficiency

Abstrak

The rapid growth of edge computing has created new opportunities for processing data closer to end users and connected devices. Federated Learning (FL) has emerged as an important distributed learning paradigm because it enables collaborative model training without requiring raw data to be transferred to a central server. However, heterogeneous edge environments involve substantial differences in computational capacity, network bandwidth, latency, energy availability, data contribution, and client reliability. Conventional client-selection strategies, such as random or uniform selection, may therefore select inefficient clients, increase communication costs, prolong training latency, and reduce convergence efficiency. This study aims to develop an adaptive client-selection mechanism to optimize federated learning performance in edge computing environments. Selected client models were aggregated using federated averaging and compared with random, uniform, and resource-based selection methods. Performance was evaluated using model accuracy, convergence rounds, communication overhead, training latency, and energy consumption. The experimental results showed an accuracy improvement of 1.2 percentage points, communication reduction of 28.0%, latency reduction of 31.5%, energy reduction of 34.5%, and convergence improvement of 34.1% compared with random client selection. These findings indicate that adaptive client selection can improve the efficiency, resource utilization, and scalability of federated learning under heterogeneous edge-computing conditions.

Referensi

Abd Aziz, A., Sekercioglu, Y. A., Fitzpatrick, P., & Ivanovich, M. (2012). A survey on distributed topology control techniques for extending the lifetime of battery powered wireless sensor networks. IEEE Communications Surveys & Tutorials, 15(1), 121–144.

Abdelzaher, T. F., & Bhatti, N. (1999). Web content adaptation to improve server overload behavior. Computer Networks, 31(11–16), 1563–1577.

Alizadeh, M., Kabbani, A., Edsall, T., Prabhakar, B., Vahdat, A., & Yasuda, M. (2012). Less is more: Trading a little bandwidth for {Ultra-Low} latency in the data center. 9th USENIX Symposium on Networked Systems Design and Implementation (NSDI 12), 253–266.

Beutel, D. J., Topal, T., Mathur, A., Qiu, X., Fernandez-Marques, J., Gao, Y., Sani, L., Li, K. H., Parcollet, T., & de GusmÃ?o, P. P. B. (2020). Flower: A friendly federated learning research framework. ArXiv Preprint ArXiv:2007.14390.

Chen, K., Zhang, J., Xiao, Y., Jo, M., & Ng, D. W. K. (2025). Completion time minimization for uav-assisted semi-decentralized hybrid federated learning. IEEE Transactions on Mobile Computing.

Cho, Y. J., Wang, J., & Joshi, G. (2020). Client selection in federated learning: Convergence analysis and power-of-choice selection strategies. ArXiv Preprint ArXiv:2010.01243.

Cong, K., Moreno, R. C., Da Gama, M. B., Dai, W., Iliashenko, I., Laine, K., & Rosenberg, M. (2021). Labeled PSI from homomorphic encryption with reduced computation and communication. Proceedings of the 2021 ACM SIGSAC Conference on Computer and Communications Security, 1135–1150.

Diao, E., Ding, J., & Tarokh, V. (2020). Heterofl: Computation and communication efficient federated learning for heterogeneous clients. ArXiv Preprint ArXiv:2010.01264.

Guo, Z., Blanco, Z., Chen, J., Li, J., Wei, Z., Dong, B., Pota, I., Shahrad, M., Xu, H., & Zhang, Y. (2022). Zenix: Efficient execution of bulky serverless applications. ArXiv Preprint ArXiv:2206.13444.

Hsu, T.-M. H., Qi, H., & Brown, M. (2020). Federated visual classification with real-world data distribution. European Conference on Computer Vision, 76–92.

Kou, W.-B., Lin, Q., Tang, M., Ye, R., Wang, S., Zhu, G., & Wu, Y.-C. (2025). Fast-convergent and communication-alleviated heterogeneous hierarchical federated learning in autonomous driving. IEEE Transactions on Intelligent Transportation Systems, 26(7), 10496–10511.

Li, E., Zeng, L., Zhou, Z., & Chen, X. (2019). Edge AI: On-demand accelerating deep neural network inference via edge computing. IEEE Transactions on Wireless Communications, 19(1), 447–457.

Li, Y., Wang, X., Li, H., Donta, P. K., Huang, M., & Dustdar, S. (2025). Communication-efficient federated learning for heterogeneous clients. ACM Transactions on Internet Technology, 25(2), 1–37.

Lin, W., Xu, Y., Liu, B., Li, D., Huang, T., & Shi, F. (2022). Contribution?based Federated Learning client selection. International Journal of Intelligent Systems, 37(10), 7235–7260.

Lin, W., & Zhou, Y. (2026). Adaptive client participation mechanism for federated learning in heterogeneous vehicular networks. Scientific Reports.

Liu, J., Du, Y., Yang, K., Wu, J., Wang, Y., Hu, X., Wang, Z., Liu, Y., Sun, P., & Boukerche, A. (2026). Edge-cloud collaborative computing on distributed intelligence and model optimization: A survey. IEEE Communications Surveys & Tutorials.

Magnússon, S., Shokri-Ghadikolaei, H., & Li, N. (2020). On maintaining linear convergence of distributed learning and optimization under limited communication. IEEE Transactions on Signal Processing, 68, 6101–6116.

Nariman, G. S., & Hamarashid, H. K. (2025). Communication overhead reduction in federated learning: a review. International Journal of Data Science and Analytics, 19(2), 185–216.

Nguyen, D. C., Ding, M., Pathirana, P. N., Seneviratne, A., Li, J., & Poor, H. V. (2021). Federated learning for internet of things: A comprehensive survey. IEEE Communications Surveys & Tutorials, 23(3), 1622–1658.

Nishio, T., & Yonetani, R. (2019). Client selection for federated learning with heterogeneous resources in mobile edge. ICC 2019-2019 IEEE International Conference on Communications (ICC), 1–7.

Niu, C., Wu, F., Tang, S., Hua, L., Jia, R., Lv, C., Wu, Z., & Chen, G. (2020). Billion-scale federated learning on mobile clients: A submodel design with tunable privacy. Proceedings of the 26th Annual International Conference on Mobile Computing and Networking, 1–14.

Rishika, R., Kumar, A., Janakiraman, R., & Bezawada, R. (2013). The effect of customers’ social media participation on customer visit frequency and profitability: an empirical investigation. Information Systems Research, 24(1), 108–127.

Rogers, S. E., & Kwak, D. (1991). An upwind differencing scheme for the incompressible Navier–Strokes equations. Applied Numerical Mathematics, 8(1), 43–64.

Savoia, M., Prezioso, E., Mele, V., & Piccialli, F. (2024). Eco-FL: Enhancing federated learning sustainability in edge computing through energy-efficient client selection. Computer Communications, 225, 156–170.

Shafique, K., Khawaja, B. A., Sabir, F., Qazi, S., & Mustaqim, M. (2020). Internet of things (IoT) for next-generation smart systems: A review of current challenges, future trends and prospects for emerging 5G-IoT scenarios. IEEE Access, 8, 23022–23040.

Tornero-Soria, M., Sánchez-Salmerón, A.-J., & Vendrell Vidal, E. (2026). Toward a Deeper Understanding of YOLO26: Block-Level Architectural Analysis and Ablation Studies. Applied Sciences, 16(13), 6758.

Veiga, R., Morais, R., Seruffo, M., Rosário, D., & Cerqueira, E. (2024). RICAM: An efficient multicriteria client selection mechanism for federated learning. 2024 IEEE 13th International Conference on Cloud Networking (CloudNet), 1–8.

Wang, X., Zhang, R., Shen, C., Kong, T., & Li, L. (2021). Dense contrastive learning for self-supervised visual pre-training. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 3023–3032.

Yuan, L., Xie, B., & Li, S. (2023). Robust test-time adaptation in dynamic scenarios. 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 15922–15932.

Yue, J., Lang, J., & Feng, R. (2025). An adaptive feature fusion strategy using dual-layer attention and multi-modal deep reinforcement learning for all-media similarity search. Discover Artificial Intelligence, 5(1), 71.