International Journal of Innovative Research in Computer and Communication Engineering
ISSN Approved Journal | Impact factor: 8.771 | ESTD: 2013 | Follows UGC CARE Journal Norms and Guidelines
| Monthly, Peer-Reviewed, Refereed, Scholarly, Multidisciplinary and Open Access Journal | High Impact Factor 8.771 (Calculated by Google Scholar and Semantic Scholar | AI-Powered Research Tool | Indexing in all Major Database & Metadata, Citation Generator | Digital Object Identifier (DOI) |
| TITLE | Neuro-Fuzzy Driven Client Selection for Efficient Federated Learning |
|---|---|
| ABSTRACT | Federated Learning (FL) enables multiple distributed devices to collaboratively train a shared machine learning model while retaining data locally, thereby improving privacy and reducing the need for centralized data collection. However, the effectiveness of FL is strongly influenced by the heterogeneous characteristics of participating clients. Variations in battery capacity, communication bandwidth, computational capability, data quality, and device reliability often lead to inefficient client participation, increased communication overhead, and slower model convergence. Conventional client selection strategies such as random or static threshold-based approaches are unable to adapt to continuously changing device conditions. Neuro-Fuzzy client selection framework for heterogeneous Federated Learning environments. The proposed framework integrates an Adaptive Neuro-Fuzzy Inference System (ANFIS) with the Federated Averaging (FedAvg) algorithm to evaluate client suitability using multiple resource-aware features. Fuzzy reasoning captures uncertainty in device characteristics, while adaptive learning continuously refines the selection policy based on feedback obtained during successive communication rounds. The framework further incorporates realistic device simulation, including dynamic battery variation, bandwidth fluctuation, client churn, and both IID and Non-IID data distributions, to emulate practical mobile computing environments. Experimental evaluation demonstrates that the proposed Neuro-Fuzzy strategy consistently selects more suitable clients than conventional selection methods, leading to faster convergence, improved mean model accuracy, reduced energy consumption, lower training time. The proposed approach provides an adaptive and resource-aware client selection mechanism that enhances the efficiency and scalability of Federated Learning in heterogeneous edge and mobile networks. |
| AUTHOR | SYED UMER, DR. M. NAGARATNA Post-Graduate Student, Department of Computer Science Engineering, Jawaharlal Nehru Technological University, Hyderabad, India Professor, Department of Computer Science Engineering, Jawaharlal Nehru Technological University, Hyderabad, India |
| VOLUME | 185 |
| DOI | DOI: 10.15680/IJIRCCE.2026.1406096 |
| pdf/96_Neuro-Fuzzy Driven Client Selection for Efficient Federated Learning.pdf | |
| KEYWORDS | |
| References | 1. Brendan McMahan, Eider Moore, Daniel Ramage, Seth Hampson and Blaise Agüera y Arcas, "Communication-Efficient Learning of Deep Networks from Decentralized Data," Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS), Vol. 54, pp. 1273-1282, 2017. 2. Zhaoyang Du, Celimuge Wu, Tsutomu Yoshinaga, Lei Zhong and Yusheng Ji, "On-device Federated Learning with Fuzzy Logic Based Client Selection," Proceedings of the 2022 International Conference on Research in Adaptive and Convergent Systems (RACS), pp. 64-70, 2022. DOI: 10.1145/3538641.3561490. 3. Narisu Cha, Zhaoyang Du, Celimuge Wu, Tsutomu Yoshinaga, Lei Zhong, Jing Ma, Fuqiang Liu and Yusheng Ji, "Fuzzy Logic Based Client Selection for Federated Learning in Vehicular Networks," IEEE Open Journal of the Computer Society, Vol. 3, pp. 39-50, 2022. DOI: 10.1109/OJCS.2022.3163620. 4. Zhihao Dong, Xu Zhu, Jie Cao, Yufei Jiang, Vincent K. N. Lau and Sumei Sun, "Fuzzy Logic Assisted Client Selection and Energy-Efficient Joint Optimization for Hierarchical Federated Learning," Proceedings of the IEEE International Conference on Communications (ICC 2023), pp. 1262-1267, 2023. DOI: 10.1109/ICC45041.2023.10279572. 5. akayuki Nishio and Ryo Yonetani, "Client Selection for Federated Learning with Heterogeneous Resources in Mobile Edge," Proceedings of the IEEE International Conference on Communications (ICC), pp. 1–7, 2019. DOI: 10.1109/ICC.2019.8761315. 6. Xiaobing Chen, Xiangwei Zhou, Hongchao Zhang, Mingxuan Sun and H. Vincent Poor, "Client Selection for Wireless Federated Learning With Data and Latency Heterogeneity," IEEE Internet of Things Journal, Vol. 11, No. 19, pp. 32183–32196, 2024. DOI: 10.1109/JIOT.2024.3425757. 7. Manas Panigrahi, Shashank Bharti and Amit Sharma, "A Review on Client Selection Models in Federated Learning," WIREs Data Mining and Knowledge Discovery, Vol. 13, No. 6, Article e1514, 2023. DOI: 10.1002/widm.1514. 8. Rituparna Saha, Sudip Misra, Aishwariya Chakraborty, Chandranath Chatterjee and Pallav K. Deb, "Data-Centric Client Selection for Federated Learning over Distributed Edge Networks," IEEE Transactions on Parallel and Distributed Systems, Vol. 34, No. 2, pp. 675–686, 2023. DOI: 10.1109/TPDS.2022.3217271. |
