A Framework Integrating Federated Learning and Fog Computing Based on Client Sampling and Dynamic Thresholding Techniques
The exponential growth in the number of Internet of Things (IoT) devices and the vast quantity of data they generate present a significant challenge to the efficacy of traditional centralized training models. Federated Learning (FL) is a machine learning framework that effectively addresses this iss...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11007538/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850232435766198272 |
|---|---|
| author | Dang van Thang Artem Volkov Ammar Muthanna Ibrahim A. Elgendy Reem Alkanhel Dushantha Nalin K. Jayakody Andrey Koucheryavy |
| author_facet | Dang van Thang Artem Volkov Ammar Muthanna Ibrahim A. Elgendy Reem Alkanhel Dushantha Nalin K. Jayakody Andrey Koucheryavy |
| author_sort | Dang van Thang |
| collection | DOAJ |
| description | The exponential growth in the number of Internet of Things (IoT) devices and the vast quantity of data they generate present a significant challenge to the efficacy of traditional centralized training models. Federated Learning (FL) is a machine learning framework that effectively addresses this issue and other concerns about data privacy. Furthermore, fog computing represents a robust distributed computing methodology with the potential to bolster and propel the advancement of FL. An integrated distributed architecture combining FL and fog computing (FC) has the potential to overcome the limitations of traditional centralized architectures, offering a promising solution for the future. One of the objectives of implementing this novel architectural framework is to alleviate the burden on communication links within the core network by training a model on distributed training data across many clients. Various techniques and frameworks have been developed and implemented, including approaches to model compression and those addressing data and device heterogeneity. These have demonstrated effectiveness in specific contexts. In this paper, we introduce a novel gradient-driven client-sampling framework that tightly couples Federated Learning with Fog Computing. By dynamically adjusting per-round thresholds based on local gradient change rates, our method selects only the most informative clients and leverages fog nodes for partial aggregation, thereby minimizing redundant transmissions, accelerating convergence under heterogeneous data, and offloading the central server. Extensive simulations on MNIST and CIFAR-10 demonstrate that our approach reduces cumulative communication by 39% and 31%, respectively, without sacrificing convergence speed or final accuracy. |
| format | Article |
| id | doaj-art-495d4fa9e1464be2aa4c60ed6b9cf8b7 |
| institution | OA Journals |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-495d4fa9e1464be2aa4c60ed6b9cf8b72025-08-20T02:03:13ZengIEEEIEEE Access2169-35362025-01-0113950199503310.1109/ACCESS.2025.357197911007538A Framework Integrating Federated Learning and Fog Computing Based on Client Sampling and Dynamic Thresholding TechniquesDang van Thang0https://orcid.org/0009-0009-2219-3767Artem Volkov1https://orcid.org/0009-0002-4296-1822Ammar Muthanna2Ibrahim A. Elgendy3https://orcid.org/0000-0001-7154-2307Reem Alkanhel4https://orcid.org/0000-0001-6395-4723Dushantha Nalin K. Jayakody5Andrey Koucheryavy6Department of Telecommunication Networks and Data Transmission, The Bonch-Bruevich Saint-Petersburg State University of Telecommunications, Saint Petersburg, RussiaDepartment of Telecommunication Networks and Data Transmission, The Bonch-Bruevich Saint-Petersburg State University of Telecommunications, Saint Petersburg, RussiaDepartment of Telecommunication Networks and Data Transmission, The Bonch-Bruevich Saint-Petersburg State University of Telecommunications, Saint Petersburg, RussiaIRC for Finance and Digital Economy, KFUPM Business School, King Fahd University of Petroleum and Minerals, Dhahran, Saudi ArabiaDepartment of Information Technology, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman University, P.O. Box 84428, Riyadh, Saudi ArabiaCOPELABS, Universidade Lusófona, Lisbon, PortugalDepartment of Telecommunication Networks and Data Transmission, The Bonch-Bruevich Saint-Petersburg State University of Telecommunications, Saint Petersburg, RussiaThe exponential growth in the number of Internet of Things (IoT) devices and the vast quantity of data they generate present a significant challenge to the efficacy of traditional centralized training models. Federated Learning (FL) is a machine learning framework that effectively addresses this issue and other concerns about data privacy. Furthermore, fog computing represents a robust distributed computing methodology with the potential to bolster and propel the advancement of FL. An integrated distributed architecture combining FL and fog computing (FC) has the potential to overcome the limitations of traditional centralized architectures, offering a promising solution for the future. One of the objectives of implementing this novel architectural framework is to alleviate the burden on communication links within the core network by training a model on distributed training data across many clients. Various techniques and frameworks have been developed and implemented, including approaches to model compression and those addressing data and device heterogeneity. These have demonstrated effectiveness in specific contexts. In this paper, we introduce a novel gradient-driven client-sampling framework that tightly couples Federated Learning with Fog Computing. By dynamically adjusting per-round thresholds based on local gradient change rates, our method selects only the most informative clients and leverages fog nodes for partial aggregation, thereby minimizing redundant transmissions, accelerating convergence under heterogeneous data, and offloading the central server. Extensive simulations on MNIST and CIFAR-10 demonstrate that our approach reduces cumulative communication by 39% and 31%, respectively, without sacrificing convergence speed or final accuracy.https://ieeexplore.ieee.org/document/11007538/Federated learningfog computingclient samplingdynamic thresholding |
| spellingShingle | Dang van Thang Artem Volkov Ammar Muthanna Ibrahim A. Elgendy Reem Alkanhel Dushantha Nalin K. Jayakody Andrey Koucheryavy A Framework Integrating Federated Learning and Fog Computing Based on Client Sampling and Dynamic Thresholding Techniques IEEE Access Federated learning fog computing client sampling dynamic thresholding |
| title | A Framework Integrating Federated Learning and Fog Computing Based on Client Sampling and Dynamic Thresholding Techniques |
| title_full | A Framework Integrating Federated Learning and Fog Computing Based on Client Sampling and Dynamic Thresholding Techniques |
| title_fullStr | A Framework Integrating Federated Learning and Fog Computing Based on Client Sampling and Dynamic Thresholding Techniques |
| title_full_unstemmed | A Framework Integrating Federated Learning and Fog Computing Based on Client Sampling and Dynamic Thresholding Techniques |
| title_short | A Framework Integrating Federated Learning and Fog Computing Based on Client Sampling and Dynamic Thresholding Techniques |
| title_sort | framework integrating federated learning and fog computing based on client sampling and dynamic thresholding techniques |
| topic | Federated learning fog computing client sampling dynamic thresholding |
| url | https://ieeexplore.ieee.org/document/11007538/ |
| work_keys_str_mv | AT dangvanthang aframeworkintegratingfederatedlearningandfogcomputingbasedonclientsamplinganddynamicthresholdingtechniques AT artemvolkov aframeworkintegratingfederatedlearningandfogcomputingbasedonclientsamplinganddynamicthresholdingtechniques AT ammarmuthanna aframeworkintegratingfederatedlearningandfogcomputingbasedonclientsamplinganddynamicthresholdingtechniques AT ibrahimaelgendy aframeworkintegratingfederatedlearningandfogcomputingbasedonclientsamplinganddynamicthresholdingtechniques AT reemalkanhel aframeworkintegratingfederatedlearningandfogcomputingbasedonclientsamplinganddynamicthresholdingtechniques AT dushanthanalinkjayakody aframeworkintegratingfederatedlearningandfogcomputingbasedonclientsamplinganddynamicthresholdingtechniques AT andreykoucheryavy aframeworkintegratingfederatedlearningandfogcomputingbasedonclientsamplinganddynamicthresholdingtechniques AT dangvanthang frameworkintegratingfederatedlearningandfogcomputingbasedonclientsamplinganddynamicthresholdingtechniques AT artemvolkov frameworkintegratingfederatedlearningandfogcomputingbasedonclientsamplinganddynamicthresholdingtechniques AT ammarmuthanna frameworkintegratingfederatedlearningandfogcomputingbasedonclientsamplinganddynamicthresholdingtechniques AT ibrahimaelgendy frameworkintegratingfederatedlearningandfogcomputingbasedonclientsamplinganddynamicthresholdingtechniques AT reemalkanhel frameworkintegratingfederatedlearningandfogcomputingbasedonclientsamplinganddynamicthresholdingtechniques AT dushanthanalinkjayakody frameworkintegratingfederatedlearningandfogcomputingbasedonclientsamplinganddynamicthresholdingtechniques AT andreykoucheryavy frameworkintegratingfederatedlearningandfogcomputingbasedonclientsamplinganddynamicthresholdingtechniques |