Adaptive federated learning for resource-constrained IoT devices through edge intelligence and multi-edge clustering

Abstract In the rapidly growing Internet of Things (IoT) landscape, federated learning (FL) plays a crucial role in enhancing the performance of heterogeneous edge computing environments due to its scalability, robustness, and low energy consumption. However, one of the major challenges in such envi...

Full description

Saved in:
Bibliographic Details
Main Authors: Fahad Razaque Mughal, Jingsha He, Bhagwan Das, Fayaz Ali Dharejo, Nafei Zhu, Surbhi Bhatia Khan, Saeed Alzahrani
Format: Article
Language:English
Published: Nature Portfolio 2024-11-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-024-78239-z
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846158587400814592
author Fahad Razaque Mughal
Jingsha He
Bhagwan Das
Fayaz Ali Dharejo
Nafei Zhu
Surbhi Bhatia Khan
Saeed Alzahrani
author_facet Fahad Razaque Mughal
Jingsha He
Bhagwan Das
Fayaz Ali Dharejo
Nafei Zhu
Surbhi Bhatia Khan
Saeed Alzahrani
author_sort Fahad Razaque Mughal
collection DOAJ
description Abstract In the rapidly growing Internet of Things (IoT) landscape, federated learning (FL) plays a crucial role in enhancing the performance of heterogeneous edge computing environments due to its scalability, robustness, and low energy consumption. However, one of the major challenges in such environments is the efficient selection of edge nodes and the optimization of resource allocation, especially in dynamic and resource-constrained settings. To address this, we propose a novel architecture called Multi-Edge Clustered and Edge AI Heterogeneous Federated Learning (MEC-AI HetFL), which leverages multi-edge clustering and AI-driven node communication. This architecture enables edge AI nodes to collaborate, dynamically selecting significant nodes and optimizing global learning tasks with low complexity. Compared to existing solutions like EdgeFed, FedSA, FedMP, and H-DDPG, MEC-AI HetFL improves resource allocation, quality score, and learning accuracy, offering up to 5 times better performance in heterogeneous and distributed environments. The solution is validated through simulations and network traffic tests, demonstrating its ability to address the key challenges in IoT edge computing deployments.
format Article
id doaj-art-4979d87aa7f540d59d48df5b2bdca758
institution Kabale University
issn 2045-2322
language English
publishDate 2024-11-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-4979d87aa7f540d59d48df5b2bdca7582024-11-24T12:22:25ZengNature PortfolioScientific Reports2045-23222024-11-0114111910.1038/s41598-024-78239-zAdaptive federated learning for resource-constrained IoT devices through edge intelligence and multi-edge clusteringFahad Razaque Mughal0Jingsha He1Bhagwan Das2Fayaz Ali Dharejo3Nafei Zhu4Surbhi Bhatia Khan5Saeed Alzahrani6Faculty of Information Technology, Beijing University of TechnologyFaculty of Information Technology, Beijing University of TechnologyCentre for Artificial Intelligence Research and Optimization (AIRO), Design and Creative Technology Vertical, Torrens UniversityComputer Vision Lab, CAIDAS, IFI, University of WurzburgFaculty of Information Technology, Beijing University of TechnologySchool of Science, Engineering and Environment, University of SalfordManagement Information System, King Saud UniversityAbstract In the rapidly growing Internet of Things (IoT) landscape, federated learning (FL) plays a crucial role in enhancing the performance of heterogeneous edge computing environments due to its scalability, robustness, and low energy consumption. However, one of the major challenges in such environments is the efficient selection of edge nodes and the optimization of resource allocation, especially in dynamic and resource-constrained settings. To address this, we propose a novel architecture called Multi-Edge Clustered and Edge AI Heterogeneous Federated Learning (MEC-AI HetFL), which leverages multi-edge clustering and AI-driven node communication. This architecture enables edge AI nodes to collaborate, dynamically selecting significant nodes and optimizing global learning tasks with low complexity. Compared to existing solutions like EdgeFed, FedSA, FedMP, and H-DDPG, MEC-AI HetFL improves resource allocation, quality score, and learning accuracy, offering up to 5 times better performance in heterogeneous and distributed environments. The solution is validated through simulations and network traffic tests, demonstrating its ability to address the key challenges in IoT edge computing deployments.https://doi.org/10.1038/s41598-024-78239-zInternet of thingsResource managementEdge artificial intelligenceHeterogeneous cluster networksEdge computingFederated learning
spellingShingle Fahad Razaque Mughal
Jingsha He
Bhagwan Das
Fayaz Ali Dharejo
Nafei Zhu
Surbhi Bhatia Khan
Saeed Alzahrani
Adaptive federated learning for resource-constrained IoT devices through edge intelligence and multi-edge clustering
Scientific Reports
Internet of things
Resource management
Edge artificial intelligence
Heterogeneous cluster networks
Edge computing
Federated learning
title Adaptive federated learning for resource-constrained IoT devices through edge intelligence and multi-edge clustering
title_full Adaptive federated learning for resource-constrained IoT devices through edge intelligence and multi-edge clustering
title_fullStr Adaptive federated learning for resource-constrained IoT devices through edge intelligence and multi-edge clustering
title_full_unstemmed Adaptive federated learning for resource-constrained IoT devices through edge intelligence and multi-edge clustering
title_short Adaptive federated learning for resource-constrained IoT devices through edge intelligence and multi-edge clustering
title_sort adaptive federated learning for resource constrained iot devices through edge intelligence and multi edge clustering
topic Internet of things
Resource management
Edge artificial intelligence
Heterogeneous cluster networks
Edge computing
Federated learning
url https://doi.org/10.1038/s41598-024-78239-z
work_keys_str_mv AT fahadrazaquemughal adaptivefederatedlearningforresourceconstrainediotdevicesthroughedgeintelligenceandmultiedgeclustering
AT jingshahe adaptivefederatedlearningforresourceconstrainediotdevicesthroughedgeintelligenceandmultiedgeclustering
AT bhagwandas adaptivefederatedlearningforresourceconstrainediotdevicesthroughedgeintelligenceandmultiedgeclustering
AT fayazalidharejo adaptivefederatedlearningforresourceconstrainediotdevicesthroughedgeintelligenceandmultiedgeclustering
AT nafeizhu adaptivefederatedlearningforresourceconstrainediotdevicesthroughedgeintelligenceandmultiedgeclustering
AT surbhibhatiakhan adaptivefederatedlearningforresourceconstrainediotdevicesthroughedgeintelligenceandmultiedgeclustering
AT saeedalzahrani adaptivefederatedlearningforresourceconstrainediotdevicesthroughedgeintelligenceandmultiedgeclustering