A Novel Multi-Server Federated Learning Framework in Vehicular Edge Computing

Federated learning (FL) has emerged as a powerful approach for privacy-preserving model training in autonomous vehicle networks, where real-world deployments rely on multiple roadside units (RSUs) serving heterogeneous clients with intermittent connectivity. While most research focuses on single-ser...

Full description

Saved in:
Bibliographic Details
Main Authors: Fateme Mazloomi, Shahram Shah Heydari, Khalil El-Khatib
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:Future Internet
Subjects:
Online Access:https://www.mdpi.com/1999-5903/17/7/315
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849418564916740096
author Fateme Mazloomi
Shahram Shah Heydari
Khalil El-Khatib
author_facet Fateme Mazloomi
Shahram Shah Heydari
Khalil El-Khatib
author_sort Fateme Mazloomi
collection DOAJ
description Federated learning (FL) has emerged as a powerful approach for privacy-preserving model training in autonomous vehicle networks, where real-world deployments rely on multiple roadside units (RSUs) serving heterogeneous clients with intermittent connectivity. While most research focuses on single-server or hierarchical cloud-based FL, multi-server FL can alleviate the communication bottlenecks of traditional setups. To this end, we propose an edge-based, multi-server FL (MS-FL) framework that combines performance-driven aggregation at each server—including statistical weighting of peer updates and outlier mitigation—with an application layer handover protocol that preserves model updates when vehicles move between RSU coverage areas. We evaluate MS-FL on both MNIST and GTSRB benchmarks under shard- and Dirichlet-based non-IID splits, comparing it against single-server FL and a two-layer edge-plus-cloud baseline. Over multiple communication rounds, MS-FL with the Statistical Performance-Aware Aggregation method and Dynamic Weighted Averaging Aggregation achieved up to a 20-percentage-point improvement in accuracy and consistent gains in precision, recall, and F1-score (95% confidence), while matching the low latency of edge-only schemes and avoiding the extra model transfer delays of cloud-based aggregation. These results demonstrate that coordinated cooperation among servers based on model quality and seamless handovers can accelerate convergence, mitigate data heterogeneity, and deliver robust, privacy-aware learning in connected vehicle environments.
format Article
id doaj-art-b66d3c08adaa41adbfc31a3255c8c162
institution Kabale University
issn 1999-5903
language English
publishDate 2025-07-01
publisher MDPI AG
record_format Article
series Future Internet
spelling doaj-art-b66d3c08adaa41adbfc31a3255c8c1622025-08-20T03:32:26ZengMDPI AGFuture Internet1999-59032025-07-0117731510.3390/fi17070315A Novel Multi-Server Federated Learning Framework in Vehicular Edge ComputingFateme Mazloomi0Shahram Shah Heydari1Khalil El-Khatib2Faculty of Business and IT, University of Ontario Institute of Technology, Oshawa, ON L1G 0C5, CanadaFaculty of Business and IT, University of Ontario Institute of Technology, Oshawa, ON L1G 0C5, CanadaFaculty of Business and IT, University of Ontario Institute of Technology, Oshawa, ON L1G 0C5, CanadaFederated learning (FL) has emerged as a powerful approach for privacy-preserving model training in autonomous vehicle networks, where real-world deployments rely on multiple roadside units (RSUs) serving heterogeneous clients with intermittent connectivity. While most research focuses on single-server or hierarchical cloud-based FL, multi-server FL can alleviate the communication bottlenecks of traditional setups. To this end, we propose an edge-based, multi-server FL (MS-FL) framework that combines performance-driven aggregation at each server—including statistical weighting of peer updates and outlier mitigation—with an application layer handover protocol that preserves model updates when vehicles move between RSU coverage areas. We evaluate MS-FL on both MNIST and GTSRB benchmarks under shard- and Dirichlet-based non-IID splits, comparing it against single-server FL and a two-layer edge-plus-cloud baseline. Over multiple communication rounds, MS-FL with the Statistical Performance-Aware Aggregation method and Dynamic Weighted Averaging Aggregation achieved up to a 20-percentage-point improvement in accuracy and consistent gains in precision, recall, and F1-score (95% confidence), while matching the low latency of edge-only schemes and avoiding the extra model transfer delays of cloud-based aggregation. These results demonstrate that coordinated cooperation among servers based on model quality and seamless handovers can accelerate convergence, mitigate data heterogeneity, and deliver robust, privacy-aware learning in connected vehicle environments.https://www.mdpi.com/1999-5903/17/7/315federated learningmulti-servermobilityhand overaggregationvehicular edge computing
spellingShingle Fateme Mazloomi
Shahram Shah Heydari
Khalil El-Khatib
A Novel Multi-Server Federated Learning Framework in Vehicular Edge Computing
Future Internet
federated learning
multi-server
mobility
hand over
aggregation
vehicular edge computing
title A Novel Multi-Server Federated Learning Framework in Vehicular Edge Computing
title_full A Novel Multi-Server Federated Learning Framework in Vehicular Edge Computing
title_fullStr A Novel Multi-Server Federated Learning Framework in Vehicular Edge Computing
title_full_unstemmed A Novel Multi-Server Federated Learning Framework in Vehicular Edge Computing
title_short A Novel Multi-Server Federated Learning Framework in Vehicular Edge Computing
title_sort novel multi server federated learning framework in vehicular edge computing
topic federated learning
multi-server
mobility
hand over
aggregation
vehicular edge computing
url https://www.mdpi.com/1999-5903/17/7/315
work_keys_str_mv AT fatememazloomi anovelmultiserverfederatedlearningframeworkinvehicularedgecomputing
AT shahramshahheydari anovelmultiserverfederatedlearningframeworkinvehicularedgecomputing
AT khalilelkhatib anovelmultiserverfederatedlearningframeworkinvehicularedgecomputing
AT fatememazloomi novelmultiserverfederatedlearningframeworkinvehicularedgecomputing
AT shahramshahheydari novelmultiserverfederatedlearningframeworkinvehicularedgecomputing
AT khalilelkhatib novelmultiserverfederatedlearningframeworkinvehicularedgecomputing