Enhancing Vehicular Network Safety through Cluster Head-based Edge Intelligence and Federated Learning

Document Type

Conference Proceeding

Source of Publication

2025 7th International Conference on Blockchain Computing and Applications (BCCA)

Publication Date

10-14-2025

Abstract

Safety in Vehicular Networks hinges on the imperative need for real-time decision-making at the network edge. As vehicles traverse dynamic and often unpredictable environments, the ability to make split-second decisions is paramount to preventing collisions and ensuring overall road safety. Given this, the role of a designated cluster head as the network edge controller becomes crucial. This research addresses the challenge of realtime decision-making to ensure vehicular safety by leveraging a cluster head architecture at the network edge. The cluster head, acting as an intelligent controller, facilitates rapid decisionmaking processes that are vital in preventing accidents and promoting safe driving practices. To enhance the intelligence of the cluster head and, consequently, the safety of Vehicular Networks, Federated Learning is employed. Federated Learning enables vehicles within the network to collaboratively train machine learning models for edge intelligence. By harnessing the collective insights from individual vehicles, these models improve collision prediction and avoidance mechanisms. The collaborative nature of Federated Learning ensures that the intelligence gained is representative of diverse driving scenarios and behaviors. Vehicles contribute to the training process without compromising sensitive data, as learning occurs locally on individual devices. The aggregated knowledge is then used to empower the cluster head with enhanced collision prediction capabilities, enabling it to make informed decisions in real time. This research contributes to advancing the safety paradigm in Vehicular Networks by integrating the cluster head-based edge intelligence architecture with Federated Learning. The synergy of these technologies not only facilitates immediate responses to potential hazards but also fosters a collective, distributed approach to safety that aligns with the collaborative nature of modern Vehicular Networks.

Publisher

IEEE

First Page

873

Last Page

881

Disciplines

Electrical and Computer Engineering

Keywords

Vehicular Ad Hoc Networks (VANETs), Privacy-Preserving Technologies in Data, Traffic control and management

Indexed in Scopus

no

Open Access

no

Link to Full Text

Share

COinS