Vehicle ad-hoc networks (VANET) have completely transformed traditional transportation systems, paving the way for intelligent transportation systems. Vehicles within a VANET cluster typically have similar speeds and distances, and traffic conflicts often occur between different clusters. Preventing secondary traffic events triggered by these conflicts requires distributing information to multiple clusters simultaneously, resulting in a high communication burden. Thus, this research proposes a semi-distributed clustering algorithm based on vehicle potential influence to improve communication performance. Specifically, the framework can be divided into three parts, i.e., cluster initialization, merging, and maintenance. Then, authors design vehicle influence areas based on motion rules and propose two influence metrics, i.e., distance influence and mobility influence which can be further divided into longitudinal influence and lateral influence. To validate the performance of the proposed algorithm in terms of cluster modularity, cluster stability, and inter-cluster safety, a simulation is conducted on a real word expressway road network with different traffic flow rates. The results show that the proposed VANET protocol produces a higher topology quality, greater cluster lifetime, and especially a lower inter-cluster conflict rate, which is more suitable for enhancing traffic safety and improving the efficiency of information dissemination in large-scale and highly dynamic scenarios.
A Semi-Distributed Influence-Aware Clustering Algorithm for Enhancing Traffic Safety in VANETs
2023-09-24
804884 byte
Conference paper
Electronic Resource
English
An intelligent clustering algorithm for VANETs
IEEE | 2014
|Distributed rate control algorithm for VANETs (DRCV)
Tema Archive | 2009
|A Traffic-Flow-Phase Adaptive Routing Algorithm for VANETs
Trans Tech Publications | 2012
|