Multilevel Cluster-Based Information Fusion in Vehicle Ad Hoc Networks

Abstract

Nowadays, the road congestion detection has playing a crucial role in vehicle ad hoc networks (VANETs), due to the increasing growth in the number of vehicles in urban roads. However, to gain more environment information, traditional road detection way may produce a lot of atomic messages. It will consume a lot of limited bandwidth resources and cost more channel collisions and conflicts. In this article, a multilevel cluster-based information fusion method through a low-level fuzzy clustering-based information fusion and a high-level modified BPA-based(basic probability assignment) evidence theory. The low-level information fusion can combine atomic messages and abstract the key attributes. While the high-level information fusion can avoid the misjudgment caused by the short-term wait events such as the traffic lights, and it also enhances the accuracy and reliability of the final decision making. This paper closes with the conclusions drawn from some simulation results that this scheme makes an accuracy decision in road congestion detection and tremendously reduces the network traffic load.