Linear Cooperative Detection for Alarm Messages in Cluster-Based Vehicular Ad Hoc Networks

Abstract

Vehicular ad hoc network (VANET) is a special class of mobile ad hoc Networks (MANET), and is an important component of Intelligent Transportation System (ITS). VANET has a future potential in terms of a rich set of applications that it can provide to its customer. Accident detection is a key functionality of ITS. Due to the effects of channel fading and shadowing, individual vehicles may not be able to reliably detect the existence of an accident. In this paper, we propose a linear cooperation framework for accident detection in a cluster-based VANET. In this framework, accident detection is based on the linear combination of local statistics gathered by individual vehicles. Our objectives are minimizing the probability of false alarm and avoiding broadcast storm. Based on the problem formulation and derivation, we proposed two efficient algorithms to do alarm detection: local detection and global detection. The numerical results show that our heuristic strategies perform well on false alarm detection.