Early car collision prediction in VANET

Traffic safety is critical to our lives. In recent years, the increasing traffic accidents have brought considerable deaths. Reducing the rate of car accident has become one of the most important tasks in transportation field. VANET is a widely used platform contributing to possible solutions. In the past, there are protocols to generate or broadcast warning messages after vehicle collision. However, since most of them are made after car accidents have already occurred, they may help accident investigation but do not prevent the accident from taking place. In this paper, we proposed a method using support vector machine(SVM) [1] for early car accident detection in VANET. Once any dangerous situation is predicted, immediately the endangered driver gets a alert along with a suggestion to avoid danger.