Traffic density estimation protocol using vehicular networks

Traffic density estimates on maps not only assist drivers to decide routes that are time and fuel economical due to less congestions and but also help in preventing accidents that may occur due to the lack of not being able to see far ahead. In this paper, we propose a protocol that exploit vehicle-to-vehicle ad hoc communication for the estimation of vehicular density and the amount of congestion on roads. The protocol forms cluster heads by a voting algorithm. These cluster heads aggregate density information and spread it to the network via few selected forwarding vehicles. The protocol does not assume all vehicles to be equipped with Global Positioning System. We analytically study the cluster head formation part of the protocol and then simulate the proposed protocol using network simulator NS2 to understand different characteristics of the protocol. © Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2013.