Connectivity Aware Routing in Vehicular Networks

Abstract

Multi-hop car to car communications are useful for supporting many vehicular applications that provide drivers with safety and convenience ranging from office on the wheel to real traffic query, vehicle safety, parking space searching and on-road advertisement. Developing multi-hop communication in vehicular ad hoc networks (VANET) is a challenging problem due to the rapidly changing topology and frequent network disconnections, which cause failure or inefficiency in traditional ad hoc routing protocols. The problem of frequent network disconnections can be partially addressed using a carry-and- forward mechanism which may incur higher delay. However, our connectivity aware routing (CAR) protocol addresses this problem by selecting an optimal route with the least probability of network disconnection and avoids carry-and-forward delay. This can be achieved using our new probabilistic model of network connectivity which takes into account a more realistic clustering phenomenon of vehicle traffic in city scenarios that is caused by traffic lights. Our simulation results show that the proposed CAR protocol outperforms existing VANET routing protocols in terms of data delivery ratio, data packet delay and network throughput. In addition, CAR improves performance for both sparse and dense networks.