An Intersection-based QoS Routing in Vehicular Ad Hoc Networks

International audienceVehicular Ad hoc Networks (VANET) face many challenging issues because of the rapid topology changes, frequent network partitions, and complicated urban environments. These special VANET characteristics seriously deteriorate the routing performance. In this paper, we propose a new vehicular routing protocol called Intersection based Routing with QoS support in VANET (IRQV). Based on Ant Colony Optimization (ACO), IRQV mainly consists of three processes, 1) terminal intersection selection process, 2) network exploration process, which is implemented using the QoS of local road segment and global end-to-end intersections, and 3) optimal routing path selection process. When initiating data packets forwarding session, IRQV dynamically chooses the optimal next intersection using latest routing information and adopts a simple greedy carry-and-forward mechanism to forward data between two adjacent intersections. In addition, we propose connectivity and transmission delay mathematical models for a 2-lane road segment scenario to estimate local QoS. Simulation results validate our derived road segment connectivity and transmission delay models, and show the effectiveness of IRQV compared with two geographical routing protocols in terms of delay, packet delivery ratio and overhead