Position-freshness based intermediate vehicle filtering layer with intersection detection for VANET

Routing in inter-vehicle communication is a challenging task owing to several intrinsic characteristics of vehicular networks, such as intermittent connectivity, dynamic topology, and inaccurate real-time location information. Routing accuracy in vehicular networks is directly proportional to the signaling overhead of the routing protocol employed. This signaling overhead is minimized by position based routing (PBR) protocols that make ondemand forwarding decisions based on the position information of neighboring vehicles. However, inaccurate position information caused by factors such as urban canyons, satellite coverage, intermittent connectivity, and error-prone channels, reduces the performance and reliability of PBR. In this paper, we propose a positionbased routing scheme that applies a filtering method over the forwarding or relaying candidate vehicles based on the angle between the vehicle's direction of motion and the line between the sender and neighboring vehicles. By filtering inappropriate forwarding vehicles, the proposed method improves communication reliability in urban environments. Simulation results show that the proposed filtering scheme improves the performance of PBR in terms of packet delivery ratio and end-to-end delay.