Research On Urban Scenario-Oriented Routing Protocols For Vehicular Ad Hoc Networks

Vehicular Ad Hoc Network(VANET), as the important application of wireless communication, is the support network for the intelligent transportation system(ITS). Routing protocols, one of the hot research topics in the VANET, are the rules and conventions of data distribution, as well as the basis of ensuring the efficient distribution of data, which have achieved many important outcomes.The routing void, known as the local optimum, is referred to the current node the one that is closest to the destination node in the forwarding process. Due to the uneven distribution of vehicles and the highly dynamic topology in the VANET, the communication links are short-lived and unstable. As a result, the routing void is a ubiquitous and challenging issue in the VANET. The existing strategies for the routing void are the perimeter forwarding,flooding broadcast, carry and forwarding, multipath routing, etc. In addition, the data packet must be forwarded along the path and can only change the route in the intersection owing to the tall buildings and other obstacles blocking in the urban environment. Hence, the anchor-based routing protocol is proposed, in which the data forwarding process is divided into the intersection-based model and the road segment-based model, and how to select the appropriate next intersection at the road intersection is an important problem.This paper, supported by the National Natural Science Foundation Project–The Technology Research Oriented to Real Scenarios and Quality of Experience in the VANET, proposed two anchor-based routing protocols for the routing void issue in the VANET. The first is the STSR, which is based on the consideration of the distance factor, combining the traffic statistics data with speed limits weights for each intersection. The intersection with the maximal weight will be selected as the next intersection. In STSR, aiming at the features of the complex urban communication environment, the greedy forwarding and the carry and forwarding are also improved. On the one hand, it reduces nodes' communication range and combines with nodes' position prediction, which improves the possibility of the successful transmission of data; on the other hand, for the carry and forwarding model, when the transmitting data suffer from the routing void, it will turn into the carry and forwarding model,and meantime, a timer will be started. When the timer stops and the current node has not found the next hop yet, the data will be transmitted back to the last intersection and reselect the next intersection. The second protocol named RTRCAR, is based on the real-time road connectivity awareness. By considering the features of the urban road segments of Xi'an, we conclude that the road segment is usually short in the urban, and a data packet can cross a section by only one or a few hops. Using these features, we design a method of real-time perception of the road connectivity, which is that letting the vehicle at the intersection send a connected probe packet to the neighbor intersection. Based on the connectivity collected by connected probe packet and the distance to the destination, the next intersection can be selected. Simulation results show that STSR can improve the overall performance of the network, compared to the GPSR and Gy TAR protocols.