Research On Historical Mobility Trajectory Routing Scheme In Urban VANETs

Communication between vehicles is an important and emerging research hotspot in the field of vehicle technology. In US, EU and Japan, governments have been carrying out the research of Intelligent Transportation Systems(ITS), which is to create a safe environment without road traffic accidents. Vehicular Ad Hoc Networks(VANETs) is one of important part in ITS. In order to make the communication between vehicles can be run effectively in mobile network environment, which is fast moving speed of the vehicle, the dynamic changes of network topology and vehicles node distributing unevenly in the network, Efficient VANETs routing protocols will play a crucial role.Since data transmission in VANETs is intermittent, its communication is used be similar to the form of DTN about "Storage-carry-forwards". If no suitable forwarding node within communication range of a node, it saves the message in the cache until it is encountered forwarding node forwards the information next time. Many scholars have done a lot of research in the field of vehicle network routing protocols for this new data forwarding mode, and they proposed a number of different type of routing protocols. Influence of network topology, obstacles in urban roads and other uncertain things, there are still a great challenge for existing routing protocols in metropolitan road environment. Hence, this paper proposes Historical Mobility Trajectory Routing Algorithm Base on Urban VANETs(HMTR). The related research work about HMTR in this paper as follows.First, the paper introduces the concept, application and challenges of VANETs, and describes and summarizes the existing routing protocols in VANETs, and points out respective advantages and problems in these protocols. Second, considering about the defects of the current mobile model in the large-scale urban network, the paper proposes Vehicle Mobile Model of Historical Mobility Trajectory(HMT). The urban traffic state map of this mobile model is drawed by the real vehicel trajectory data, which is compose of a plurality of square units. Then the model applies the k-means clustering algorithm to thes square units, and proposes global vehicle mobility pattern and single vehicle mobility pattern, which are uesd to generalized the characteristics and laws of metropolitan vehicle daily activities. Third, this paper uses two vehicle mobility patterns to propose HMTR routing algorithm, which is applicable for the large-scale urban VANETs. This new algorithm uses the contact of vehicles and zones and the traveling direction of relay vehicle to enhance the forwarding rate and reduce vehicle data forwarding delay. At last, we use ONE simulation platform to compare GeoEpidemic, GeoPRoPHET and GeoMobCon with HMTR. The result of simulation show that, HMTR has improved for four performance, which are include of forwarding rate, transport cost, average latency and average number of hops. Generally speaking, HMTR has better data forwarding capability, lower overhead and transmission delay to the network, and a good stability of the network.