Data Transmission Strategy Research Based On Vehicle AD HOC Network

With the development of society and economy, the vehicle has gradually become one of the most important transport tools for people’s daily life. On one hand, the increase of vehicle ownership is convenient for people’s daily life and accelerated the development of social economy, on the other hand, it triggered by problems such as traffic congestion, pollution and traffic accidents are caused great trouble to people. To solve these problems, vehicle ad hoc network came into being. Vehicle ad hoc network is a special multi hop wireless ad hoc networks designed for vehicles. It consists of a mobile vehicle using wireless communication technology.Vehicular ad hoc network not only inherits the basic characteristics of ordinary ad hoc network including multiple hops and no center, but also has its own characteristics of network topology quick changes, predictable vehicle trajectory and owning global navigation and high-performance on-board computer, etc. These characteristics determine that data transfer mode of vehicular ad hoc network is difficult to learn from the original solution of mobile ad hoc network.The paper mainly studied vehicular ad hoc network’s data transfer mode. The existing data transfer mode of vehicular ad hoc network has the following problems: 1. Data transfer efficiency of wireless networks and utilization rate of wireless resources are a pair of negative correlation parameters. The data transfer efficiency is difficult to balance between speed and resource. 2. In the data transfer, selection strategy of the next hop node is the key to affect the overall network. Now most of the transfer strategies simply choose the next hop node from the position relation, and do not use of other important parameters of on-board network to improve the strategies.Through reading the related literatures and experimental research, this paper was divided into the following two parts according to the above two typical data transfer problems:Firstly, we improved the algorithm on the basis of GPSR in order to balance between speed and resource. The algorithm filtrated data based on Naive Bayes model before data transferred, gave preference to transmitted the more important information wirelessly, and adopted vehicle carries transmission to decrease the wireless resource utilization. When transmission met local optimization, the algorithm also adopted vehicle carries transmission to decrease the wireless resource utilization if a car’s driving direction was good for transmission, and it transferred data through surrounding transmitting data if not. Finally, through comparing the method proposed in this paper with several classical algorithms in the simulation experiment we proved the validity of the method.Secondly, this paper chose the next hop node according to the location of the relationship between vehicle and its historical movement track comprehensively. We predicted vehicle future track according to historical track in the same road, thus we find a vehicle as the next hop node so that we can get to the destination eventually. Finally, we proved that our method can improve the performance of the whole network based on taxi driving data simulation in Shanghai.