Research On Topological Structure Characteristics Of VANETs

With the development of wireless communication technology and network technology,VANETs has become a hot spot in the field of intelligent transportation system. As a special mobile ad-hoc network, dynamic network topology is one of the important characteristics of VANETs, accurately describing the variation trend of dynamic network topology is of great significance in development, design, optimization of network protocol and network management and so on for VANETs.Based on Intelligent Driver Model with Lane Changes(IDM-LC), this paper apply VanetMobiSim to set up VANETs and study the statistical characteristics of dynamic network topology for VANET in detail. The main work includes the following four aspects:(1) Studying the instantaneous topological characteristics of VANETs. It shows that the degree distribution of VANETs is not power-law by analyzing degree distribution characteristics of instantaneous topology, that is to say, VANETs are not scale-free networks; the distribution of degree correlation presents that there is a positive correlation in VANETs, namely VANETs are assortative networks; the analysis of clustering-degree correlation shows that VANETs have no hierarchy.(2) Analyzing the evolutionary characteristics of VANETs topology over time based on complex network theory. The experimental results show that topological structure of VANETs has a strong correlation with time and parameters such as the average degree, clustering coefficient and global efficiency are all long range dependence. Therefore, the common short range dependence can't depict the statistical distribution characteristics.(3) Studying the spatio-temporal characteristics of connectivity for VANETs.The simulation results show that the connectivity is poor since the network is segmented seriously when the communication radius is small. With the radius increasing, the connectivity of VANETs is efficiently improved. And statistical characteristics of some connectivity measurement parameters including the number of isolated nodes and connected components as well as the size of the largest connected component are studied, which present they are also long range dependence. Finally, the influence of the communication radius and node density on connectivity of VANETs are researched.(4) Studying empirically the evolution characteristics of topology for VANETs. The real topological evolution of VANETs with time is analyzed based on the VANETs made up by taxis in Shanghai. The results show that the parameters measuring network topology are all of long range dependence and have a very strong correlation with time. Increasing communication radius can effectively enhance network connectivity, and when the communication radius reaches a certain threshold, the size of the largest connected component remain relatively stable, which is helpful to the transmission of data packets over the VANETs.