Research Of Vehicular Ad-hoc Network Based On Complex Network Theory

With the rapid development of urbanization process, it makes the traffic condition more complex in the city. Vehicular Ad-hoc Network (VANET), as the basic element for the application of Intelligent Traffic System (ITS), will be the key ways to relieve traffic jam, prevent traffic accident, improve traffic safety, optimize the traffic flow condition and realize autonomous. VANET is a special subset of multi-hop Mobile Ad hoc Networks in which vehicles can not only communicate with each other but also with the fixed equipments along the roads through wireless interfaces, which is more flexible in deployment and has a better development outlook. The prior knowledge for traditional VANET, such as communication protocol and its topology, didn’t reveal the essence from the macroscopic angle. From the perspective of complex network theory, VANET model, topological structure and control scheme are proposed, which is a fresh perspective overcoming the deficiency of the assessment from a single perspective.Firstly, VANET system architecture and the model algorithm are introduced. The paper focuses on the macroscopic Vehicle-to-Vehicle, Vehicle-to-Infrastructure and communication network using complex network theory, and then the instantaneous degree distribution and stability degree distribution are analyzed using mean-field theory to determine exponent range. According to the assumptions, module model with life cycle is also proposed considering that the actual VANET cannot grow infinitely. Secondly, special circumstances are discussed in particular. The algorithm calculates degree distribution and recognizes the relationship, while degree distribution obeys the power-law to verify the Scale-free property and Gini coefficient to verify the network heterogeneity. What’s more, the paper carries out extensive simulations of information propagation under normal circumstances and capacity constraints. Finally, we not only derive the network state equation of controlling to equilibrium point according to network state equation from the perspective of improving the existing network performance, efficiency and stability, but also calculate the maximum eigenvalue of matrix and network eigenvalue, as well as propose VANET pinning control strategy. The paper discusses how the topological structure parameters affect network synchronization dynamics, then describes the basic concepts of network controllability, and analyzes the relationship between degree correlation and the controllability. Moreover, the paper analyzes different pinning control methods of controllability; accordingly, we propose specific pining control, random pinning control and mixing control methods of complex network to implement our strategy by numerical example and simulation.Simulation results show that VANET degree distribution exponent is2<γ≤3and mathematically prove the power law property and Scale-Free network features. Those larger degree nodes have a larger effect on the network, which information flow and nodes degree obey the power law distribution, while the degree distribution doesn’t obey the power law when considering nodes life cycle. The smaller the degree distribution exponent is, the more difficult synchronization is to achieve; the synchronization of preferential attachment network is stronger than rewired attachment network. Dynamic network shows stronger robustness under random attack. Contrast with different control strategies, specific pinning control strategy is the highest efficiency. All the results demonstrate that the increase ratio of specific pinning control nodes has little impact on the synchronization for the same network, that is, function on the whole network becomes more and more small; in vehicular ad-hoc disassortative networks, higher degree nodes pinning controlled status can spread to the whole network more easily, and thus the network is easier to control.