Research On Route Selection Problem Of Patrol Vehicles Based On Social Analysis Of VANET

The two major tasks of the patrol vehicles of communication are patrolling and repairing the communication optical cable,and these tasks are the basic guarantee for realizing high-speed,stable and uninterrupted communication.The maintenance tasks of these vehicles in urban areas are mainly affected by problems such as poor real-time performance,low efficiency and limited communication conditions.The movement of vehicle nodes is not random but implies relationships of drivers.Conducting social analysis of vehicle nodes can discover the implicit movement relationships and provide optimal route selection scheme for communication patrol vehicles in real-time road traffic conditions which can be obtained by communication among vehicles through VANET routing technology,and provide an application-oriented idea for solving the problems in patrol vehicle tasks.In this thesis,the historical trajectory data of vehicle nodes is obtained,the historical encounter situation among nodes and the mobile preference of regions are analyzed,and the social attributes of vehicle nodes are defined.Based on the social attributes of the nodes,the three kinds of unicast routing protocol of VANET are improved.And based on the real-time road traffic conditions,the path selection of patrol vehicles can be conducted and the feasibility of these improved methods can be verified.The details of these work are listed as following..Firstly,according to the historical movement trajectory of the node,the encounter judgment model is designed,and the historical encounter probability of each node is calculated.The node moving area is meshed,and the movement preference of the node is defined by analyzing the number of movements of the node to each grid.The historical encounter probabilities,recent encounter situations and movement preferences are regarded as the social attributes of nodes.Secondly,according to the social attributes of each node,the topology based,geo-based and clustering based unicast routing protocols of VANET are improved.And SAODV,SGPSR and SCBR are presented.The performance of three improved protocols is verified by simulation and the SAODV protocol is chosen for communication.Then,based on the historical movement trajectory of the node,the road traffic condition can be predicted and used for designing the improved Dijkstra path selection algorithm based on traffic weight.Compared to the Dijkstra algorithm with well efficiency and the Genetic algorithm with well accuracy,the simulation results of the improved traffic-based Dijkstra algorithm show that the improved algorithm has a great improvement in efficiency and timeliness.Finally,the communication maintenance prototype system of patrol vehicles isdesigned.With the improved methods as the cores of each module in the system,work flow scheme can be proposed,and the feasibility of the improved methods are verified,resulting in providing ideas of solving the route selection problem of communication patrol vehicles.