Research About Cooperative Downloading Optimization Strategies Of Clustering Based VANET

VANET(Vehicular Ad-Hoc Network)is a special Ad Hoc network.It is a kind of network which is specially adapted to the characteristics and requirements of the vehicle.It is a kind of Internet of things used in the field of traffic.With the rapid development of the Internet and VANET technology,car users hope that even in the car can enjoy network services at anytime and anywhere.China's current VANET technology and infrastructure construction are still not perfect.In the highway scenario,the high-speed movement characteristics of the vehicle,the sparsity of AP/RSU distribution and the limitation of AP/RSU communication area make car users in the process of moving intermittently access the Internet.The VANET cooperative download method is recognized as an effective method to solve the problem of the car users intermittently access to Internet and so on in VANET.However,there is still a need to improve the performance of the cooperative downloading method in the packet loss rate of the target vehicle and the data throughput target during the transmission between target vehicle and cooperative vehicles.This dissertation presents a cluster-based VANET cooperative download optimization strategy based on the above problems.This dissertation mainly completes the following research work:(1)This dissertation establish a two directions VANET cooperative download model in the highway scenario.The core idea is that the target vehicle uses the same and opposite vehicles to forward data,to achieve real-time and rapid delivery of AP/RSU data and real-time acquisition of data required by vehicle users.(2)This dissertation proposes a single-link cooperative downloading strategy based on motion consistency clustering to improve the utilization of the communication blind area.First of all,the cluster is initialized according to the distribution state of the vehicle.Secondly,If the gap between the speed of a vehicle and the average speed of the cluster is large,the vehicle is deleted.And based on the weight value of vehicle speed and distribution state,we select the cluster heads and dynamically maintain the cluster.In the end,we choose cooperative vehicle cluster from the same direction and opposite direction vehicles to provide a single-link cooperative download based on dynamic time slot.Simulation results show that compared with similar cooperative methods,the average download bandwidth is increased by about 35%.(2)This dissertation presents a multi-vehicle cooperative download strategy based on Serial Interference Cancellation(SIC)algorithm,to further improve the data download during the assistance of vehicle clusters and target vehicles in the dark area.Firstly,according to the communication process of the cluster members in the cooperative vehicle cluster with the target vehicle,a transmission model for opposite direction cooperative downloading based on successive interference cancellation algorithm is established.Secondly,a constraint transmission area based on SIC algorithm and the largest set of serial interference are proposed to further reduce the model solution time,and the objective function is solved by maximum SIC feasible cluster member set algorithm.Finally,in the dark area,the cooperative downloading transmission scheme to complete the cooperative download transmission between the vehicle cluster and the target vehicle.Comparing with single-link cooperative downloading strategy based on motion consistency clustering,the average throughput increased by about 8%.