Research On Caching Strategy Of Vehicle Network Based On Heuristic Algorithm

With the development of intelligent transportation system,mobile vehicles will need not only map information,but also more entertainment contents,but the capacity of the entire network is limited,which brings great pressure to the vehicular ad hoc network(VANET).Edge caching is the deployment of caches on the edge of the system,introducing edge caching into VANET can effectively relieve the pressure of the network,and with the diversity of vehicle user requests,how to properly cache and transfer content has become a hot research topic.This thesis studies content caching and transmission in VANET,and corresponding collaborative caching strategies are proposed based on vehicle clustering for the scenes of highways and urban roads in vehicle networks,which can effectively improve the success rate of vehicle users’ requests and reduce the delay of users’ content acquisition.The main research contents and innovations of this thesis are as follows:(1)A collaborative edge cache strategy based on vehicle clustering is proposed for highway scenarios in vehicle network.Firstly,in order to improve the stability of communication links between vehicles,a weight-based clustering method is used to cluster moving vehicles.Content can be cached in moving vehicles and roadside units,and vehicles can obtain required contents from cluster heads and roadside units.Different content acquisition methods will bring different request delays and costs.Then,the delay and cost of vehicle content acquisition are analyzed.Under the constraint of cache capacity,an optimization problem of joint minimization of delay and cost is constructed,and the ant colony algorithm is used to solve the problem to obtain the optimal cache placement strategy.Simulation results show that the proposed caching strategy reduces the delay and cost of the requested content.(2)A cooperative cache strategy using parked vehicles is proposed for urban road scene in vehicle network.Firstly,appropriate mobile cache nodes are selected according to the social similarity and centrality between moving vehicles to improve the cache efficiency.Then LSTM neural network is used to predict the content prevalence,and contents are cached cooperatively in mobile cache nodes,roadside units and parked vehicles.Finally,a cache hit ratio maximization optimization problem is constructed and solved by genetic algorithm.Simulation results show that the proposed cache strategy effectively improves the cache hit ratio and reduces the content acquisition delay.(3)Considering the problem of unstable communication link between base station and vehicles in urban scene due to occlusion of buildings,a UAV-assisted vehicle network cooperative cache strategy is proposed.Firstly,the moving vehicles are clustered,and the optimal hover position of UAV in the vehicle cluster is found according to the activity of vehicles in the cluster,so as to reduce the content request delay of vehicles in the cluster.Then the flight path and hover time of UAV are optimized according to different activity of vehicle clusters to improve the success rate of users’ requests of contents.Then the global prevalence of the contents are obtained according to the activity of the cluster and the prevalence of the contents within the cluster.Finally,the optimal cache placement scheme is obtained by particle swarm optimization algorithm.Simulation results show that the proposed caching strategy improves the request success rate of vehicle users and reduces the delay.