Research On Vehicular Network Caching Strategy Based On ICN

Nowadays,the vehicular network has changed from the traditional security applications for assisting traffic management to supporting diversified,sea-quantized entertainment information service applications.Because the Information Center Network can provide effective support for large-scale Internet of Vehicles services and massive content,the integration of ICN and Internet of Vehicles has become a hot spot in the research of the vehicular network,which is called V-ICN for short in this thesis.The V-ICN caching strategy is deeply studied aimed at entertainment information service applications in this thesis.For the problem of content placement and content acquisition in the V-ICN,the cache discovery,pre-caching and cache update strategies are designed separately.The main work of this thesis is as follows:(1)A cache discovery strategy is designed.Through the analysis and maintenance of the geographical location information and historical message information,the problem that the vehicle users outside the RSU coverage cannot effectively obtain the content is solved.This strategy is also the foundation of the pre-caching and cache update strategy.(2)A RSU pre-caching strategy is designed.The Simpson time estimation method is used to estimate the request delay of the content.The optimal placement model and the content acquisition problem in the RSU cable network are considered to solve the problem.On this basis,the bus route and the user's preference for the content on the route are analyzed.And the pre-caching strategy of the bus is designed to maximize the one-hop completion probability of the content request.(3)Based on the Policy Gradient algorithm in reinforcement learning,the cache decision algorithm PGCD and the coordinated cache update strategy C-PGCU are designed to solve the problem of cache decision and cache replacement after receiving the Data message.(4)Build a V-ICN simulation platform composed of traffic modules and network modules for system simulation.After verification on the built V-ICN simulation platform,the strategy designed in this thesis improve the success rate of content and the average request delay of content.