Content Cooperative Caching Strategy For Internet Of Vehicles

With the coming of the era of big data,data traffic has taken an explosive growth trend,thus wireless communication networks are facing unprecedented opportunities and challenges.As an important component of intelligent transportation systems,internet of vehicles(Io V)mainly relies on low-latency and highly-reliable content request services to bring safety,comfort and convenience to users.Edge caching could accelerate the response process of content requests to a certain extent.However,the edge nodes in Io V,like vehicles and roadside units,do not have enough caching capacity to cache all contents.Therefore,the efficient content caching strategy design will play a major role in the implementation of Io V.Firstly,the research background and key technologies of Io V are introduced.In addition,the research background and significance of content caching in Io V are summarized;on this basis,the principles,characteristics and classifications of content centric network and machine learning are summarized,and the advantages and applications of content centric networks and machine learning in vehicular caching are analyzed.Secondly,a cooperative caching strategy with content request prediction is proposed in Io V systems,in which vehicles are clustered using K-means to reduce the impact of vehicles’ mobility on the communication link and simplify the process of content requesting and distribution;the number of content requests is predicted adopting the long-short term memory network to obtain more accurate content popularity;based on above,a Q-learning-based cooperative caching algorithm in Io V is designed to acquire final content caching decision.Thirdly,a vehicular trajectory and request aware cooperative caching strategy is proposed in Io V systems,in which K-order Markov chain is used to predict vehicles’ trajectories;considering the difference of content popularity in the coverage areas of RSU,each roadside unit collects and summarizes the historical content request information of the set of vehicles in next time period,and uses long-short term memory network to predict the content popularity in their respective coverage areas;based on this,a cooperative caching algorithm for Io V based on deep Q network is designed to obtain final content caching decision.Finally,the main work and innovation points of this thesis are summarized,and the future research work is prospected.