The Research Of Cooperative Caching Strategy For Content-Centric Networking

With the increasing demand of application services,which makes the traditional TCP/IP network architecture has been exposed many problems in security,mobility,and so on.In order to deal with these problems,many researchers have proposed many new future network architecture,the Content-Centric Networking(CCN)is one of the most representative of the future network architecture.In the CCN,each node possesses the capability to store the data.In a limited cache space,through nodes collaboration to cache the valid content,which enables the users to obtain the required data more quickly.Take into account the question that the data transmission cause the link consumption and the cache space of on the path nodes is limited,a Popularity-based Neighbor Collaborative Caching strategy(PNCC)was proposed in the third chapter.Through mutual cooperation between neighbor nodes,according to the popularity of data in the local node and request number of the same data from the neighbor nodes,the cost difference of data transmission is computed.Based the cost difference,the return data is decided to be stored in the either maximum popularity node or cooperative caching of the neighbor node,in order to reduce the cost of data transmission and improve the utilization of cache space in the neighborhood.The simulation results show that the policy can improve the network's hit ratio,reduce the delay for users to fetch data and the hop of data transmission by placing the data into the appropriate node and effective utilization of cache resources in the neighbor node.If the node have many redundant contents and cache content of accessing with the downtrend,as well as the all cache contents is noticed,which makes the problem that the availability of store content and the accuracy of announce content is reduced,for this problem,a Neighborhood Cooperative caching strategy based on Cache Value(NCCV)was proposed in the fourth chapter.For improve validity of the cache content by forecasting model to predict and delete the cache content of accessing with the downtrend,according to the different cache content with cache value in the node is different and different interfaces in the node receive the same content with number of the accessing is different,through selecting the cache contents is noticed,which get the cache resources in the neighborhood and the local redundant content is reduced.Thesimulation results show that the strategy is effectively improved in terms of the network's hit ratio,user's access delay and data transmission hops.