Caching Strategy Based On K Shell Value In Content Centric Networking

In order to cope with the current Internet challenges, one of the popular ideas is to reconstruct an Information Centric Networking (ICN), and the Content Centric Networking (CCN) is one of them. Each routing node in CCN can cache content, the whole networking caching system can not only reduce the redundant traffic in the network, but also improve the responding speed of the request. Therefore, optimizing the performance of the caching system is of great significance to CCN. Because of the new characteristics of CCN caching system, which has the characteristics of transparency, universality and fine granularity, optimizating the algorithm of CCN caching system becomes a new challenge. In this thesis, the performance of the caching system is optimized from two aspects:the cache decision strategy and the cache space allocation strategy. The main work of this thesis is as follows:Firstly, a cache decision strategy based on K shell value is proposed. At present, the main idea of CCN cache decision strategy is to select the most influential node in the path, than cache the content on this node. It will not only reduce the cache redundancy, but also imiprove the diversity of the content in cache. Therefor measuring the node information dissemination force becomes the key point of the algorithm. The popular betweenness based index has high computational complexity and node has an inaccurate recognition to the influence of information dissemination. Considering this, the author introduced the K shell decomposition method into CCN cache study, and defined the K shell value of the node based on the K shell decomposition method in this thesis. K shell value of the node not only has low computational complexity, but also can identify the node more accurately which has the greatest influence of information dissemination. For the case that there are more than one nodes have the maximum K shell value, the author propose the decision strategy based on remaining cache space of the node and hops the node to the content requester. The simulation results show that the proposed algorithm can not only converge faster, but also get higher cache hit ratio and faster request response speed.Secondly, a cache space allocation strategy based on betweenness and K shell value is proposed. The simulation result shows that the cache hit ratio of different cache space allocation strategy is different when the cache decision strategy is the same. From the point of view that the node importance is related to the traffic characteristics of node and the node’s topologic connectivity in the network, the author defined the BK value of the node, and proposed a cache space allocation strategy based on BK value. Through simulation experiments, the author determined the weightings of betweenness and K shell value in the BK value formula. And the author also got the result that the cache hit ratio is further improved when the cache decision strategy based on K shell value combined with the cache space allocation strategy based on BK value.