Research On Collaborative Caching Policy Based On Node Centrality And Content Popularity In NDN

Ubiquitous network cache of NDN contributes to enhancing the efficiency of net-work data distribution and avoiding repeatly sending data packets.The default caching decision policy LCE(Leave Copy Everywhere)will introduce high redundancy in the network,which causes low usage rate of the cache.Besides,the default routing mech-anism does not utilize caching informations of other nodes,caching on each node is essentially non-collaborative.For the above problems,this thesis optimizes the caching decision policy and ob-ject availability in Named Data Networking,and particularly focusing on raising cache hit ratio,improving the availability of cached content and reducing server loads.The main work includes the following:(1)The default caching decision policy introduces high redundancy,and the Be-tweenness policy causes high replacement rate.Aiming at the above problems,this the-sis proposes a novel caching policy based on betweenness and edge popularity(BEP),which caches the most popular contents in the most important nodes.BEP raises the cache resource utilization and improve caching system performance.(2)The default caching decision policy does not make full use of cache informa-tions of the neighbor nodes,and content popularity is nottaken into account.To deal with these problems,this thesis proposes a cooperative caching policy based on neighbor content popularity(NCP),which make use of the information of neighbor caches.NCP further improves caching system performance.The simulation results show that BEP and NCP can obtain higher hit ratio,and lower server load compared with some other policies.The main innovations of this the-sis includes the following:(1)The concept of Edge Popularity is used according to the characteristics of the low-pass filter with the node;(2)Building up horizontal caching community using neighborhood cache informations.(3)Nodes will advice their neigh-bors to cache contents elimated by themselves based on content popularity,which ends up elimating contents with lower popularity,raising the cache resource utilization.