Research On The Caching Strategy For Content-Centric Networking

The current TCP/IP network architecture employs host-oriented communication model,which was designed for transmitting special data in the early days of Internet.However,with the use and popularity of the Internet,application requirements have gradually turned into content distribution and acquisition.The conflict between the network architecture and application requirements has brought many problems in the current Internet.To solve these problems,the Content Centric Networking(CCN)is proposed.As a key component of CCN,the ubiquitous caching nodes can store hot contents flowing through them,thereby effectively reducing content retrieval time.Therefore,it is necessary to design an efficient caching policy in CCN.The thesis mainly investigates two important metrics:profit and energy efficiency(EE)in the CCN caching policies,and proposes a cooperative caching policy based on the provider profit and a caching policy for videos based on the energy efficiency,respectively.The main research of this thesis are summarized as follows:1.Based on the two-level CCN,a cooperative caching policy is proposed by maximizing the provider profit.Since this optimization problem is NP-hard problem,inspired by the twofold benchmark cases,this thesis proposes a heuristic cooperative caching policy to solve this problem.In this policy,the cache space of control node is divided into a duplicate fraction and unique fraction.In the former fraction,nodes cache duplicated contents,which are the most popular contents in the content library,and in the later fraction only unique contents are allowed to be cached.A replacement algorithm of control node is also proposed to support the caching policy.The simulation results express that the proposed caching policy is superior to other traditional strategies in terms of earned profit.2.Based on the Cognitive-Content Centric Networking(C-CCN),a caching policy is proposed by maximizing the EE.Since users have various needs for different types of video,a scalable video coding(SVC)-based probabilistic caching policy is proposed,which employs SVC technology to divide the video files into two layers.According to the proposed caching policy,the expressions of the successful content delivery rate(SCDR)are derived.Based on the derivations and the established sum of energy consumption,the optimal EE problem can be formulated,which is efficiently sloved by a standard gradient projection algorithm.The numerical results validate theoretical analysis,and express that the proposed caching policy is superior to other traditional strategies in terms of EE.This thesis deeply researches the key component of cache in CCN,and through the analysis of problems in existing researches,proposes a cooperative caching policy based on the provider profit and a caching policy for videos based on the energy efficiency.The simulation results express the superiority of the proposed two caching strategies in terms of profit and EE,respectively.