基于内容流行度的移动CCN缓存策略研究

摘要

Recently,in order to adapt to the shift from the sender-driven end-to-end communication mode to the receiverdriven mass content acquisition mode for Internct applications,and provide native support for scalable and efficient content acquisition from the network architecture level,researchers have proposed a content-centric networking (CCN),which has great potential to better support content distribution in the future Intemet.With the increasing number of mobile users,future mobile users will also visit the CCN.For user mobility in CCN,a mobility caching strategy based content popularity is proposed.This strategy considers user mobility,uses semi Markov model to model the mobility of users,and also uses multiple linear regression theory to predict content popularity.It combines with the user mobility and the original content popularity to derive a mobility cache strategy.Simulation results show that the addition of mobility will have a greater impact on the popularity of the content,and this strategy can effectively improve the cache hit rate.%近年来,为了适应互联网应用由发送者驱动的端到端通信模式向接收者驱动的海量内容获取模式的转变,并从网络体系架构层面提供对可扩展和高效内容获取的原生支持,研究界近年来提出以内容为中心的新型网络(content-centric networking,CCN)体系架构,它能够高效地支持未来互联网上的内容分发.随着移动用户数量的不断增加,未来移动用户也将访问CCN网络,针对内容中心网络中的用户移动性问题,提出了一种基于内容流行度的移动性缓存策略.该策略考虑用户移动性,利用半马尔科夫模型对用户移动性进行建模,同时采用多元线性回归模型对内客流行度进行预测,结合用户移动性与原有内容流行度提出一种移动性缓存策略.仿真实验表明,移动性的加入会对内容的流行度产生较大的影响,该策略能有效地提升缓存命中率.