基于混合高斯模型的移动ICN缓存机制研究

Caching Mechanism of Mobile ICN based on Hybrid Gaussian Model

针对移动互联网环境下信息中心网络的缓存放置问题,提出一种基于混合高斯模型的缓存优化机制(Gaussian Mixture Cache,GMC)。该策略通过网内的核心感知节点,周期性请求其余路由节点信息,依据混合高斯模型生成节点边缘度,最后将节点边缘度分发至路由节点。基于边缘优先的原则,路由节点考查数据内容的流行度和自身的边缘度得到细粒度的缓存概率,以实现节点移动条件下的高效缓存。仿真结果表明,GMC机制能够在移动ICN环境下有效发挥缓存的作用,降低了平均路由跳数,减少了数据请求时延,提高了缓存命中率。

Aiming at the cache placement problem of the information center network in the mobile Internet environment,a cache optimization mechanism GMC(Gaussian Mixture Cache)based on a hybrid Gaussian model is proposed.This strategy,through the core sensing nodes in the network,periodically requests the information of other routing nodes,generates the node edge degree according to the mixture Gaussian model,and finally distributes the node edge degree to the routing nodes.Based on the principle of edge priority,the routing node examines the popularity of data content and its own edge degree to obtain a fine-grained cache probability and realize effective caching under the condition of node moving.Simulation results indicate that the GMC mechanism can effectively play the role of cache in the mobile ICN environment,reduce the average number of routing hops,decrease the data request delay,and improve the cache hit rate.