QoE驱动的基于CDN缓存命中率的码率自适应算法

QoE-driven code rate adaptive algorithm based on CDN cache hit rate

随着移动流媒体用户群的增长,内容分发网(CDN)已经成为提高网络容量和用户体验质量(QoE)的重要途径。为了研究如何提升移动流媒体用户的QoE,提出基于CDN节点缓存命中率的QoE模型,该模型参数包含CDN节点缓存命中率、带宽吞吐量和客户端缓存变化量。在该模型的基础上,提出基于CDN节点缓存命中率的码率自适应算法,解决在网络条件不佳的情况下,根据用户码率请求参数的分布情况调整缓存策略,以提高CDN缓存命中率,降低时延,提高用户QoE。通过实验结果分析可以得出在网络条件相对较差的情况下,基于CDN缓存命中率的码率自适应算法能明显降低用户的RTT,提高用户QoE。

With the growth of mobile streaming media user base,content delivery network(CDN)has become an important way to improve network capacity and quality of experience(QoE).In order to improve the QoE of mobile streaming media users,a QoE model based on CDN node cache hit rate is proposed.The model parameters include CDN node cache hit rate,bandwidth throughput and client cache change.Based on the model,a bit rate adaptive algorithm based on CDN node cache hit rate is proposed,which can adjust the cache strategy according to the distribution of user bit rate request parameters under poor network conditions.This improves cache hit rate,reduces latency and improves user QoE.Through analysis of the experimental results,it can be concluded that under relatively poor network conditions,the code rate adaptive algorithm based on CDN cache hit rate can significantly reduce the user's RTT and improve the user's QoE.