An Optimization Method To Improve Qoe In Http Streaming

As a new round information age arrived, mobile internet has been more and more important in people’s daily lives. Now the3G and4G technology has been developed rapidly as well as all kinds of intelligent terminals in China which brings a big opportunity for mobile streaming media. In HTTP Streaming servers provide different bitrate versions of the same video for clients to choose when network condition change which make it more popular than HTTP progressive download and RTP/RTSP among content providers. Differed from traditional network, mobile network suffers a lot from disturbance which may lead to a bad impact on users’QoE when they are enjoying their videos. In the past most researches focus on optimize the QoE in the network layer or the application layer which ignore the effect of users’behavior. This may cause inefficient to those solutions when individual use’s behavior differs a lot from others’. How to optimize the QoE for users involved with their behaviors is the main purpose of this paper.In this paper we first analysis the HTTP Streaming including the major mature technology nowadays and research status, and summarize some basic rules of the user’s behavior,specially jump forward, in their process of watching streaming videos. Then we put forward a segment cache mode with the support of user’s behavior to improve the segments storage and add bandwidth prediction to choose the right version of the segments when jump forward occurs. At last we deploy a HTTP Live Streaming system and develop the client application with the cache mode on an Android phone which proves the efficiency of the optimization.