基于HTTP的流媒体自适应码率切换算法

基于HTTP的动态流媒体自适应(DASH)系统已经广泛地应用于互联网上的视频传输.针对客户端缓冲区受限和网络环境不稳定的情况,提出了一种新的基于缓冲区的码率控制算法.该算法将缓冲区分成不同等级,并采用滑动窗技术对视频的下载时间进行平滑处理,使缓存尽最大可能地维持在均衡区内.该算法在向上切换码率时采用保守策略,而在向下切换时则利用逻辑斯谛方程来决定码率切换的幅度.此外,该算法采用快速启动策略,能使系统开始时下载码率迅速匹配网络带宽.实验结果表明,算法能在复杂的网络环境中,充分利用带宽和缓冲区,降低码率的切换次数.

Adaptive rate switch algorithm based on HTTP streaming

Dynamic adaptive streaming over HTTP (DASH) has been widely used in the video transmission.With the limited client buffer and varying network,this paper proposes a new adaptive bitrate switching algorithm based on the client buffer.The proposed algorithm divides the client buffer into several buffer levels and uses sliding window to analyze the download time of recent video segments in order to maintain the media buffer size in the balance interval.The algorithm uses a conservative strategy when switching to bit rates upwards,and when switching down,the logistic equation is used to determine the magnitude of the bit rate switching.In addition,this paper also proposed a fast start strategy,in which download rate could match the network bandwidth quickly after the system starts.Experiment results show that the proposed algorithm can make full use of bandwidth and buffer,and reduce the number of rate switching in complex network environment.