Algorithm Research And System Implementation Of DASH-based Streaming Media Transmission In Wireless Environment

The rapid growth of the demand for audio and video services has promoted the continuous evolution of the streaming media technology.Due to the inherent randomness of changes in wireless network throughput,the traditional streaming media technology are difficult to meet the business needs of the wireless video users.With its good compatibility and flexibility,the streaming media technology based on MPEG-DASH has attracted wide attention in the industry since its birth.DASH client can download various bitrate media segments adaptively from HTTP server,thus providing reliable guarantee for efficient transmission and smooth playback of video.This thesis first introduced the development of streaming media transmission technology and the main content of MPEG-DASH standard.Secondly,concerning the fact that existing rate adaptation algorithm based on MPEG-DASH have frequent bitrate switching and low average bitrate in wireless network,a rate smooth switching algorithm based on joint control of bandwidth and buffer was proposed.The algorithm was mainly composed of two parts: bandwidth detection mechanism and rate decision mechanism.The bandwidth detection mechanism used the sliding window to capture the download information of the historical segments in the most recent period to detect the bandwidth fluctuation,and calculate the bandwidth prediction value under different fluctuation.The rate decision mechanism smoothed the available bandwidth by considering the bandwidth fluctuation and the buffer state,and then adopted corresponding adjustment strategy to select the segment bitrate.This thesis also determined the QoE evaluation index and statistical function of the rate adaptive algorithm based on the QoE influencing factors of streaming users.By simulating the performance of the algorithm in a single-client exclusive link scenario and a multi-client shared link scenario,the experimental results show that compared with the fuzzy-based DASH rate adaptation algorithm and the modulated throughput driven rate adaptation algorithm,the proposed algorithm can not only improve the average bitrate in playback but also reduce the number of bitrate jitters in the case of sharp fluctuations in network throughput and multi-client contention for bandwidth.In order to verify the actual effect of the proposed algorithm,an adaptive streaming transmission system based on MPEG-DASH was also designed and built by using theserver/client structure in this thesis.The server used Apache to distribute and store video segment and related configuration files,and the DASH client realized functions such as network monitoring,bitrate selecting,user interface interaction and video playback by using Android application.Through integrating the proposed algorithm into the client,the adaptation effect of the algorithm was tested with the traffic shaping tool.The tested results are basically consistent with the simulation results,which demonstrate that the rate smoothing switching algorithm can provide better experience quality for wireless streaming media users.