Research On DASH Based QoE Optimization Strategy For Mobile Video Streaming

Dynamic Adaptive Streaming over HTTP(DASH)could select the streaming video bit-rate according to the network environment and cache level of the user terminal.However,the unreasonable using of device's remaining power and user mobility mode seriously could affect the user's Quality of Experience(QoE).Therefore,we deeply study the impact of device's remaining power and user mobility mode on mobile streaming user's QoE.Therefore,it is extremely important to propose the optimization strategy of mobile streaming user s QoE.In order to improve the user's experience in the mobile DASH application,we study the relationship between remaining power and user mobile mode and the user's QoE,and design the optimization strategy based on the user QoE feedback,thereby improving the mobile streaming video user's QoE.The main research contents of this thesis are as follows:(1)We propose the User Battery and Mobility QoE Model called UBMQM,based on remaining battery power and device mobility.We add remaining power and device mobility into traditional QoE model parameters,including initial delay,pause event,switching event and video perceived quality,to jointly evaluate the user QoE in the case of bandwidth fluctuations,for guiding the streaming rate adaptive strategy to optimise.(2)We propose the Battery-Mobility Bitrate Adaptive Scheme called BMBAS,based on device's remaining battery,device mobility and network bandwidth.In order to realize the trade-off between remaining power and video quality during video playback and reduce the video playback and switching events in different user's mobile modes,We propose the BMBAS bit-rate adaptive strategy combined the battery-aware strategy and the mobility-aware strategy to provide users with a optimal video bit rate and improve user QoE.We build up a real application system and test the QoE optimization of DASH performance in terms of device's remaining power and mobility.The experimental results show that the proposed rate control algorithm shows better adaptability by acquiring the remained power of the device and predicting the user's scene.Especially when device's remained power is in low status and the user walks,the user experience is significantly improved.Therefore,the algorithm could better adapt to different power modes and mobile scenarios,and improve user QoE.