Research On User Traffice Optimazation Based On Mobile Edge Computing

With the rapid development of mobile communication technology,mobile terminals have become more and more popular in life,and people have thus obtained great convenience.The number of edge terminals has exploded,and the surge in user traffic has generated massive amounts of data in the mobile Internet,causing a huge load on the backbone network and backhaul links,bringing higher network latency to users.Class-based services,especially streaming media businesses,have raised higher demands.The mobile edge computing proposes to sink the business from the cloud computing network to the edge of the network,and provides services such as computing,storage,and information opening to provide technical support for effective optimization of user services.Since streaming media services already account for more than 75%of total network traffic,optimization for streaming media services will become an important part of user service optimization and an urgent need in the commercial field.In this paper,the typical business-video service in the user service is optimized.In the mobile edge network,the optimization of the adaptive code rate,caching strategy and wireless network information opening of the streaming media service is carried out.The main work of the thesis includes:The first part is the research on stream media rate adaptation based on mobile edge computing.Considering that mobile users share a common radio access link in a wireless network,when the network is congested,the user experience is degraded.A network-aware streaming adaptive selection strategy is proposed in the scenario of mobile edge computing.The strategy is based on the premise of optimizing the user to watch the video experience,and dynamically modifying the MPD file according to the situation of network blocking,so as to limit the range of video rate required by all users.The test found that the scheme can respond to network congestion in real time,and can increase the MOS value by one point in the case of blocking,thereby promoting the user to select a more appropriate video quality,which is a user-driven adaptive algorithm.The second part is the research of QoE-based streaming media caching strategy:The main research on the design and implementation of streaming media caching strategy for DASH format in mobile edge networks.Considering the storage capacity in the edge network,the streaming media characteristics in the DASH format,and considering the limitations of classical algorithms such as LFU and LRU,a more suitable caching scheme is needed in the edge network.In this paper,the caching strategy under the mobile edge network is used to ensure the user QoE as the optimization goal.From the perspective of the continuity of the video cache block,a caching strategy is proposed to guarantee the user's experience quality.The strategy is evaluated in the relevant platform,and the scheme can increase the bitrate by 12%compared with the LFU,ensuring the QoE of the user watching the video.The third part is the research of streaming media acceleration based on wireless sensing:The research focuses on the streaming media acceleration strategy based on wireless sensing in the mobile edge network scenario.Based on the basis of the first two chapters,the video information sensing capability in the mobile edge network is used to adaptively change the video request.Based on the user experience quality and multi-user perspective,the related caching strategy is proposed from the perspective of video fluency.The entire solution can dynamically offload user requests to the mobile edge computing platform,improving the bit rate and fluency experience of the user watching the video.Compared with LFU,our algorithm can increase the code rate by 18.6%and reduce the card time by 62.5%,which gives users a better viewing experience.Compared with the previous work,radio network-awared can increase the bitrate by 14%.