Cache-based Video Performance Optimization In Mobile Edge Computing Networks

In recent years,the mobile data traffic dominated by video services has exploded,and the repeated transmission of a large number of popular video contents have led to a surge in backhaul link load,resulting in a large data transmission delay and a reduced user experience.In order to meet the user's demand for high-definition video,Mobile Edge Computing(MEC)effectively solves the problem of delay in the video transmission process while reducing;the backhaul link load by sinking the core network(Core Network,CN)content distribution function to the edge of the network and leveraging its data caching,computing and network awareness capabilities.Video clarity and transmission delay are the key factors affecting the user's video experience.The ultra-high definition video has higher requirements for the network in video encoding,storage and transmission.Therefore,how to effectively utilize video caching and transmission resources in the MEC network will become a key issue.Based on the characteristics of video services,this thesis focuses on video caching strategy and transmission and jointly optimizes cache resources,transcoding resources and wireless communication resources provided by MEC to improve the overall performance of video transmission and user experience.The main contributions are as follows:Firstly,this thesis conducts a full investigation on the network ar-chitecture and deployment methods of MEC.The key technologies such as edge caching and transcoding of MEC involved in this thesis are summarized.Then the current researches on wireless edge caching strategies are classified.Secondly,from the perspective of MEC system,the backhaul bandwidth saved by caching and transcoding is regarded as the system gain and the system resources(cache resources,transcoding resources)are consumed as cost.This thesis proposes a caching and transcoding strategy that maximizes the system utility.Given the wireless edge network topology and storage space,the popularity of different definition versions of video is considered.If the video requested by the user is cached at the MEC server,then the backhaul link is unoccupied.Due to the time-vary-ing channel conditions,this paper considers the channel conditions in the transmission phase when designing the buffering and transcoding strategy.This thesis considers the network awareness function of the MEC server and combines the adaptive bitrate technology to adjust the transmitted video version in real time.Then the optimization problem is solved by the two-layer many-to-many matching algorithm.The algorithm is compared with the exhaustive search algorithm and the random matching algorithm and the experimental results show that the proposed algorithm can obtain a relatively better solution at a lower complexity.In addition,the cache strategy has the best system performance compared to the traditional conservative cache and minimum redundant cache strategy.Finally,from the user level,in order to meet the user's low latency requirements for delay-sensitive video,this thesis analyzes the video transmission delay problem by comprehensively considering backhaul delay,wireless transmission delay and MEC transcoding time.The optimization goal of minimizing video transmission delay is established under the condition of guaranteeing users' QoS(Quality of Service)and backhaul bandwidth limitation.Aiming at the complex mixed integer nonlinear programming problem of modeling,this thesis proposes a successive convex approximation-assisted genetic algorithm to solve the problem and then obtains the suboptimal solution of the problem.Finally,compared with other baseline algorithms,the effectiveness of the proposed algorithm under different storage capacities of MEC servers and different users is verified by simulation.