Research On Cooperative Video Caching Strategy For Mobile Edge Computing

The rapid growth of video traffic has brought huge burden to the Base Station(BS)and core network links.Meanwhile,it will seriously restrict users' pursuit of high service experience.Edge caching technology caches popular video content near users,which effectively reduces the backhaul traffic load,so that the end users can quickly obtain the required content.Therefore,it is of great significance to study the efficient video caching strategy enabled by mobile edge computing(MEC)for alleviating the backhaul link pressure and improving the quality of experience(Qo E).The main contents of the thesis are as follows:In view of the problem that the increasing demand of multimedia services such as video leads to the excessive backhaul link load,which leads to the deterioration of transmission delay and the reduction of Qo E,this thesis proposes a popularity aware cloud side cooperative video caching strategy.Considering the caching and computing capabilities of MEC,a distributed cooperative caching and computing framework is constructed to support the caching and distribution of video streaming.Considering the difference of global and regional video popularity and the change of video popularity with time,a two-stage caching scheme is proposed to minimize the average initial buffering delay of video delivery in the network,so as to make rational use of limited caching and computing resources.Simulation results show that the proposed scheme can reduce the initial buffer delay,backhaul traffic load,improve the cache hit rate and the quality of user experience.Aiming at the shortage of wireless resources caused by frequent requests for short videos from a large number of devices in the network,this thesis proposes a device to device(D2D)cooperative caching strategy based on social relationship awareness.Through the deep analysis of the physical and social properties of the device,a cooperative cache strategy based on user social relations is proposed to improve the sharing intention of end users,reduce network traffic load,and reduce transmission delay.In order to maximize traffic offloading,users are divided into communities according to their physical and social attributes,and users with strong physical and social attributes are divided into the same community,so as to improve users' willingness to share.For the users in the community,the relay nodes are determined according to the contribution of the users in the community when caching the content,and further according to the popularity of the video content in the community predicted by the collaborative filtering algorithm,so as to realize the personalized caching placement of the video content for the needs of users.Simulation results show that the proposed algorithm can increase the traffic offloading and reduce the transmission delay.