Research On Edge Computing And Pre-storage Resource Allocation Technology Based On Social D2D Network

With the proliferation of handsets and other derivative devices,the network traffic has witnessed a trend of explosive growth in recent years.In the next generation of communication technologies,the wireless caching and the mobile edge computing based on social network can offload tasks of to devices.This can improve utilization rate of resource,reduce download delay and computation delay.However,how to select optimal devices for caching and computing,how to allocate the caching and computing tasks are urgently to be solved.This thesis focuses on the above problems,and obtains the following research results:(1)Aiming at the cellular communication system based on social network,a many-to-one matching algorithm is proposed to solve the files allocation problem.This strategy can effectively reduce the average download delay in the network system.Simulation results show that the proposed algorithm has better performance than other benchmark algorithms.(2)Aiming at the problem that the important users in social network is hard to choose and rent,a many-to-one important users selection algorithm and a many-to-many files allocation algorithm are proposed.In the many-to-one important users selection algorithm,users are taken as competitive resources,combining the users’ social characteristics and the physical characteristics to achieve the selection,allocation and renting of important users.The many-to-many files allocation matching algorithm solves the problem of popular files allocation.Simulation results show that the proposed algorithms can effectively maximize social welfare and minimize average download delay.(3)Aiming at the lack of computing resources of devices,an edge computing node selection algorithm based on social network is proposed.Combining the communication characteristics,computing characteristics and social characteristics of user devices,the edge computing nodes can be effectively selected.The simulation results show that the proposed algorithm can effectively maximize the social welfare and minimize the average computational cost.Finally,the thesis summarizes the whole work and points out the existing deficiencies.In addition,the prospect of further research is discussed.