Research On Edge Caching Technology Based On Cooperative Strategy And Coding In Ultra Dense Networks

To alleviate the pressure brought by the explosive growth of mobile data,ultra dense network has been introduced as a new network architecture.This network can increase network capacity and meet user needs by increasing the deployment density of small base stations,but it also leads to backhaul congestion,huge energy consumption,and excessive cost.At the same time,edge caching technology has attracted much attention as a means to solve the rapid increase in mobile data.This technology can store popular content at the edge of the network near end users,thereby reducing service latency for user-requested content,use of network bandwidth and energy consumption.Therefore,the application of edge caching technology in ultra dense network is is of great significance.In this thesis,considering the deployment characteristic of ultra dense network,the application of edge caching technology in ultra dense network is studied in depth.According to different optimization goals,two efficient content placement caching schemes are designed.The first scheme considers that users may move back and forth frequently in small base stations that are close to each other;The second scheme considers that when the coverage areas of small base stations overlap,the user equipment may be connected to multiple small base stations.The main research contents are summarized as follows:Aiming at the service latency of users in ultra dense network,combined with the short-range deployment characteristic of small base stations and the mobility of users,this thesis proposes a cooperative caching scheme based on mobility awareness.Firstly,based on the distance,a hierarchical clustering algorithm is used to group all small base stations.Small base stations within a group can share cached resources.Secondly,the user's movement trajectory in all small base stations is modeled as a Markov chain.Then,a content placement caching optimization problem is set up with the goal of minimizing the average delay of the user requests files,and a greedy algorithm is proposed to solve the problem.Finally,compared with the simulation analysis of the most popular file caching scheme and the cooperative caching scheme,it proves that the proposed caching scheme has obvious performance advantages in reducing latency.Aiming at the load of backhaul links in ultra dense network,combined with the overlapping coverage characteristics of small base stations and the independent attribute of Maximum Distance Separable(MDS)codes,this thesis proposes a coded caching scheme based on multiple connections.Firstly,split the file into fragments,encode the fragments into MDS-coded segments,cache the encoded segments of the file,and calculate the number of encoded segments that each file is at least encoded into.Secondly,for ultra dense network scenario,consider that users may be connected to multiple base stations,and multiple base stations provide services.Then,the optimization problem of the placement cache of the file encoded segments is set up with the goal of minimizing the average backhaul rate of the small base stations,and the optimization problem is sorted into a convex optimization problem for solving.Finally,compared with the simulation analysis of the randomly placed caching scheme and the most popular file caching scheme,it proves that the proposed cache scheme has obvious performance advantages in reducing the backhaul rate.