Research On Content Caching And Delivery Scheme In D2D Mobile Network

With the rapid development of intelligent mobile terminal and diversified Internet services,the traditional cellular network has been unable to load the explosive growth of mobile data traffic demand.To resolve the conflict between the continuous surge in mobile data traffic and limited capacity of backhaul network,many solutions have been proposed in5 G era.Specially,the utilization of D2D(device-to-device)content caching and delivery mechanism in traditional communication network is one of the current promising solutions.In D2 D caching network,parts of contents are cached in advance on the terminals when the network is idle.When the content request is initiated,the content is delivered through D2 D communication.The direct communication mode between users offloads data traffic to the local area and effectively relieves the burden on the backhaul link during peak traffic time and reduces the transmission delay caused by network congestion.However,it is difficult to generate appropriate content caching and delivery strategies due to the limited cache space,device energy,diversified content requirements,location changes of users in mobile scenarios,and diverse social relationships among users.In view of above problem,this paper carries out research on the following two aspects:(1)In order to improve the matching degree between cached content and user requests and reduce the system overhead of content requests,this paper designs a content caching strategy based on user interest and similarity.Firstly,considering the content coding storage mode and the inter-contact model of user mobility,the optimization problem is established with the goal of minimizing cost of the D2 D content caching network.Secondly,according to the subject of content,user’s content request history and personal characteristic information,the joint influence of user’s interest preference and similarity between users on the request probability is analyzed.Thirdly,the influence of user mobility on system cost is analyzed theoretically.And then,a heuristic placement algorithm for content cache is proposed.Finally,the system average cost and cache offloading ratio of the proposed scheme are simulated through the data set to verify the effectiveness of the proposed strategy.(2)In order to improve the amount of content for requesting users through D2 D communication,this paper proposes a content delivery strategy based on user features and mobile awareness.Firstly,MD2D(Multi-Devices-to-Device)communication mode which collaborate multiple available D2 D links is introduced in this chapter to provide more efficient content sharing service for requesting users.And then,considering the mobility of user,the limited caching storage and energy of device and the trust relationship in society,we construct an evaluation framework to jointly consider the content offloading quality in physical level and the user relationship in social level.Subsequently,we define the D2 D content successful transmission amount as the performance metric,and content delivery link selection problem is modeled.Given the NP-Hard property of the problem,we first obtain a performance benchmark by brand and bench algorithm which is much more efficient than exhaustive search.Next,we prove that the problem falls into the category of monotone submodular function over one matroid and knapsack constraints.And then,we proposed a corresponding near-optimal algorithm based on greedy approach.The simulation results show that the strategy we proposed has better performance than other schemes under multiple parameter scenarios.In addition,the content delivery strategy based on branch-andbound algorithm is optimal,while the greedy-based algorithm sacrifices part of performance in exchange for the increase of computing speed.