Research On Data Forwording Scheme And Content Caching Strategy In D2D Communication Networks

Recent years,along with the large-scale promotion of smart mobile devices and rapid development of wireless network technology,the problem caused by dramatical growth of wireless traffic data has becoming more and more severe.The traditional cellular network is unable to satisfy the ongoing demand for mobile data communications due to the limited wireless band resource.People need a new type of wireless communication technology to meet the demand of higher band,lower delay and lower power consumption.Device-to-Device communication is a new type of wireless communication technology which permits mobile terminals to communicate with each other directly.Because of the benefits in increasing spectrum utilization efficiency,enhancing the transmission rate and saving the power,D2 D communication has become an important developing direction of future network and drawn much attention recently.In this paper,we analyze the performance of D2 D communication underlaying cellular Networks,which includes data forwarding scheme and caching strategy.The main contributions of our work are demonstrated as following:(1)We model an architecture of D2 D communication underlaying cellular Networks,assuming the distribution of user nodes follow Poisson point process and derive the probability distribution function of the distance between user nodes.An flexible scheme for mobile UEs to select communication mode individually is present,which is based on the threshold of channel SINR value.Besides,we assume the edge nodes can communicate with base station(BS)in a two-hop manner.Finally,we derive the network weighted capacity and verify the superiority of the proposed architecture by numerical results.(2)Based on the aforementioned architecture,we propose a distributed caching strategy.The network files are modeled by Zipf distribution function and ordered by popularity.We define the “D2D group” that the user nodes in the region can communicate with each other through D2 D links.The storage space of all the nodes in the D2 D group are treat as a whole which permits the nodes can cache the network files together.(3)We consider the sociability of user nodes and derive the preference degree of network files by recommended algorithm to recommend to users the files they prefer.As the same time,the weighted delay is derived based on file popularity and we define the preference degree-delay index which combines the preference degree and weighted delay.Finally,we propose cache selection algorithm through the solution of knapsack problem to determine the cache matrix of the network.