Research On Delay Performance Of Wireless Caching Network

The explosive growth of global mobile data traffic,which is mainly multimedia data,resulting in severe increases in user latency and a serious burden on backhaul links which connect the base station to the core network.The network congestion in backhaul links may lead to more seriously long transmission delays and thus reduces the overall quality of experience(QOE).In order to support the growth of mobile data traffic,it is an effective method to reduce the transmission delay and backhaul traffic by deploying caches at BSs.The base station caching technology is used to cache the frequently requested data to the small base station or wireless terminal which is closer to the users,thereby satisfy the user's request without increasing the burden over backhaul links.This paper is to study and discuss how to deploy base station caching to reduce system transmission delay and improve system performancem in the context of the 5th-Generation heterogeneous network.In this paper,firstly consider the delay performance of the two layer wireless caching network model un-der the condition of heterogeneous network.The nodes of the two layers are modeled as mutually independent homogeneous Poisson Point Process.All the helpers have caching ability,but only a part of the users have caching ability.So the total transmission delay includes D2D-delay,helper-delay and BS-delay.According to the stochastic characteristics of the PPP distribution,the probabilities of three kinds of delays are calculated,and the average transmission delay is obtained.Based on the optimization of the average transmission delay,the optimal caching placement of the two layer wireless caching network is proposed.That is,when only a few cache-enabled users and helpers are involved in the communication network,more popular files should be cached.When there are many cache-enabled users and helpers participating in the communication network,we should consider the caching diversity and cache more files as much as possible.Using DC(difference of convex)algorithm to solve the optimization problem and get a good convergence.The simulation results show that the performance of the optimal caching strategy is better than that of the popular caching strategy and the even caching strategy.Nextly,we analyzes the delay performance of wireless caching network from another point of view.Firstly,the caching network model is established,which considering each base station does'nt cache com-plete files,but a certain percentage of the file.Thus a new user connection strategy is considered,which is quite different from the former one.When a user requests a file,there is a cluster of candidate base stations.According to the sorting of the connection delay between user and BSs,we choose the base station with lower delay from the cluster to transmission files to user.The transmission delay includes two parts,the first is the average transmission delay when user has the cluster of candidate base station;the second is the transmission delay when user does not have a cluster of candidate base station and will be satisfied from the BSs which get files from the core network through the backhaul.The backhaul delay of the second part is known.According to the two partial delays,the objective function for minimizing the transmission delay problem is established.The optimization problem is solved by the Lagrange Relaxation algorithm,which achieves a good conver-gence.The simulation results are compared to the random complete caching strategy and the fair caching strategy,and the delay performance of the optimal caching stratey with known backhaul delay is proved to be better.