Research On Content Caching Strategy For Wireless Access Networks

In recent years,with the widespread popularization of various wireless devices,people’s demands for instant multimedia services have become stronger and stronger.However,due to the overburdened backhaul link,the traditional centralized network architecture cannot meet the everincreasing traffic demands of users.In addition,a feature of existing networks is that there may be severe congestion problems during peak hours,while communication system resources are not fully utilized during off-peak hours.In order to improve this problem,the industry proposes wireless caching technology based on the redundant nature of repeated traffic in the network.By equipping cache nodes near users,it shortens the distance between users and their requested content,greatly reduces the transmission of redundant traffic in the network,and alleviates the problem of shortage of network resources.The basic problem of radio access network caching technology research is how to deploy cached content under the condition of limited storage space to improve system utility.This thesis considers the scenarios of base stations and UAVs as edge nodes to study the caching problem,and takes the profit maximization of each caching participant as the optimization goal to model and solve the problem.The main work and innovations are as follows:Firstly,for the caching scenario where the base station is used as an edge node,in order to ensure the safety and reliability of file transactions,this thesis proposes a caching strategy based on blockchain technology,which implements a decentralized file caching and transmission process.as a cache chain.Among them,users serve as lightweight nodes,and content providers and base stations serve as full nodes.Based on the smart contract mechanism,the three store the transaction content and automatically execute the predetermined transaction process.Through the analysis of the file transfer process,the optimization goal of maximizing the profit of the content provider is established.In order to solve the optimization objective function,a content placement algorithm based on greedy strategy is proposed,which can optimize the profit obtained by content providers under limited storage space.In addition,in order to determine the dynamic connection relationship between the base station and the user,a linear model is used to predict the user’s location.The simulation results show that when the user mobility is weak,the most basic linear model can achieve a very good prediction effect.At the same time,blockchain technology can prompt operators to actively offload network traffic by formulating incentive mechanisms.Secondly,In order to cope with the new caching requirements in the network,This thesis introduces UAVs to achieve additional traffic offloading.In the scenario where the base station and the UAV act as edge nodes together,an improved cache deployment algorithm is proposed.Firstly,an optimization problem is constructed to minimize the average transfer cost of files.By disassembling the original problem,it is proved that the solution process of the constructed optimization problem has a high degree of complexity.For this reason,this thesis divides the problem into two modules to solve separately.In the process of solving the base station cache problem,due to the existence of overlapping coverage of base stations,a greedy idea is used to solve the problem.When solving the UAV cache problem in the constructed model,the UAV cache problem is transformed into a user assignment problem,and an improved cache deployment algorithm is proposed based on agglomerative hierarchical clustering algorithm.The simulation results show that the improved algorithm can achieve a better user allocation scheme.Compared with the baseline scheme,the proposed algorithm can significantly reduce the overhead of transferring files.