Propagation Dynamics Based Resource Deployment Strategies For Edge Networks

In recent years,with the increase of the number of mobile devices and the rapid growth of Internet data traffic,the burden of core network,access network and backhaul link is increasing.The rapid growth of mobile data service demand brings great challenges to the effective transformation of wireless network resources into serving capabilities.To tackle this challenge,in addition to increasing spectrum bandwidth to increase system capacity and adopting advanced wireless transmission technology to improve spectrum efficiency,mobile edge computing is also one of the promising solutions.As an important part of mobile edge computing,edge cache aims to reduce redundant content transmission,reduce the burden of backhaul link,reduce content acquisition delay,and improve quality of experience by deploying content resources to edge networks in advance.Research shows that one of the important sources of explosive growth of mobile communication data traffic is repeated requests and transmission of few popular contents.Considering that content popularity will dynamically change with the propagation process,and network resource deployment have impact on content popularity.Meanwhile,content popularity will directly affect the network resource deployment scheme.Dynamically analyzing the relationship between edge network resource deployment strategy and content popularity can improve the accuracy of resource deployment and the efficiency of edge network resources.Focus on these problems,the main contributions are as follows:Firstly,a propagation dynamics based content propagation model is proposed.With the rapid development of social networks,the distribution of users’ requests in wireless networks is not stable.Users’ interest is influenced by the surrounding users in social networks,resulting in the time-varying nature of content popularity in the network.Based on the theory of epidemic model,we propose a content propagation model to analyze the time-varying characteristics of content popularity and its interaction with network resource deployment.Firstly,the edge network scenario is proposed to provide a basis for the establishment of propagation model.Secondly,network resource deployment,an important factor affecting the content propagation process,is added to the content propagation model to describe the interaction between network decision and content propagation.Finally,considering that the basic propagation model will evolve with content propagation process,the corresponding evolution model needs to be built to improve the description of the content communication process.Secondly,an edge base station(BS)resource deployment strategy based on propagation dynamics is proposed.In the small cell edge network,each BS provides a cache placement strategy according to the content request characteristics in the coverage area,so that users can directly obtain the required content from the BS.The pressure of backhaul link and the content acquisition delay can be reduced.Based on this scenario,we apply the propagation model suitable for the edge network scenario and establish the problem of maximizing the user demand satisfaction rate.Based on factor graph and belief propagation algorithm,the resource deployment strategy of edge BS is proposed in this thesis.Simulation results show that compared with the benchmark schemes,our proposed strategy can effectively improve the user satisfaction rate,cache hit ratio and average content acquisition delay.Thirdly,a resource deployment strategy in cooperation edge network is proposed.In the scenario of cooperation between BS communication and D2D communication,the propagation process of content in the coverage area of BS will change,and the characteristics of users’ requests will also become different.Under this condition,the deployment strategy of edge network resources needs to be transformed accordingly to improve the utilization efficiency of network resources and service quality,and reduce the burden of backhaul links.In this thesis,the propagation dynamics model suitable for BS and D2D cooperative communication scenario is reconstructed,as well as the problem of maximizing user demand satisfaction rate,and the resource deployment strategy of edge cooperative network is proposed.Simulation results show that compared with the benchmark schemes,the network performance of our proposed scheme is improved.