Resource Allocation In Cache-enabled Wireless Access Network

With the development of wireless communication technology,the demand for the communication ability of mobile terminals increased rapidly in the last few years.The radio resource is more and more sparse and operating costs of mobile communication services are increasing day and day.In order to solve these problems,the network architecture has been constantly evolving.The research shows that by deploying a large number of small base stations with low energy consumption and low cost,the system capacity can be increased and the spectrum efficiency rate can be improved.In addition,in mobile communication services,multimedia applications have replaced voice service as the most important service,and users' requests for multimedia content such as videos are mostly concentrated in a few hot content.Considering the reuse of content,caching at the radio access network have also been mentioned more.However,the intensive deployment of small cells brings more serious interference and requires a more reasonable resources allocation method to solve the problem.In addition,for caching technique,it is also necessary to give full play to the network cache storage through better resource allocation.In this paper,we propose three new resource allocation methods in different scenarios and with different control types.First of all,we introduce the development of mobile network architecture and the development direction of radio access network in the future,and we also introduce a cloud-based caching-enabled small cell network which is considered in this paper.Then,for a downlink network composed of multiple small base stations with cache storage,a resource allocation method based on evolutionary algorithm is proposed to improve the system throughput.The algorithm consists of two phases.Firstly,the small base stations determine their access users according to the distance between the user and the small base station and the caching condition of small base station.Then we use an improved evolutionary algorithm to allocate power to each user.Simulation results show that the proposed resource allocation algorithm can effectively improve the system throughput performance.Then,for a similar network architecture,we consider a hybrid control method in which the information in the network can not be fully known.We proposed a resource allocation method based on game theory and reinforcement learning.Finally,we analyze the resource allocation performance with mobile users,and we propose a resource allocation method with the consideration of user mobility.In this method,the most part of the content obtained by each user comes from the cache of the small base station.Simulation results show that the algorithm can effectively improve the system throughput and energy efficiency.