Research On Optimization Of Access Relationship In Green Wireless Communication Network

In recent decades,with the continuous development of communication technology and the widespread use of wireless smart devices,the number and demand of users has exploded unprecedentedly.The number and types of communication base stations(BSs)added to ensure service quality have also increased.As a result,the energy consumption of wireless communication networks continues to increase.In a wireless cellular network,the power consumption caused by maintaining base station operation and frequent user switching accounts for most of the total wireless network consumption.Therefore,optimizing the BS-user relationship in a wireless network is an effective way expected to reduce the total network overhead.Scholars at home and abroad have done a lot of research on this issue.The base station-user access relationship is mainly divided into two aspects: user access control and base station activation control.This article mainly researches from the above two aspects.The current user access control is generally optimized in conjunction with power control,which try to support the largest number of users with minimum power under the condition of meeting the quality of service(QoS)of the user,finally achieving a balance between the number of service users and the transmission power.However,the current related algorithms all make decisions based on real-time channel state information(CSI),which leads to the risk of frequent disconnection of user access status,causing the network to spend extra energy to overcome its influence.Therefore,this paper considers the user link switching frequency over a period of time based on the original problem.It is expected to find a balance between the stability and flexibility of the access user set,and meanwhile minimize the total energy consumption of the network,i.e.the joint design of long-term user admission control and instant beamformingIn terms of base station activation,the usual method is that base stations share certain information with each other and work together according to the requirements of the application scenario to optimize resource allocation and reduce unnecessary losses.At present,the best solution is to jointly optimize the transmission power and the activation of the base station.However,due to the dynamic characteristics of CSI,the impact of the switching frequency of the base station on network resource consumption is not considered.Therefore,the second part of this article considers the switching frequency of the working state of the base station over a period of time on the original problem,which aims to find a balance between the stability and flexibility of the active base station set,and meanwhile minimize the total network energy consumption,that is the joint design of long-term base station activation and instant beamforming.Because the non-deterministic polynomial(NP)problems that are presented in this paper are difficult,some effective approximations to these two joint optimization problems are utilized to make them convex problems.In the joint long-term optimization of admission control and beamforming problem,this paper uses the Successive Upper-bound Minimization(SUM)to design offline and online algorithms,respectively.To meet the needs of practical applications,this paper uses the Alternating Direction Method of Multipliers(ADMM)framework to design a distributed algorithm with low complexity to iteratively solve online problems.In the joint long-term optimization of base station activation and beamforming problem,this paper also presents offline and online algorithms and ADMM-based online algorithms that can be solved in parallel using similar methods.In addition,this paper uses the Deep Reinforcement Learning(DRL)framework to design a data-oriented intelligent online algorithm to solve the problem of base station activation.The experimental simulation results show that the algorithm proposed in this paper for user control and base station activation can effectively reduce the switching frequency of BS-user link,and achieves better performance on the total network energy consumption.In addition,through a large number of experiments,it is also found that the intelligent online algorithm can surpass the performance of the offline algorithm in terms of base station activation.