Distributed Beamforming Desigh For Energy Efficiency In MU-MISO Interfence Channels

With the popularization of mobile smart terminals and the booming development of mobile Internet services,the number of mobile data services is exploding,and users are increasingly demanding the speed of wireless mobile networks.Multi-user and multi-antenna technology can greatly improve spectrum efficiency and energy efficiency.It is one of the key technologies of post-4G mobile communication systems.At the same time,multi-user linear beamforming technology is a low-complexity method for implementing multi-user antenna systems.Due to factors such as volume and energy consumption,the base station uses multiple antenas and the terminal uses a single antenna.In this scenario,because there is coupling between downlink users and the downlink rate has a greater impact on the user experience,this paper studies the MU-MISO downlink.In addition,energy efficiency has gradually become a very important criterion due to the explosive growth of energy consumption in wireless communication networks.Therefore,this paper focuses on the energy-efficient distributed beamforming design in MU-MISO interference channels.This paper mainly studies the problem of beamforming design based on the maxi-mum energy efficiency in the MISO interference channel.And proposed two kinds of distributed algorithms.Since the objective function of the problem of maximizing energy efficiency contains a fractional form,it is a non-convex function.First,the commonly used non-convex optimization methods such as semi-definite re-laxation,D.C.planning,continuous convex approximation,fractional programming,and WMMSE algorithm are introduced.Based on this mathematics,this paper studies the problem of single-user scenario,gives the system model under this scenario,and then solves the scenario problem.For this scene optimization problem,it contains the rank 1 constraint.Firstly,the semi-definite re-laxation algorithm is used.Discard the rank 1 constraint,find the relaxation solution of the original problem,and then expand into the form of the difference(DC)of the two-convex function by introducing auxiliary variables.Then use the continuous convex approxima-tion algorithm for the non-convex part of the DC function to make the entire objective function.For the convex function,by solving the approximate solution of the relaxation problem,we continuously approach the optimal solution of the original function.Finally,an iterative algorithm for finding the optimal solution of the original function is given.Then,the above problems are extended to multi-user scenarios,and a distributed beamforming design algorithm based on energy efficiency under multi-user scenarios is proposed.Due to the coupling between multiple users,the scenario is much more com-plex than the single-user scenario,but its algorithmic idea is similar.Firstly,there is a rank 1 constraint in the objective function,a semi-definite relaxation algorithm is used to discard the rank 1 constraint,and it is proved in the following that the rank of the opti-mal solution to the original problem is 1,so the original problem can be solved without loss from the final solution.Excellent solution.For the semi-definite relaxation problem,by introducing variables,the fractional form is transformed into the form of the differ-ence between the two convex functions(DC).Then the continuous convex approximation algorithm is used to approximate the entire DC function by convex approximation and a series of sub-questions after decoupling are obtained.,and introduced the concept of interference pricing.Finally,for multi-user scenarios,a second distributed algorithm for iteratively solv-ing beamforming designs with maximum energy efficiency in multi-user scenarios is pro-posed.The solution of the closed-form solution is given for each iteration,which greatly reduces the time complexity of the algorithm.The algorithm uses fractional planning,weighted mean squared error(MSE)and the equivalence relationship between the mini-mization problem and the weighted sum rate problem,respectively.The solution is based on the WMMSE algorithm,the algorithm is less complex,and similar energy efficien-cy can be obtained.This algorithm is also a distributed algorithm and each user solves independently.