Research On The Design Of Physical Layer Multicast Beamforming Based On ADMM Algorithm

Since the beginning of the 21 st century,people's communication has evolved from simple text messages and phone calls in the past to many multimedia communication forms such as browsing short videos,long videos and live broadcasting.Compared with the traditional point-to-point transmission mode,physical layer multicast is a point-to-multi-point content distribution technology,which can better provide efficient and reliable transmission services to deal with the emerging scenarios and services.In the physical layer multicast system,given the channel information of the base station and the receiver,the spectrum efficiency and transmission quality of the system can be improved by beamforming the transmitted signals.However,the beamforming technology of existing multicast systems is not as mature as that of unicast systems.The beamforming problem of multicast systems is usually nondeterministic nolynminal-hard(NP-hard)problem,and the existing heuristic algorithms generally have high computational complexity and poor scalability.To solve the above problems,this paper will carry out research on the beamforming technology of multicast system and build an optimization solution Method based on Alternating Direction Method of Multipliers(ADMM).It is a method used to solve separable optimization problems.It can be widely used in machine and statistical learning,signal processing,data mining,compressed sensing and other large-scale data analysis and processing fields.When ADMM and its improved form are used to solve convex optimization problems,there are some general research results in terms of convergence and convergence speed.In this paper,based on the quality of service(Qo S)criterion,the beamforming design method for single group multicast scenarios and multi-group multicast scenarios based on ADMM is studied,which ensures that the signal noise ratio(SNR)of each user meets the given requirements and minimizes the transmission power of the system.The main content and work of this paper are as follows:1.This paper describes the single group multicast scenario,and models it according to the requirement of the correlation characteristic in the actual communication.A kind of efficient parallel execution ADMM algorithm is proposed,and the detailed solution process is given.Secondly,a series of analyses including convergence and computational complexity are performed on the algorithm presented in this chapter.Finally,this paper compares with the existing algorithms and verifies that the algorithms presented in this paper are better than the existing algorithms in terms of computational complexity and transmission power improvement.2.Expanding a single group multicast scenario to a multi-group multicast scenario differs from a single group multicast scenario in that there is intergroup interference between multicast groups,and the user's minimum channel signal to interference plus noise ratio(SINR)is considered in the modeling process.Based on the analysis of the transmission power and the requirement of the user's SINR in the physical layer multi-multicast system,the optimization model of beamforming design is established.Aiming at this optimization model,by introducing auxiliary variables to achieve variable splitting,the corresponding ADMM algorithm is proposed.By fully exploiting the structural characteristics of the problem during the implementation of the algorithm,the large-scale optimization problem can be decomposed into several small-scale subproblems in each iteration,and each smallscale subproblem has a closed solution.Secondly,a series of analyses including convergence and computational complexity are carried out on the algorithm proposed in this chapter.Finally,the simulation results verify the effectiveness of the algorithm proposed in this chapter.