Research On Hybrid Precoding Algorithm For Millimeter Wave Massive MIMO System

Mobile communication is playing an increasingly important role in today's society,and it has brought us great convenience.With the upgrading of electronic equipment and the generation of various new services,people have put forward higher requirements on the spectrum efficiency,energy efficiency,and low delay of the new generation communication system.In this context,Massive MIMO(5G)technology,which is a new key communication technology,has become the most challenging and interesting subject in wireless research.It combines millimeter-wave technology and the gain brought by large-scale antenna arrays.It can make full use of resources to improve spectrum utilization to a large extent,and meet multi-device connections and high data rates,more bandwidth,low latency quality of service,and low interference.The demand has received great attention from many researchers.This thesis mainly researches massive MIMO technology.Aiming at the large number of RF links required by the all-digital precoding method,it brings relatively high cost,energy consumption,complexity and other issues.The hybrid precoding method is studied to improve the system.The energy efficiency performance of the work can be summarized as follows:(1)The hybrid zero-forcing(ZF)precoding method is studied.In this paper,an improved hybrid precoding algorithm based on the minimum mean square error(MMSE)is proposed for the downlink massive MIMO millimeter wave system.The method is mainly divided into two phases: by ignoring the interference between users to obtain the best analog precoder and analog combiner between the base station and the user terminal;using the limited feedback channel of the base station and the user terminal,using the MMSE algorithm to design digital Precoding section.In addition,quantized beamforming codebooks are used to generate analog and digital beamforming vectors with small training and feedback overhead.The performance of the improved algorithm proposed in this paper is close to the all-digital precoding algorithm.This can effectively reduce the number of RF links required,and achieve higher spectral efficiency than traditional hybrid precoding algorithms.(2)The hybrid precoding design of the downlink Massive MIMO system is studied in order to achieve the goal of minimizing the mean square sum of multiple antenna users.The proposed beamforming technique takes into account both horizontal and vertical planes.The hybrid precoding scheme is implemented by the following steps.For the combiner at the user and the baseband precoding at the base station,the solution can be obtained by convex optimization methods;the particle swarm algorithm(PSO)is used to solve the non-convex problem,and the RF is quickly obtained by updating the global optimal position of the particles.Precoding.When the mean square error(MSE)is minimal,an approximate optimal hybrid precoding can be obtained.Can significantly improve the achievable spectrum efficiency and system reliability.(3)The problem of pilot pollution in downlink Massive MIMO systems is studied,and a pilot allocation scheme based on soft pilot multiplexing and weighted graph coloring is proposed.First,the cell users and pilot sequences are divided into two groups,and the celledge users are assigned orthogonal pilot sequence groups due to severe pilot pollution.Then,for cell-center users,the text uses weighting based on classic graph coloring The interference pattern method further purifies the slight pilot pollution among users in the center of the cell.Compared with traditional methods,the proposed new method can effectively reduce pilot pollution and improve the energy efficiency performance of hybrid precoding.