Investigation Of Downlink Transmission Technology In Multiuser Multiantenna System

Recently, Multiantenna technology is widely recognized as a modern technology that can effectively improve the spectral efficiency of wireless communication system. In Multiantenna systems, Multiantenna technology can achieve higher speeds and better communication quality with its unique spatial degrees of freedom, without taking up additional spectrum resources and time resources, or make a well tradeoff between transmission rate and transmission reliability.In this paper, we study the downlink transmission technology in multiuser multiantenna systems and make some contribution to the investigation of channel limited feedback technology, user selection technology and precoding technology in multi-user multi-antenna communication systems. The contents of the work are listed as follows:Firstly, we did research on limited feedback techniques in multiantenna systems. After the introduction of several common limited feedback technology, we focused on the codebook based limited feedback technology which is used more wildly in multi-antenna system recently. We described the basic strategy of codebook based limited feedback technology and we discuss in detail of two very important codebook design method:vector quantization codebook design method and Grassmann codebook design method. We propose a limited feedback mechanism which is based on valuebook and proposed the method to generate the valuebook. This feedback mechanism can effectively feeds the channel gain information (CGI), the base station can use CGI to do the user selection or user identification work.Secondly, the precoding methods in multiuser multiantenna systems are studied. The Dirty Paper Coding Technology which can achieve the capacity of downlink MIMO channel is elaborated and zero-forcing dirty paper coding method which can implement dirty paper coding capacity is also described. Because of the actual realization of dirty paper coding system is not feasible, we discusses Tomlinson-I larashima precoding whose computation complexity is much lower. In addition, we study two linear precoding methods:one is Block Diagonalization pre-encoding, which can eliminate interference between multiple users; the other is Signal to jamming plus noise Ratio precoding method, which can suppress both the interference between the users and the noise.Finally, we studied the user selection strategies in downlink multiuser multiantenna system, including the greedy user selection strategy, quasi-orthogonal user selection strategy and heuristic GS-orthogonal user selection strategy. On this basis, we proposed an optimization scheme in downlink multi-user multi-antenna transmission system under condition of imperfect channel state information in Base Stations. This optimization method use CGI from valuebook to select the optimal users, and then get CSI from codebook feedback and precoded only for optimal users. This optimization transmission scheme can for one thing guarantee the reliability of communications, for another it can significantly reduce the amount of feedback and increase the effective throughput of the system.