The Single-user Mimo Wireless Communication Systems And Multi-user Pre-coding Technology Research

The precoding technology is a pre-processing stage with exploiting channel state information (CSI) at the transmitter, to improve system performance by adapting the variational wireless channel.Then, the key problem in precoding can be described as: how to exploit CSI, and how to process with CSI. The latter one in MIMO system is the main work of this dissertation.There are two methods for transmitter to exploit CSI: feedback and reciprocity.To explain the gain of precoding in a simple way, in chapter 3, we firstly discuss in condition of perfect CSIT(no distortion and no feedback delay),and analyze the optimal transmitter structure with precoding,then compare two precoding criteria(based on capacity and based on PEP) in matlab simulation.Due to the time variation of wireless channel, error of estimate at receiver, and feedback delay, etc, it becomes nearly impossible to obtain perfect CSIT, so we discuss the condition of perfect CSIR and partial CSIT, which is called limited feedback.Then we compare 2 quantization modes(quantizing channel matrix and quantizing precoding matrix) in simulation.The discussion above is base on sigle-user.In Multi-user scenario, there are more issues to solve. It will be mentioned in chapter 4 that: how to eliminate (CCI) co-channel interference, and how to successfully schedule users to exploit multi-user diversity, etc.Firstly, to solve the CCI eliminating matter, we firstly introduce block diagonalization (BD) method, and we find that in such a system, the spatial degrees of freedom are used for interference mitigation.Then we point out that when interference is at the same level as noise, such as in medium and low SNR regions, it is not effective to pre-cancel CCI (co-channel interference) while sacrificing spatial degrees of freedom. Therefore, according to different SNR, we can choose whether to pre-cancel CCI that is based on BD. In this way, because of the spatial multiplexing(SM) gain, it has an improved performance in low and medium SNR regions; besides, the complexity is greatly reduced due to ignoring CCI. Then, to solve the Multi-user scheduling problem, on the basis of optimal full search, we give a low-complexity method. Through simulation validation, we can find out that due to keep the Multi-user diversity gain well, the performance of new method is torlerably worse than optimal search, with the computing complexity reduced a lot.Finally, some conclusion to the whole dissertation will be given, and further research directions and issues are pointed out.