Research On V-BLAST Detection Algorithm And Multi-User MIMO Precoding

With the rapid developments in communication technologies, the demands for services of high data-rate and quality are growing. As the available radio spectrum is very limited, multiple-input multiple-output (MIMO) technology draws much attention due to the high channel capacity it can provide. Possessing high spectral efficiency and relatively simple transmitter and detector structure, Vertical-layered space-time (V-BLAST) system proposed firstly by Foschini of Bell Labs in 1998 is one of the most important signal processing techniques for MIMO system and has acquired extensive applications in the passed a dozen years.Due to the high complexity of mobile communication channel, the analysis and classifications of mobile channels build the foundation for researches on MIMO techniques. Starting with the propagation mechanism of electromagnetic wave in mobile communications, the thesis analyzes various parameters of small-scale fading, on the basis of which, the small-scale fading channels are divided into six types according to the impacts on the signals exerted by these parameters during the transmission. Further, depending on the statistical characteristics of small-scale fading, the thesis discusses two classical small-scale fading models and presents three multi-antenna MIMO fading channel models.On the basis of the in-depth analysis of V-BLAST structure, simulation results on V-BLAST detection algorithms are presented in this thesis to evaluate the advantages and disadvantages of existing algorithms. Then a new algorithm, minimum mean-square-error ordered-grouped interference-suppression (MMSE-OGIS), is proposed based on the combination of singular value decomposition (SVD), maximum likelihood (ML) detection and MMSE ordered successive interference cancellation. On the platform of the V-BLAST system, the simulation is performed and the results obtained show that MMSE-OGIS algorithm can improve system performance with an acceptable increase in complexity. Being suitable for various modulation schemes and group lengths, the proposed algorithms can provide flexibility in practical use.In addition, due to the fact that the multi-user MIMO (MU-MIMO) is becoming the trend of mobile systems, the suppression of interference among multiple users in MIMO system has become the new research hotspot. Starting with researches on pre-equalization technique used at the transmitter to suppress the interference among multiple streams on the basis of single-user MIMO model, the thesis provides simulation result to show the satisfactory performance by means of precoding at the base station in the downlink. Further, the precoding model of MU-MIMO is presented. The block diagonalization method, which decomposes the multi-user channel matrix through SVD decomposition to perform transmitter precoding, is used to eliminate the co-channel interference from other users. In this way, the MU-MIMO interference suppression can be transformed into the problem of canceling multiple parallel single-user MIMO interference. Finally, simulation results obtained in a variety of situations show the good performance of the MU-MIMO model.