Nonlinear Precoding Mimo System

Precoding is a key technology in wireless communication system. It increases the adaptation of transmitte signals effectively and improves the system performance by pre-process the transmitte signals using channel state information at the transmitter (CSIT).Tomlinson-Harashima Precoding (THP) is a typical nonlinear precoding and it was initially introduced for the pre-equalization of intersymbol interference in single-input single-output (SISO) channels, and now it is widely used to cancel the interference among the symbols, carriers and users in MIMO channels. This paper focuses on Tomlinson-Harashima Precoding, and researches the performances of QR-type decomposition THP based on zero-forcing (ZF) or minimum mean-square error (MMSE) criterion in perfect and imperfect channel state imformation cases. To improve the bit error rate performeances of conventional THP methods, a new MMSE based THP is proposed using Geometric Mean Decomposition (GMD). This paper has an intensive study on the improved MMSE-THP scheme.In part one and two, the paper gives an overall introduction of the research background, the development of precoding techologies and some key technologies of the future wireless communication system which are the basis of following researches.Then, this paper gives some detailed introductions on ZF-THP and MMSE-THP basing on QR decomposition in single user and multi-user cases. Meanwhile, it compares and analyses the performances of ZF-THP and MMSE-THP systems, and summarizes the research acheivements of THP. However, there are significant performance differences between specific mobile stations in multi-user case. In this paper,"best-first"ordering is used to minimize the maximum noise variance for the ZF-THP and error variance for the MMSE-THP. A new ordering method is proposed when the number of users is not too large, the new method is easy to understand and the performance of it approximates to that of the"best-first"ordering method. Ordering can cancel the quality diversity of different users and achieve better performance compared to those systems without ordering.The fourth part of this paper has further reseaches on the performances of QR decomposition based THP systems with imperfect channel state information at the transmitter, which is close to the actual communication environment. The main focuses of this part are the effects of Doppler frequency offset and error estimate. Researches in this part can be considered expansiones of the existing researches.The fifth part is the key innovation of this paper. An improved MMSE-THP using GMD decomposition is proposed here and simulations are used to show the advantages of the proposed algorithm. The effects of Doppler frequency offset and error estimate to the improved method will be discussed in this part. Considering error estimate of channel matrix, a robust scheme of this new MMSE-THP is proposed at the end of this part.Finally, some conclusions to the whole dissertation will be given, and further research directions and issues will be pointed out.