Research On Iterative Detection And Channel Estimation In Wireless MIMO Systems

Wireless multiple-input-multiple-output (MIMO) antenna systems offer significant improvements in performance and capacity when used in wireless communications, and have received much attention recently. In the last few years, MIMO techniques involved in 2G, 3G and beyond 3G systems have been considered and developed extensively in research. Among them, the iterative receiver and channel estimation techniques are the focus of this dissertation.Firstly, the detection algorithms are studied for wireless multi-input multi-output systems. To achieve a better trade-off between performance and complexity, iterative soft decision interference cancellation algorithms are investigated. It is shown that the probabilistic data association algorithm and linear minimum mean-square-error (LMMSE) based iterative soft decision interference cancellation (ISDIC) algorithm are equivalent. Meanwhile, the block-wise ISIDC is proposed and a new ordered serial ISDIC which is based on the instantaneous signal-to-interference-plus-noise ratio is also presented. Simulation results show that the ISDIC algorithms can achieve the optimal detector with low-order modulation.Furthermore, low complexity space-frequency turbo detection and decoding schemes are proposed for frequency selective MIMO systems. For the MIMO systems with a large number of transmit antennas, when high-order modulation is applied, traditional soft-input soft-output (SISO) detector will suffer from imprac-tically high computational complexity, especially in large rms-delay-spread environments. Using the algebraic properties of cyclic-prefix based block transmission systems, we implement the LMMSE based turbo detector in the discrete Fourier transform (DFT) domain to reduce the detector complexity. The algorithms can be applied to cyclic prefix based block transmission systems, such as precoded orthogonal frequency division multiplex (OFDM) systems and space-time coded OFDM, etc. Simulation results show that space-frequency turbo receivers significantly improve the performance of the systems, and the performance is close to the channel capacity.Due to the matrix inversion involved in the LMMSE based ISDIC algorithm, the computational complexity is still very high. Then, a low complexity polynomial expansion linear detector based soft-input soft-output detector is proposed. Since only matrix-vector multiplication is involved, the complexity of the SISO detector is very low. The algorithm can be applied to the uncoded iterative detection and the turbo receiver. Simulation results show that in most cases even with...