Transmit beamforming for MIMO wireless systems

A new trend in wireless communications is to use multiple transmit and receive antennas to attain higher data rate, better signal quality, and superior system flexibility. In this dissertation, we aim to design novel transmit beamforming schemes for multiple-input multiple-output (MIMO) wireless systems.; The space-time coding (STC) technology has shown great potential in exploiting the high capacity from the MIMO systems. In this dissertation, we address the application of transmit beamforming in the STC systems. Based on a mean feedback model, we have developed a linear beamforming scheme for the quasi-orthogonal space-time block codes. The proposed quasi-orthogonal beamformer accomplishes high transmission rate as well as high order spatial diversity. In a related study, we have derived a new performance criterion based on channel phase feedback. This design criterion is used to construct a new class of codes called co-phase space-time trellis codes (CPSTTCs). The performance of these new codes is evaluated using numerical simulations.; We also investigate the application of transmit beamforming algorithms for time-selective and frequency-selective fading channels. For the time-selective case, we develop two classes of beamforming algorithms that exploit the mutual correlations in the fading channel. The predictive vector quantization (PVQ) beamformer accomplishes superior power delivery. However, it has a high implementation complexity. The successive beamforming (SBF) algorithm has a much simpler implementation, and its performance is very close to that of the PVQ beamformer. In a time-selective fading environment, both the PVQ and SBF algorithms are superior to the existing memoryless beamformers. For the frequency-selective fading case, we investigate the finite rate beamformer design for the orthogonal frequency division multiplexing (OFDM) systems. To reduce the feedback requirement, we exploit both the time and frequency domain correlations. Based on the SBF approach, we develop several classes of beamforming algorithms. All these algorithms require little feedback, and they outperform other existing finite rate beamformers in the literature.