Statistical Modeling of Multi-Polarized MIMO Channels

In this thesis we synthesize a new model for correlated Rician dual-polarized MIMO channels that enables analysis of MIMO system performance in asymmetric channels. A number of factors distinguish our work from existing models , including the incorporation of a K-factor matrix that is used to assign arbitrary K-factors for to the different sub-channels, asymmetric power imbalance parameters, and space-polarization correlation characterizations. The model is used to analyze MIMO system capacity bounds, where lower and upper bounds for the ergodic capacity are derived. Two approximations are also considered to estimate the outage capacity for the 2x2 dual-polarized MIMO system, where the accuracy of the approaches are investigated for different 2x2 dual-polarized MIMO channel realizations. An extension from the 2x2 case to an arbitrary number of dual-polarized antennas at the transmitter and the receiver is also presented. Based on the extended model for a 4x4 system, we propose transmission schemes for the full-dimension architecture as well as for spatial beamforming, polarization beamforming and mixed beamforming architectures assuming that channel state information is available at the transmitter. The performance of the proposed transmission schemes are compared for different channel conditions using the developed channel model.