Multi-Antenna Mobile Radio Channels: Modelling and System Performance Predictions

Implementing multiple antennas at one or both ends of wireless communication systems has a significant impact on improving system performance. However, the gains in system capacity are sensitive to the spatial and temporal structure of the wireless channels upon which the system is built. The focus in this dissertation, is on using measurements to extract detailed information about the spatio-temporal behaviour of mobile wireless channels, modelling their dynamics, and utilizing this knowledge to improve transmission and receiving strategies.;In the second part of the dissertation, the information gained from studying the spatial structure of wireless channels is applied to improve the performance of multiple antenna systems. A new space division multiple access transmission technique is proposed, along with the specification of appropriate implementation algorithms. These algorithms utilize knowledge of the existence of multipath clusters, and angles of departure at a base station associated with transmission to specific users through selected clusters, for efficient beamforming in a hypothetical system, and it is demonstrated that the capacity is increased. Examples are considered in which the proposed transmission technique is applied in both broadcasting and cognitive networks.;In the final part of the dissertation, the information extracted from the SIMO sounding measurements is used to study the effect of several aspects of the receive antenna arrays on system performance. The extracted multipath parameters are used to reconstruct fields at the arrays under consideration, and subsequently study the effect of array configuration and element efficiency on mean effective link gain, eigenvalue dispersion and outage capacity under real-world channel conditions. Three configurations of a dual inverted-F receive antenna array in a generic mobile handset are studied.;The first part of the dissertation reports a measurement-based analysis that was conducted during the thesis project to characterize the dynamics of the spatial and temporal structure of wireless channels. Measurement data from a wideband single input multiple output channel sounding system with a 32-element uniform circular array that were recorded in a typical urban environment (downtown Ottawa) using a fixed transmitter and a moving receiver were analysed. Using a high resolution algorithm, the power, delay and angular parameters of the different multipath components were estimated. Following this identification of multipath parameters, major multipath clusters were identified, and their behaviour, such as the length of their active intervals, changes in their powers, as well as delay and angular characteristics within their active intervals were then modelled The objective is to contribute to the development of a channel model having statistical characteristics that change in time to replace current models that generate outputs with the same statistics throughout total simulation intervals. It is believed that this will aid in the development of more efficient, adaptive, transmission techniques for use in future wideband systems.