Against the background of the rapid evolution of mobile communication systems in the areas of service provision and capacity enhancement described above, the main focus of the research is on coded space-time processing techniques. The use of space-time processing is an attractive solution because it can mitigate the effects of multipath fading as well as suppress co-channel interference, therefore, significantly improving system performance. The topics are presented in the context of designing mobile communication systems where the two core areas of spatial processing and error coding are to be integrated in an optimum way. Of particular importance in this thesis, will be those CDMA based solutions for the mobile sector and the new performance analysis issues that need to be addressed as a result of the introduction of heterogeneous services and service environments into a single, mobile cellular access network. Furthermore, novel applications of turbo transmit and receive antenna diversity and beamforming techniques to mobile cellular access networks aimed at increasing the efficiency of such networks are considered. The thesis has the following goals: • To establish a general spatial/temporal channel model for use in the evaluation of coded space-time processing concepts applied to CDMA networks. • To analyze the performance of uncoded cellular CDMA systems incorporating space-time techniques using analytical methods in a number of realistic application scenarios. • To design, implement and evaluate coding strategies for incorporation into the space-time CDMA systems. This objective can be broken down into the following items: --- Space-time coding systems when considering multiple transmit antennas for the downlink. --- Coded space-time systems when considering multiple receive antennas for the uplink. • To establish the performance of coded space-time CDMA cellular networks under realistic scenarios. This thesis introduces many (some novel) space-time turbo coded techniques to increase the downlink capacity of a cellular CDMA network using multiple transmit antennas. For improving the uplink capacity, coded space-time diversity and beamforming techniques, employing multiple receive antennas, are considered. In order to quantify the performance improvements that may be achieved, a framework for the evaluation of these systems are constructed. Using this framework the BEP of all the space-time coding systems are derived analytically, and evaluated under identical propagation scenarios. The results presented show that the use of space-time turbo coded processing is an attractive solution since it can improve system performance significantly under conditions of multipath fading for both the uplink and downlink. It is shown that the two core areas of spatial processing and channel coding can be integrated in an optimum way to increase the capacity of existing cellular CDMA networks.
Thesis (DPhil (Electronic Engineering))--University of Pretoria, 2007.