Antenna Selection for MIMO Systems with Space - Time Coding

Multiple-Input multiple-output (MIMO) technology is capable of enhancing capacity and coverage of wireless links. Implementation of MIMO devices is intricate because of the increased number of radio frequency chains (one for each antenna element). Antenna selection is one attractive approach to mitigate this requirement. It only utilizes an optimal subset of all available antennas, adapting it to the current channel state in order to minimize the performance loss compared to the full complexity system. This diploma thesis is dedicated to the analysis of space-time block coded MIMO transmission systems with antenna selection. The orthogonal Alamouti space-time block coding scheme as well as an extended non-orthogonal variant of it are studied in detail, the performance of the latter is evaluated using low-complexity linear zero-forcing receivers. The respective exact instantaneous bit error ratio expressions are acquired, providing the basis for the formulation of optimal selection criteria for the considered non-frequency selective, quasi-static MIMO channel. All three possible variants namely receive, transmit and combined selection are compared to each other and the results of numerical simulations are presented. Achievable diversity gains as well as changes in average signal to noise ratio (SNR) at the receiver are given. For the case of the non-orthogonal space-time block coding scheme the performance figures using the optimal criterion are compared to the results for a given set of sub-optimal, generally less complex selection algorithms. It will be shown that the simple selection procedures that are able to maximize performance for orthogonal space-time block codes only deliver moderate selection gains when applied to non-orthogonal codes. Obviously, the usability of antenna selection largely depends on the statistical characteristics of the MIMO channel. This motivated the execution of additional simulations using a simple stochastic channel-correlation model. The hereby obtained results allow to assess the performance deterioration when the MIMO channel is not spatially uncorrelated anymore.