Real Time Signal Processing for Multi-Antenna Systems and experimental Verification on a reconfigurable Hardware Test-bed von Diplom-Physiker

Multiple-input multiple-output (MIMO) systems will be applied in wireless communications in order to increase the system performance, spectral efficiency, and link reliability. Theoretically, the channel capacity of MIMO systems can grow linearly with the number of transmit and receive antennas. An important performance metric beneath capacity is the normalized mean square error (MSE) under the assumption of optimal linear reception. Both performance measures depend on the properties of the MIMO channel as well as on the available channel state information (CSI) at the transmitter. In this thesis, we derive optimum transmission strategies of singleand multiuser MIMO systems with respect to the different types of CSI at the transmit and receive side. The optimization is taken under the assumption of the MSE as the objective function. The results differ therefore from those known for ergodic capacity optimizations. We start with a derivation of the optimum transmission strategy for the single user MIMO scenario with perfect channel knowledge at the receiver and no or full channel knowledge at the transmitter. Furthermore, we derive the optimum transmission strategy for a multiple access channel (MAC) with only one antenna per user and several antennas at the base station. We look very close on how the SNR gap approximation, often used for bit-loading approaches, affects the behavior of the sum rate functional which has to be maximized. To bridge from the well studied multi-antenna theory towards implementations on real-world systems some basic and practical considerations will be made for wireless MIMO systems. Here, the emphasis is put on antenna configurations with respect to the number and kind of antennas at each side of the link. Furthermore we analyze the effect of a line-of-sight on system relevant parameters as the rank of the transmission channel, capacity and achievable bit-error-rates. The MIMO broadcast channel as the duality equivalent of the MIMO MAC is investigated for a multi-antenna base station and several distributed users. When CSI is available at the BS appropriate pre-coding techniques can be applied. We will look into SVD-MIMO transmission and linear and non-linear pre-coding techniques. Further emphasis is put on a comparison towards the necessary transmit power needed for transmit pre-coding. In reality the systems can suffer from performance degradation caused by e.g. channel estimation errors, a limited transmitter dynamics or co-channel interference. The impact of each degradation factor will be evaluated and strategies to combat or limit the undesired effects will be proposed. To achieve optimum system performance adaptive transmission is an important issue. The aim is to adapt the data transmission to the actual channel realization, thus avoiding transmission over bad channels. We will extend channel aware bit-loading with discrete modulation alphabets to multi-user scheduling policies. In a cross-layer approach the optimization considers the instantaneous channel state and quality of service parameters e.g. the data queues of all users as well. This approach allows high cell throughput and stable data queues at the MTs or the BS, both