In the area of MIMO wireless systems employing multiple antennas at the transmitter and the receiver we analyse space-time codes for improved link reliability. The emphasis is on systems utilizing incomplete channel state information at the transmitter to select the space-time code and the active antennas. Furthermore, we develop computationally very efficient methods for data detection in MIMO systems using spatial multiplexing. Here we focus on the approximate computation of reliability information (soft values) for the transmit data which needs to be provided to the channel decoder. The use of this soft information in the context of iterative (turbo) receivers is a further aspect of our ongoing research. For multi-user systems we develop techniques for transmit precoding which allow that the individual users recover their data in a simple and non-cooperative fashion. To this end, we use fundamental mathematical results and tools from algorithmic number theory.
Our information theoretic research deals with the application of methods from information geometry for the analysis and improvement of iterative algorithms like those used for the calculation of channel capacity and in turbo receivers. We further analyse the maximum achievable data rates in systems using incorrect channel state information at the receiver (this is particular relevant for wireless communication systems). To this end, we developed a generalization of the information theoretic notion of typical sequences that features enhanced robustness against channel estimation errors.