Multi-user detection for frequency hopped networks and improved space-time codes for wireless communication

We develop a useful class of multi-user detectors for the frequency hopped networks. The successive interference canceler uses multi-user demodulation to mitigate the other user's interference and iterative multi-user decoding using the error correcting code for more reliable interference cancellation. To the best of our knowledge, this is the first reported multi-user detection work for the FH networks. Both hard and soft decision based methods are considered. An iterative decoding structure for the Reed Solomon codes is developed for the hard decision based approach and this is then extended for the soft decision based approach by employing a Block Turbo code formed by shortened RS codes. Space-time codes for multi-antenna systems have received great attention recently for supporting higher data rates. The space-time codes assume that the transmitter is completely ignorant of the channel. We show that in the presence of a feedback channel of low bandwidth, one can greatly improve the performance of the space-time codes by adaptively controlling the power, code and modulation. While this problem is well understood theoretically, it is in general, difficult to find the feedback statistic from the receiver to the transmitter based on the measured multi-antenna channel matrix. We show that the sufficient statistic for the space-time orthogonal block codes is a scalar and the optimal adaptation strategies are derived given the band-width of the feedback channel. The feedback based on this statistic gives excellent results and is shown to be useful for other space-time codes as well. The space-time codes are difficult to design and decode when the number of transmit antennas become large. We propose low complexity space-time codes which give excellent performance for large antenna systems which is better than the group interference cancellation methods proposed elsewhere. The iterative decoding of coded data streams from each transmit antenna with interleaving and power allocation forms a powerful code with simpler decoding. Further we explore the use of transmit diversity makes the system robust against jamming.