Abstract

Merging layered space-time coding (LST), which provides high data rate transmission with space-time block coding (STBC) provides transmit diversity and improved performance to wireless communication systems. This architecture is known as multilayered space-time coding (MLSTC). In this thesis we propose a decreased complexity minimum mean squared error (MMSE) detector based on the sorted QR decomposition (SQRD), intended for the detection of multi-user STBC systems. The proposed decoder has been proven to enhance the overall performance of the MLSTC architecture. It offers a lower computational complexity than that offered by conventional detectors such as group interference cancellation (GIC) detectors. In addition, the number of receive antennas required to perform detection is smaller than that required by the GIC detector. Further, in order for all groups, which can be viewed as individual users, to benefit from a similar performance, unequal power allocation strategy will be employed to the system. We also concatenate OFDM to the MLSTC scheme in order to combat a frequency selective channel. We then extend this scheme to space-frequency-time (SFT) codes in order to extract the maximum frequency diversity available from the channel. In this thesis, we also propose a new transceiver architecture for multiple-input multiple-output (MIMO) systems. The proposed scheme borrows ideas from the MLSTC and threaded space-time coding (TSTC) schemes in an effort to maximize the diversity order while maintaining a low complexity detection. Specifically, the proposed scheme has a structure similar to that of the MLSTC while it employs a spatial interleaver (SI) in front of the MLSTC encoder. We analyze the proposed scheme where we derive an upper bound on the bit error rate and derive the diversity-multiplexing gain trade-off curve. We present several examples through which we demonstrate the superiority of the proposed scheme over existing schemes.