Generalized suboptimum decoding for space-time codes in quasistatic flat fading channels

Abstract

This thesis work present a generalized version of principal ratio combining (PRC) [4], which is a near-optimum detection scheme for space-time codes in quasistatic flat fading environments. Quasistatic flat fading is assumed because frequency flat, slow Rayleigh fading may be considered the most critical and frequent disturbance in a wireless communications. Space-time coding, which integrates channel coding, modulation, and multiple transmit antennas, achieves higher data rates and provides diversity to combat fading at the same time. It was proved in [3] that the designed codes are optimal in terms of the tradeoff between diversity advantage, transmission rate, decoding complexity, and constellation size. When maximum-likelihood (ML) detection is employed at the receiver [42], the decoding complexity increases as the number of transmit and receive antennas increases. In [4], the proposed suboptimal detection algorithm is independent of the number of receive antennas resulting in some sacrifice in performance. But the performance penalty increases as the number of receive antennas increases. In this thesis, a more generalized version of suboptimum decoding scheme is proposed, which shows a flexible tradeoff between performance and decoding complexity. In the proposed scheme, receive antennas are optimally divided into K groups, and the PRC detection method is applied to each group. The metric of the generalized version of PRC is shown to include ML and PRC. When K$\geq$ 2, the problem becomes how to determine the number of elements in each group. The definitions on PI (performance index) are also proposed to find the optimum configuration of each group. In terms of performance index, Grouping Rules can be established. Correlated receive antennas case is also briefly reviewed. Then, simulation results are presented to justify the PI and the grouping rules.