Combined interference cancellation and ML decoding of orthogonal space-tme block code over time-selective fading channels

We propose a novel simplified maximum-likelihood (ML) decoder of orthogonal space-time block code (OSTBC) [1] for wireless communications over time-selective fading channels. When the channel varies from symbol to symbol, there exists inter-transmit-antenna interference (ITAI). The proposed decoder can recognize and completely cancel ITAI and provide excellent diversity advantage of the OSTBC. When the channel does not vary from symbol to symbol, the proposed simplified ML decoder becomes the optimum ML decoder for OSTBC. We derive a theoretical upper bound for the bit error probability of the proposed decoder. Comprehensive simulation and theoretical results show that the proposed decoder does not exhibit error floors at high signal-to-noise ratios like the currently proposed decoder [2] and [4]. Furthermore, the proposed decoder significantly outperforms the currently proposed decoder at the same decoding complexity.