Joint channel estimation and symbol detection for SFBC-OFDM systems via the EM algorithm

Joint channel estimation and symbol detection for space-frequency block coded OFDM (SFBC-OFDM) systems is considered. The expectation-maximization (EM) algorithm is employed to devise both soft-in-hard-out (SIHO) maximum likelihood (ML) iterative receivers and soft-in-soft-out (SISO) maximum a posteriori (MAP) iterative receivers for SFBC-OFDM systems. Besides the low computational complexity and fast convergence, the EM-based iterative receivers also have the ability of tracing the variation of fast fading channels, and thus provide quite satisfying performance and bandwidth efficiency even in high-vehicular-speed environment. Numerical results are provided to corroborate the theoretical designs.