Semiblind ML OSTBC-OFDM Detection in Block Fading Channels

This paper presents a semiblind maximum-likelihood (ML) detector for the orthogonal space-time block coded orthogonal frequency division multiplexing (OSTBC-OFDM) system. Many existing blind/ semiblind OSTBC-OFDM receivers typically require that the channel is static over a multitude of OSTBC-OFDM blocks. The proposed method is specifically for detection over one OSTBC-OFDM block only, and hence is well suited to block fading channels. The presented identifiability analysis shows that the data can be uniquely identified in a probability one sense by using one pilot code only, in contrast to the pilot-based least-squares channel estimator which requires at least L pilot codes where L is the channel length. Simulation examples are then presented to show the efficacy of the proposed detector.