Subspace-based (semi-) blind channel estimation for block precoded space-time OFDM

Space time coding has by now been well documented as an attractive means of achieving high data rate transmissions with diversity and coding gains, provided that the underlying propagation channels can be accounted for. We rely on redundant linear precoding to develop a (semi-)blind channel estimation algorithm for space time (ST) orthogonal frequency division multiplexing (OFDM) transmissions with Alamouti's (see IEEE J. Select. Areas Commun., vol.16, p.1451-58, Oct. 1998) block code applied on each subcarrier. We establish that multichannel identifiability is guaranteed up to one or two scalar ambiguities, regardless of the channel zero locations and the underlying signal constellations, when distinct or identical precoders are employed for even and odd indexed symbol blocks. With known pilots inserted either before or after precoding, we resolve the residual scalar ambiguities and show that distinct precoders require half the number of pilots than identical precoders to achieve the same channel estimation accuracy. Simulation results confirm our theoretical analysis and illustrate that the proposed semi-blind algorithm is capable of tracking slow channel variations and improving the overall system performance relative to competing differential ST alternatives.