Kalman filtering for channel estimation in space-time coded systems

The paper considers channel estimation, in the context of space-time coding. Kalman filtering (KF), in its prediction form, is employed to track the fading channel in a multiple-antenna system, using orthogonal space-time codes and three different decoders. The system's performance is evaluated through computer simulations. The effect of the Doppler rate on the channel estimation and decoding scheme is investigated. The results show that the performance degrades as the Doppler frequency increases. Furthermore, the effect of channel estimation on the decoders' performance is examined. It is shown that when channel estimation errors cannot be kept sufficiently small, then a simple symbol by symbol decoder provides the same performance as a joint decoder.