Kalman filter of channel modes in time-varying wireless systems

In mobile communications the movement of the users makes the propagation channel to be time-varying. Algorithms that track channel variations have to trade between complexity and accuracy. Since the second-order statistic of time-varying channels is stationary, estimation of the channel can be reduced to track a set of r uncorrelated parameters. Based on this decomposition, in this paper we propose to simplify the optimum Kalman filter (KF) by tracking the r channel modes separately and by using the steady-state solution of the KF gain.