Performance of an adaptive Kalman equaliser on time variant multipath channels

This paper develops an adaptive equaliser which utilises the Kalman filtering to reconstruct the transmitted sequence in time variant environments. The adaptive Kalman equaliser(AKE) addressed by Mulgrew and Cowan is modified by adopting a channel estimator, coefficients of which are updated by a gradient algorithm with fading memory prediction. By computer simulations, the performance of the AKE is investigated, and shown to be superior to that of the decision feedback equaliser(DFE) involving the adaptation of recursive least squares(RLS) algorithm in the case of a second order Markov communication channel model.