On the inclusion of channel's time dependence in a hidden Markov model for blind channel estimation

We introduce several modifications to the Baum&Welch (BW) formulas used to reestimate the parameters of a hidden Markov model (HMM). The estimated parameter is the channel impulse response (CIR) of a communication system which is known to be time-varying. With these modifications, channel tracking properties of a BW-based algorithm are improved. The resulting algorithm is tested in a specific mobile radio environment (the GSM system), exhibiting good performance at expenses of higher computational complexity.