Efficient estimators for an HF radio link

A technique for the estimation of the sampled impulse response of an HF radio link was developed several years ago and was shown by computer simulation tests to have a substantial potential advantage in performance over more conventional systems. A number of further developments have led to new channel estimators. In some of these, a Kalman filter is incorporated into the system in such a way that the filter operates on only a few variable quantities and is therefore considerably less complex than a conventional Kalman filter. Three different Kalman filters are studied, two of these being designed for a channel varying linearly (at a constant rate) with time, and the third being a conventional Kalman filter that is designed for a time-invariant or very slowly time-varying channel. All of these employ an exponential window and therefore operate with a fading memory. A description is given of the various estimators, the results of a series of computer simulation tests are presented, and the accuracies of the channel estimates given by the different systems are compared.<
>