An iterative algorithm for inverse filtering

If the Z transform of a filter has some zero points which are on or very close to the unit circle, the inverse filter will extend for many significant terms instead of vanishing in a few terms, and a tremendous truncation error will result. An iterative algorithm has been found to handle this zero-point problem. This paper describes the iterative algorithm and shows mathematically that it converges and that the limit is equal to the original signal. Numerical results indicate that the restoration of the original signal is reasonably good.