Spectral Factorization and Spectral Estimation by Kalman Filtering
Date:
Abstract
In this talk I review some signal processing theory including common techniques of spectral estimation. Some shortcomings of these common techniques are exposed under curtain conditions. I then demonstrate how the Kalman filter may be employed in spectral factoring (which is not new) and then show how this can be extended to spectral estimation (which to the best of my knowledge is new). I end by reporting that the aforementioned shortcomings of certain spectral estimation technique are not shared by this new technique. Examples are frequently used to demonstrate concepts and one of the example timeseries is a solution to the Kuromoto-Sivashinsky equation.