A unified approach to discrete-time distributed parameter estimation by the least-squares method

Abstract

Optimal prediction, filtering, and smoothing for a discrete-time distributed parameter system are approached in a unified manner using the least-squares method. Based on an unibiased, minimum variance estimation error criterion, the optimal filter is derived. Then the optimal prediction and smoothing estimators are derived, including fixed-point, fixed-interval, and fixed-lag smoothers.

Additional details