Minimax-Robust Filtering and Finite-Length Robust Predictors

Purchase on Springer.com

$29.95 / €24.95 / £19.95*

* Final gross prices may vary according to local VAT.

Get Access

Abstract

This paper consists basically of two parts. In the first part (Sections 2 and 3) , we consider some aspects of linear filtering under uncertainty about the spectral features of the relevant time series. The second part (Sections 4 and 5), considers a time domain approach to the general problem of filtering under probabilistic uncertainty and applies this general formulation to aspects of the linear filtering problem. Each of these approaches represents a different means for solving the minimax robust filtering problem, and both are applied to the specific problem of robust q-step prediction to illustrate the similarities of the two approaches.