Model Based Estimation

This monograph provides a tool-set for hybrid estimation that can successfully monitor the behavior of complex artifacts with a large number of possible operational and failure modes such as production plants, automotive or aeronautic systems, and autonomous robots. For this purpose, ideas from the fields of System Theory and Artificial Intelligence are taken and hybrid estimation is reformulated as a search problem.

An accessible introduction to indirect estimation methods, both traditional and model–based. Readers will also find the latest methods for measuring the variability of the estimates as well as the techniques for model validation.

The purpose of this book is to provide graduate students and practitioners with traditional methods and more recent results for model-based approaches in signal processing.
Firstly, discrete-time linear models such as AR, MA and ARMA models, their properties and their limitations are introduced. In addition, sinusoidal models are addressed.
Secondly, estimation approaches based on least squares methods and instrumental variable techniques are presented.
Finally, the book deals with optimal filters, i.e. Wiener and Kalman filtering, and adaptive filters such as the RLS, the LMS and their variants.

The purpose of this book is to provide graduate students and practitioners with traditional methods and more recent results for model-based approaches in signal processing.