System Identification
A Frequency Domain Approach

How to model a linear dynamic system from noisy data? This book presents a general approach to this problem, with both practical examples and theoretical discussions that give the reader a sound understanding of the subject and of the pitfalls that may occur on the road from raw data to validated model. The emphasis is on robust methods that can be used with a minimum of user interaction. Readers in many fields of engineering will gain knowledge about
  • choice of experimental setup - experiment design
  • automatic characterization of disturbing noise
  • generation of a good plant model
  • detection, qualification, and quantification of nonlinear distortions
  • improved model validation tools
and, from the theoretical side, about
  • system identification
  • interrelations between time and frequency domain approaches
  • stochastic properties of the estimators
  • stochastic analysis

System Identification: A Frequency Domain Approach is written for several levels. Practicing engineers and scientists who do not want to delve into mathematical details of proofs will find it useful, as will researchers who wish to learn more about the theoretical aspects of the proofs. Several of the introductory chapters are suitable for undergraduates. Each chapter starts with an abstract and ends with exercises, and examples are given throughout.

About the Authors

Rik Pintelon and Johan Schoukens are professors of electrical engineering at the Vrije Universiteit Brussel, Brussels, Belgium. They share research interests in system identification, signal processing, and measurement techniques. They are the co-authors of a software package with user friendly graphical user interface, Frequency Domain System Identification toolbox for MATLAB, which covers the methods discussed in this book.

View the Table of contents

©2006 • Vrije Universiteit Brussel • Dept. ELEC • Pleinlaan 2 • 1050 Elsene • Tel.: +32-2-629.27.67
• Fax: +32-2-629.28.50 • info.ELEC@vub.ac.be