Design of Gaussian inputs for Wiener model identification

Research output: Chapter in Book/Report/Conference proceedingConference paper

1 Citation (Scopus)

Abstract

We develop a tractable algorithms for finding the optimal power spectral density of the Gaussian input excitation for identifying a Wiener model. This problem is known as a difficult problem for two reasons. Firstly, the estimation accuracy depends on the higher order joint moments of the potentially infinitely many past samples of the input signal. In addition, the covariance matrix of the parameter estimates is thought to be a highly non-convex function of the power spectral density function. In this contribution we show that under Gaussian assumption it is possible to completely parameterize the set of all admissible information matrices with only a finite number of parameters. We present a convex algorithm to solve the D-optimal design problem. This idea can be extended further to design Gaussian mixture designs.
Original languageEnglish
Title of host publication17th IFAC Symposium on System Identification (SYSID 2015), Beijing, China, October 19-21, 2015
PublisherElsevier
Pages614-619
DOIs
Publication statusPublished - 19 Oct 2015
Event17th IFAC Symposium on System Identification (SYSID 2015) - Beijing, China
Duration: 19 Oct 201521 Oct 2015

Publication series

NameIFAC-PapersOnline
PublisherElsevier
Number28
Volume48
ISSN (Electronic)2405-8963

Conference

Conference17th IFAC Symposium on System Identification (SYSID 2015)
CountryChina
CityBeijing
Period19/10/1521/10/15

Keywords

  • identification
  • Design

Fingerprint

Dive into the research topics of 'Design of Gaussian inputs for Wiener model identification'. Together they form a unique fingerprint.

Cite this