Identification of Volterra Kernels using interpolation

József G. Németh, I. Kollár, Johan Schoukens

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

This paper presents a new method for the identification of frequency-domain Volterra kernels. Since the nonlinear kernels often play a secondary role compared to the dominant, linear component of the system, it is worth establishing a balance between the degree of liberty of these components and their effect on the overall accuracy of the model. This is necessary in order to reduce the model complexity, hence the required measurement length. Based on the assumption that frequency-domain kernels are locally smooth, the kernel surfaces can be approximated by interpolation techniques, thus reducing the complexity of the model. Similarly to the unreduced (Volterra) model, this smaller model is also i) linear in the unknowns; ii) only locally sensitive to its parameters; and iii) free of structural assumptions about the system. The parameter estimation boils down to solving a linear system of equations in the least-squares (LS) sense. The design of the interpolation scheme is described and the performance of the approximation is analyzed and illustrated by simulation. The algorithm allows a significant saving in measurement time compared to other kernel estimation methods.

Original languageEnglish
Pages (from-to)770-775
Number of pages6
JournalIEEE Transactions on Instrumentation and Measurement
Volume51
Issue number4
DOIs
Publication statusPublished - Aug 2002

Fingerprint

interpolation
Interpolation
linear systems
Time measurement
Parameter estimation
Linear systems
time measurement
approximation
simulation

Keywords

  • B-spline
  • Interpolation
  • Nonlinear system
  • Random multisine
  • System identification
  • Volterra kernel
  • Volterra series

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Instrumentation

Cite this

Identification of Volterra Kernels using interpolation. / Németh, József G.; Kollár, I.; Schoukens, Johan.

In: IEEE Transactions on Instrumentation and Measurement, Vol. 51, No. 4, 08.2002, p. 770-775.

Research output: Contribution to journalArticle

Németh, József G. ; Kollár, I. ; Schoukens, Johan. / Identification of Volterra Kernels using interpolation. In: IEEE Transactions on Instrumentation and Measurement. 2002 ; Vol. 51, No. 4. pp. 770-775.
@article{3ceae3b9c3f7468fa8accc3b5293dad2,
title = "Identification of Volterra Kernels using interpolation",
abstract = "This paper presents a new method for the identification of frequency-domain Volterra kernels. Since the nonlinear kernels often play a secondary role compared to the dominant, linear component of the system, it is worth establishing a balance between the degree of liberty of these components and their effect on the overall accuracy of the model. This is necessary in order to reduce the model complexity, hence the required measurement length. Based on the assumption that frequency-domain kernels are locally smooth, the kernel surfaces can be approximated by interpolation techniques, thus reducing the complexity of the model. Similarly to the unreduced (Volterra) model, this smaller model is also i) linear in the unknowns; ii) only locally sensitive to its parameters; and iii) free of structural assumptions about the system. The parameter estimation boils down to solving a linear system of equations in the least-squares (LS) sense. The design of the interpolation scheme is described and the performance of the approximation is analyzed and illustrated by simulation. The algorithm allows a significant saving in measurement time compared to other kernel estimation methods.",
keywords = "B-spline, Interpolation, Nonlinear system, Random multisine, System identification, Volterra kernel, Volterra series",
author = "N{\'e}meth, {J{\'o}zsef G.} and I. Koll{\'a}r and Johan Schoukens",
year = "2002",
month = "8",
doi = "10.1109/TIM.2002.803301",
language = "English",
volume = "51",
pages = "770--775",
journal = "IEEE Transactions on Instrumentation and Measurement",
issn = "0018-9456",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "4",

}

TY - JOUR

T1 - Identification of Volterra Kernels using interpolation

AU - Németh, József G.

AU - Kollár, I.

AU - Schoukens, Johan

PY - 2002/8

Y1 - 2002/8

N2 - This paper presents a new method for the identification of frequency-domain Volterra kernels. Since the nonlinear kernels often play a secondary role compared to the dominant, linear component of the system, it is worth establishing a balance between the degree of liberty of these components and their effect on the overall accuracy of the model. This is necessary in order to reduce the model complexity, hence the required measurement length. Based on the assumption that frequency-domain kernels are locally smooth, the kernel surfaces can be approximated by interpolation techniques, thus reducing the complexity of the model. Similarly to the unreduced (Volterra) model, this smaller model is also i) linear in the unknowns; ii) only locally sensitive to its parameters; and iii) free of structural assumptions about the system. The parameter estimation boils down to solving a linear system of equations in the least-squares (LS) sense. The design of the interpolation scheme is described and the performance of the approximation is analyzed and illustrated by simulation. The algorithm allows a significant saving in measurement time compared to other kernel estimation methods.

AB - This paper presents a new method for the identification of frequency-domain Volterra kernels. Since the nonlinear kernels often play a secondary role compared to the dominant, linear component of the system, it is worth establishing a balance between the degree of liberty of these components and their effect on the overall accuracy of the model. This is necessary in order to reduce the model complexity, hence the required measurement length. Based on the assumption that frequency-domain kernels are locally smooth, the kernel surfaces can be approximated by interpolation techniques, thus reducing the complexity of the model. Similarly to the unreduced (Volterra) model, this smaller model is also i) linear in the unknowns; ii) only locally sensitive to its parameters; and iii) free of structural assumptions about the system. The parameter estimation boils down to solving a linear system of equations in the least-squares (LS) sense. The design of the interpolation scheme is described and the performance of the approximation is analyzed and illustrated by simulation. The algorithm allows a significant saving in measurement time compared to other kernel estimation methods.

KW - B-spline

KW - Interpolation

KW - Nonlinear system

KW - Random multisine

KW - System identification

KW - Volterra kernel

KW - Volterra series

UR - http://www.scopus.com/inward/record.url?scp=0036703614&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036703614&partnerID=8YFLogxK

U2 - 10.1109/TIM.2002.803301

DO - 10.1109/TIM.2002.803301

M3 - Article

VL - 51

SP - 770

EP - 775

JO - IEEE Transactions on Instrumentation and Measurement

JF - IEEE Transactions on Instrumentation and Measurement

SN - 0018-9456

IS - 4

ER -