Simulation of homogeneous two-dimensional random fields: Part II-MA and ARMA models

Pol D. Spanos, Marc Mignolet

Research output: Contribution to journalArticle

26 Citations (Scopus)

Abstract

Alternatively to the autoregressive (AR) models examined in Part I, the determination of moving average (MA) algorithms for simulating realizations of two-dimensional random fields with a specified (target) power spectrum is presented. First, the mathematical form of these models is addressed by considering infinitevariate vector processes of an appropriate spectral matrix. Next, the MA parameters are determined by relying on the maximization of an energy-like quantity. Then, a technique is formulated to derive an autoregressive moving average (ARMA) simulation algorithm from a prior MA approximation by relying on the minimization of frequency domain errors. Finally, these procedures are critically assessed and an example of application is presented.

Original languageEnglish (US)
Pages (from-to)S270-S277
JournalJournal of Applied Mechanics, Transactions ASME
Volume59
Issue number2
DOIs
StatePublished - 1992

Fingerprint

autoregressive moving average
Power spectrum
simulation
power spectra
optimization
matrices
approximation
energy

ASJC Scopus subject areas

  • Condensed Matter Physics
  • Mechanics of Materials
  • Mechanical Engineering

Cite this

Simulation of homogeneous two-dimensional random fields : Part II-MA and ARMA models. / Spanos, Pol D.; Mignolet, Marc.

In: Journal of Applied Mechanics, Transactions ASME, Vol. 59, No. 2, 1992, p. S270-S277.

Research output: Contribution to journalArticle

@article{d1ba3ffe15964a3c82b3ac2875d8a7ea,
title = "Simulation of homogeneous two-dimensional random fields: Part II-MA and ARMA models",
abstract = "Alternatively to the autoregressive (AR) models examined in Part I, the determination of moving average (MA) algorithms for simulating realizations of two-dimensional random fields with a specified (target) power spectrum is presented. First, the mathematical form of these models is addressed by considering infinitevariate vector processes of an appropriate spectral matrix. Next, the MA parameters are determined by relying on the maximization of an energy-like quantity. Then, a technique is formulated to derive an autoregressive moving average (ARMA) simulation algorithm from a prior MA approximation by relying on the minimization of frequency domain errors. Finally, these procedures are critically assessed and an example of application is presented.",
author = "Spanos, {Pol D.} and Marc Mignolet",
year = "1992",
doi = "10.1115/1.2899500",
language = "English (US)",
volume = "59",
pages = "S270--S277",
journal = "Journal of Applied Mechanics, Transactions ASME",
issn = "0021-8936",
publisher = "American Society of Mechanical Engineers(ASME)",
number = "2",

}

TY - JOUR

T1 - Simulation of homogeneous two-dimensional random fields

T2 - Part II-MA and ARMA models

AU - Spanos, Pol D.

AU - Mignolet, Marc

PY - 1992

Y1 - 1992

N2 - Alternatively to the autoregressive (AR) models examined in Part I, the determination of moving average (MA) algorithms for simulating realizations of two-dimensional random fields with a specified (target) power spectrum is presented. First, the mathematical form of these models is addressed by considering infinitevariate vector processes of an appropriate spectral matrix. Next, the MA parameters are determined by relying on the maximization of an energy-like quantity. Then, a technique is formulated to derive an autoregressive moving average (ARMA) simulation algorithm from a prior MA approximation by relying on the minimization of frequency domain errors. Finally, these procedures are critically assessed and an example of application is presented.

AB - Alternatively to the autoregressive (AR) models examined in Part I, the determination of moving average (MA) algorithms for simulating realizations of two-dimensional random fields with a specified (target) power spectrum is presented. First, the mathematical form of these models is addressed by considering infinitevariate vector processes of an appropriate spectral matrix. Next, the MA parameters are determined by relying on the maximization of an energy-like quantity. Then, a technique is formulated to derive an autoregressive moving average (ARMA) simulation algorithm from a prior MA approximation by relying on the minimization of frequency domain errors. Finally, these procedures are critically assessed and an example of application is presented.

UR - http://www.scopus.com/inward/record.url?scp=84988280239&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84988280239&partnerID=8YFLogxK

U2 - 10.1115/1.2899500

DO - 10.1115/1.2899500

M3 - Article

AN - SCOPUS:84988280239

VL - 59

SP - S270-S277

JO - Journal of Applied Mechanics, Transactions ASME

JF - Journal of Applied Mechanics, Transactions ASME

SN - 0021-8936

IS - 2

ER -