Adaptive modified covariance algorithms for spectral analysis

Kyriakos Kitsios, Andreas Spanias, Bruno Welfert

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

An optimum block modified covariance algorithm is developed for computing time-varying autoregressive (AR) parameters. The method presented here differs from those presented previously in that it derives optimally selected time-varying convergence factors such that the block mean square error is minimized from one iteration to the next. In particular, the algorithm developed here, called block modified covariance algorithm with individual adaptation of parameters (BMCAI), uses individual time-varying convergence factors computed using modified covariance matrix approximations along with the Gauss-Seidel method. Even though the BMCAI is gradient based, it retains the attractive spectral matching properties of fixed-window least-squares modified covariance algorithms while at the same time providing capabilities for time-varying spectral estimation.

Original languageEnglish (US)
Pages (from-to)715-720
Number of pages6
JournalSignal Processing
Volume82
Issue number5
DOIs
StatePublished - May 2002

Fingerprint

Spectrum analysis
Covariance matrix
Mean square error

Keywords

  • Modified covariance algorithms
  • Spectral analysis

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Adaptive modified covariance algorithms for spectral analysis. / Kitsios, Kyriakos; Spanias, Andreas; Welfert, Bruno.

In: Signal Processing, Vol. 82, No. 5, 05.2002, p. 715-720.

Research output: Contribution to journalArticle

@article{ca9b851b502f4c4783f3043c6fc6a931,
title = "Adaptive modified covariance algorithms for spectral analysis",
abstract = "An optimum block modified covariance algorithm is developed for computing time-varying autoregressive (AR) parameters. The method presented here differs from those presented previously in that it derives optimally selected time-varying convergence factors such that the block mean square error is minimized from one iteration to the next. In particular, the algorithm developed here, called block modified covariance algorithm with individual adaptation of parameters (BMCAI), uses individual time-varying convergence factors computed using modified covariance matrix approximations along with the Gauss-Seidel method. Even though the BMCAI is gradient based, it retains the attractive spectral matching properties of fixed-window least-squares modified covariance algorithms while at the same time providing capabilities for time-varying spectral estimation.",
keywords = "Modified covariance algorithms, Spectral analysis",
author = "Kyriakos Kitsios and Andreas Spanias and Bruno Welfert",
year = "2002",
month = "5",
doi = "10.1016/S0165-1684(01)00178-5",
language = "English (US)",
volume = "82",
pages = "715--720",
journal = "Signal Processing",
issn = "0165-1684",
publisher = "Elsevier",
number = "5",

}

TY - JOUR

T1 - Adaptive modified covariance algorithms for spectral analysis

AU - Kitsios, Kyriakos

AU - Spanias, Andreas

AU - Welfert, Bruno

PY - 2002/5

Y1 - 2002/5

N2 - An optimum block modified covariance algorithm is developed for computing time-varying autoregressive (AR) parameters. The method presented here differs from those presented previously in that it derives optimally selected time-varying convergence factors such that the block mean square error is minimized from one iteration to the next. In particular, the algorithm developed here, called block modified covariance algorithm with individual adaptation of parameters (BMCAI), uses individual time-varying convergence factors computed using modified covariance matrix approximations along with the Gauss-Seidel method. Even though the BMCAI is gradient based, it retains the attractive spectral matching properties of fixed-window least-squares modified covariance algorithms while at the same time providing capabilities for time-varying spectral estimation.

AB - An optimum block modified covariance algorithm is developed for computing time-varying autoregressive (AR) parameters. The method presented here differs from those presented previously in that it derives optimally selected time-varying convergence factors such that the block mean square error is minimized from one iteration to the next. In particular, the algorithm developed here, called block modified covariance algorithm with individual adaptation of parameters (BMCAI), uses individual time-varying convergence factors computed using modified covariance matrix approximations along with the Gauss-Seidel method. Even though the BMCAI is gradient based, it retains the attractive spectral matching properties of fixed-window least-squares modified covariance algorithms while at the same time providing capabilities for time-varying spectral estimation.

KW - Modified covariance algorithms

KW - Spectral analysis

UR - http://www.scopus.com/inward/record.url?scp=0036566142&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036566142&partnerID=8YFLogxK

U2 - 10.1016/S0165-1684(01)00178-5

DO - 10.1016/S0165-1684(01)00178-5

M3 - Article

AN - SCOPUS:0036566142

VL - 82

SP - 715

EP - 720

JO - Signal Processing

JF - Signal Processing

SN - 0165-1684

IS - 5

ER -