Spectral estimation in frequency-domain by subspace techniques

Akcay H.

SIGNAL PROCESSING, vol.101, pp.204-217, 2014 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 101
  • Publication Date: 2014
  • Doi Number: 10.1016/j.sigpro.2014.02.015
  • Journal Name: SIGNAL PROCESSING
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.204-217
  • Keywords: Identification, Spectrum, Subspace method, Periodogram, Welch estimator, Dynamic factor model, IDENTIFICATION, ALGORITHMS, MODELS, INFORMATION, VARIABLES, MATRICES
  • Anadolu University Affiliated: Yes

Abstract

In this paper, frequency-domain subspace-based algorithms are proposed to estimate discrete-time cross-power spectral density (cross-PSD) and auto-power spectral density (auto-PSD) matrices from samples of the Welch cross-PSD and auto-PSD estimators on uniform grids of frequencies. The proposed algorithms are shown to be strongly consistent. A link between the well-known time-domain covariance-based estimation methods and the frequency-domain realization-based methods of this paper is also established. The consistency of the proposed algorithms is not expected a priori since the periodograms are not consistent spectrum estimators even if the true power spectrum is rational. The algorithms are tested on numerical examples and a real-life application example concerned with the modeling of acoustic spectra for fault detection in induction motors. (C) 2014 Elsevier B.V. All rights reserved.