Subspace-based signal analysis using singular value decomposition

Veen, Alle-Jan van der; Deprettere, E.F.; Swindlehurst, A. Lee

doi:10.1109/5.237536

Cited by 282 publications

(130 citation statements)

References 101 publications

Supporting

Mentioning

129

Contrasting

Order By: Relevance

“…So-called 'subspace methods' have proved extremely successful at black-box identification of multivariable linear state-space models from data, and are consequently of great current interest [6,7,8,10,11,13]. In this paper we show that stability of the models obtained can be guaranteed very simply and in a uniform manner for many of the published variations of subspace methods.…”

Section: Introductionmentioning

confidence: 85%

Guaranteed stability with subspace methods

Maciejowski

1995

Systems & Control Letters

View full text Add to dashboard Cite

We show how stability of models can be guaranteed when using the class of identification algorithms which have become known as 'subspace methods'. In many of these methods the 'A' matrix is obtained (or can be obtained) as the product of a shifted matrix with a pseudo-inverse. We show that whenever the shifted matrix is formed by introducing one block of zeros in the appropriate position, then a stable model results. The cost of this is some (possibly large) distortion of the results, but in some applications that is outweighed by the advantage of guaranteed stability.

show abstract

Section: Introductionmentioning

confidence: 85%

Guaranteed stability with subspace methods

Maciejowski

1995

Systems & Control Letters

View full text Add to dashboard Cite

show abstract

“…The matrices A, B, and C in Eq. (1) were calculated using the SVD method [9], [10]. The transfer function G(z) was calculated as…”

Section: Discussionmentioning

confidence: 99%

System Identification of Mechanomyogram at Various Levels of Motor Unit Recruitment

Uchiyama

Tamura

2014

SICE Journal of Control, Measurement, and System Integration

View full text Add to dashboard Cite

: The mechanomyogram from a single motor unit and the induced mechanomyogram at various levels of recruitment were measured with an acceleration sensor. The transfer functions between motor unit action potential (or electrical stimulation) and the mechanomyogram were identified using the singular value decomposition method. The purpose of this study is to clarify how the model order of the transfer function depends on the recruitment level. The second-to tenth-order transfer functions were calculated, and the difference between the observed and the estimated mechanomyograms using the transfer function, the fitness, was calculated. The relationship between the model order and the fitness was tested using the Holm-Bonferroni multiple comparison. At low levels (single motor unit, 20, and 40%) of recruitment, there were significant differences between the fourth-and higher-order models, but there were no significant differences between the fifth-and higher-order models. In contrast, at high levels (60, 80, and 100%) of recruitment, the fourth-order model did not show significant differences between the fifth-or higher-order models. As a result, the fifthand fourth-order models were appropriate at low and high recruitment levels, respectively. The differences in the order might be caused by interactions between active and resting motor units.

show abstract

“…The nonlinearity related to signal pole estimation can be circumvented by using the so-called HTLS algorithm by Van Huffel et al [29], which is a total least squares (TLS) based variant of Kung et al's original state space algorithm [28]. These algorithms belong to the class of single shift-invariant methods within the set of subspace-based signal analysis algorithms [30]. The HTLS algorithm is not immediately suited for solving the weighted problem in (4).…”

Section: Estimation Of Perceptually Relevant Esm Parametersmentioning

confidence: 99%

“…For the ESM, the parameter estimation schemes can roughly be divided into two main groups: analysis-by-synthesis schemes such as the matching pursuit (MP) based algorithms described in [16], [27] and subspace-based schemes (e.g., [28]- [30]). While some work has been done for extracting perceptually relevant sinusoids using MP based schemes (e.g., [31], [32]), less efforts have been directed toward estimating perceptually relevant ESM parameters using subspace techniques [21].…”

mentioning

confidence: 99%

A Perceptual Subspace Approach for Modeling of Speech and Audio Signals With Damped Sinusoids

Jensen

Heusdens

Jensen

2004

IEEE Trans. Speech Audio Process.

View full text Add to dashboard Cite

Abstract-The problem of modeling a signal segment as a sum of exponentially damped sinusoidal components arises in many different application areas, including speech and audio processing. Often, model parameters are estimated using subspace based techniques which arrange the input signal in a structured matrix and exploit the so-called shift-invariance property related to certain vector spaces of the input matrix. A problem with this class of estimation algorithms, when used for speech and audio processing, is that the perceptual importance of the sinusoidal components is not taken into account. In this work we propose a solution to this problem. In particular, we show how to combine well-known subspace based estimation techniques with a recently developed perceptual distortion measure, in order to obtain an algorithm for extracting perceptually relevant model components. In analysis-synthesis experiments with wideband audio signals, objective and subjective evaluations show that the proposed algorithm improves perceived signal quality considerable over traditional subspace based analysis methods.Index Terms-Complex exponentials, perceptually relevant sinusoids, psycho-acoustical distortion measure, sinusoidal modeling, speech and audio processing, subspace-based signal analysis.

show abstract

Subspace-based signal analysis using singular value decomposition

Cited by 282 publications

References 101 publications

Guaranteed stability with subspace methods

Guaranteed stability with subspace methods

System Identification of Mechanomyogram at Various Levels of Motor Unit Recruitment

A Perceptual Subspace Approach for Modeling of Speech and Audio Signals With Damped Sinusoids

Contact Info

Product

Resources

About