Robust selection of clean swept-sine measurements in non-stationary noise

Prawda, Karolina; Schlecht, Sebastian J.; Välimäki, Vesa

doi:10.1121/10.0009915

Cited by 7 publications

(6 citation statements)

References 28 publications

(31 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This makes their energies equal, too, E h 2 i = E h 2 j . A similar relation holds for the noise energies under the assumption of stationary background noise, E u 2 i = E u 2 j [14]. When the conditions of time-invariance and stationary background noise are fulfilled, the correlation between two RIR measurements can be expressed by means of SNR:…”

Section: Rir Correlationmentioning

confidence: 91%

“…where h i (t) is the RIR itself and u i (t) is the stationary background noise of the i th measurement. Here, we treat the background noise as stationary since any nonstationary behavior is considered contamination which renders the measurement unfit for analysis [14,25]. Thus, RIRs containing non-stationary noise were not used in this study.…”

Section: Rir Correlationmentioning

confidence: 99%

“…We assume that the noise terms u i and u j , being random, are uncorrelated with the RIRs as well as with each other [14,25], yielding E[u i u j ] = 0, E[h j u i ] = 0, and E[h i u j ] = 0, and transforming Eq. ( 2) into…”

Section: Rir Correlationmentioning

confidence: 99%

“…In Eq. ( 8) only the parts regarding the RIR itself are calculated in the short windows, as short-term noise energy is hard to separate from the RIR [25,28]. Thus, E u 2 is computed as one number for the entire measurement, using the part of the signal where the RIR has already decayed below the noise floor level.…”

Section: Short-term Running Cross-correlationmentioning

confidence: 99%

“…The presence of transfer-function variation violates the assumption of a linear time-invariant system, leading to artifacts when noise-based excitation signals are used [1,4,6,12], and impairing sweep-based measurements, especially when using synchronous averaging [3,10,13,14]. This is critical when performing measurements in the ultrasound frequency range, such as scale-model measurements [15] as well as when multiple RIRs are compared, e.g., in scattering coefficient [16] and loudspeaker directivity measurements [17].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Time Variance in Measured Room Impulse Responses

Prawda,

Schlecht,

Välimäki

2022

Proceedings of the 10th Convention of the European Acoustics Association Forum Acusticum 2023

View full text Add to dashboard Cite

Time variance is unavoidable in room impulse response (RIR) measurements, as the atmospheric conditions fluctuations and air movement prevent any room from being perfectly steady. However, the effect of such changes on RIRs has received little attention so far, although it is known to cause energy loss when RIRs are averaged to enhance their signal-to-noise ratio. In this work, we focus on the influence of spatio-temporal fluctuations in the speed of sound on RIR energy loss. We introduce a new approach to examining the relationship between time-variance-related energy loss of averaged RIRs and the decrease in their cross-correlation over time. We show that even when the measurements are compensated for the global changes in the speed of sound, the small fluctuations still contribute to the RIR dissimilarity. The results also indicate that the RIRs correlation decreases with the growing time separation between measurements. It shows that the short-term running cross-correlation is a good metric for estimating the time-variance-related energy loss in RIRs. The research presented in this work is a step towards more robust acoustic measurements and compensation of time-variance-related artifacts in measured signals.

show abstract

Section: Rir Correlationmentioning

confidence: 91%

Section: Rir Correlationmentioning

confidence: 99%

Section: Rir Correlationmentioning

confidence: 99%

Section: Short-term Running Cross-correlationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Time Variance in Measured Room Impulse Responses

Prawda,

Schlecht,

Välimäki

2022

Proceedings of the 10th Convention of the European Acoustics Association Forum Acusticum 2023

View full text Add to dashboard Cite

show abstract

Calibrating the Sabine and Eyring formulas

Prawda

Schlecht

Välimäki

2022

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

Of the many available reverberation time prediction formulas, Sabine's and Eyring's equations are still widely used. The assumptions of homogeneity and isotropy of sound energy during the decay associated with those models are usually recognized as a reason for lack of agreement between predictions and measurements. At the same time, the inaccuracy in the estimation of the sound-absorption coefficient adds to the uncertainty of calculations. This paper shows that the error of incorrectly assumed sound absorption is more detrimental to the prediction precision than the inherent error in the formulas themselves. The proposed absorption calibration procedure reduces the differences between the measured and predicted reverberation time values, showing that an accuracy within ±10% from the target reverberation time values can be achieved regardless of the absorption distribution in a room. The paper also discusses the oft neglected air absorption of sound, which may introduce considerable bias to the measurement results. The need for an air-absorption compensation procedure is highlighted, and a method for the estimation of its parameters in octave bands is proposed and compared with other approaches. The results of this study provide justification for the use of the Sabine and Eyring formulas for reverberation time predictions.

show abstract

Sparse optimization for nonlinear group delay mode estimation

Liang

Ding

Jakobsson

et al. 2022

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

Nonlinear group delay signals with frequency-varying characteristics are common in a wide variety of fields, for instance, structural health monitoring and fault diagnosis. For such applications, the signal is composed of multiple modes, where each mode may overlap in the frequency-domain. The resulting decomposition and forming of time-frequency representations of the nonlinear group delay modes is a challenging task. In this study, the nonlinear group delay signal is modelled in the frequency-domain. Exploiting the sparsity of the signal, we present the nonlinear group delay mode estimation technique, which forms the demodulation dictionary from the group delay. This method can deal with crossed modes and transient impulse signals. Furthermore, an augmented alternating direction multiplier method is introduced to form an efficient implementation. Numerical simulations and experimental data analysis show the effectiveness and advantages of the proposed method. In addition, the included analysis of Lamb waves as well as of a bearing signal show the method's potential for structural health monitoring and fault diagnosis.

show abstract

Robust selection of clean swept-sine measurements in non-stationary noise

Cited by 7 publications

References 28 publications

Time Variance in Measured Room Impulse Responses

Time Variance in Measured Room Impulse Responses

Calibrating the Sabine and Eyring formulas

Sparse optimization for nonlinear group delay mode estimation

Contact Info

Product

Resources

About