First- and Second-level Bayesian Inference of Flow Resistivity of Sound Absorber and Room’s Influence

Sang-Hyeon Choi; Ikjin Lee; Cheol-Ho Jeong

First- and Second-level Bayesian Inference of Flow Resistivity of Sound Absorber and Room’s Influence

Sang-Hyeon Choi
, Ikjin Lee
, Cheol-Ho Jeong

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

219 Downloads (Orbit)

Abstract

Sabine absorption coefficient is a widely used one deduced from reverberation time measurements via the Sabine equation. First- and second-level Bayesian analysis are used to estimate the flow resistivity of a sound absorber and the
influences of the test chambers from Sabine absorption coefficients measured in 13 different reverberation chambers. The first-level Bayesian analysis is more general than the second-level Bayesian analysis. Sharper posterior distribution can be acquired by the second-level Bayesian analysis than the one by the first-level Bayesian analysis because more data are used to set more reliable prior distribution. The estimated room’s influences by the first- and the second-level Bayesian analyses are similar to the estimated results by the mean absolute error minimization.

Original language	English
Title of host publication	Winter meeting proceedings of the Korean Society of Mechanical Engineers
Number of pages	4
Publication date	2016
Publication status	Published - 2016
Event	Winter Meeting proceedings of the Korean Society of Mechanical Engineers - Kangwon, Korea, Republic of Duration: 14 Dec 2016 → 16 Dec 2016

Conference

Conference	Winter Meeting proceedings of the Korean Society of Mechanical Engineers
Country/Territory	Korea, Republic of
City	Kangwon
Period	14/12/2016 → 16/12/2016

Keywords

Bayesian Inference
Absorption Coefficient
Flow Resistivity
Room’s Influence

Access to Document

KSME csh jchFinal published version, 543 KB

OpenUrl availability

Check availability in DTU Findit

Cite this

@inproceedings{1a9cbb07ce5a4f17892403b4cf62a619,

title = "First- and Second-level Bayesian Inference of Flow Resistivity of Sound Absorber and Room{\textquoteright}s Influence",

abstract = "Sabine absorption coefficient is a widely used one deduced from reverberation time measurements via the Sabine equation. First- and second-level Bayesian analysis are used to estimate the flow resistivity of a sound absorber and the influences of the test chambers from Sabine absorption coefficients measured in 13 different reverberation chambers. The first-level Bayesian analysis is more general than the second-level Bayesian analysis. Sharper posterior distribution can be acquired by the second-level Bayesian analysis than the one by the first-level Bayesian analysis because more data are used to set more reliable prior distribution. The estimated room{\textquoteright}s influences by the first- and the second-level Bayesian analyses are similar to the estimated results by the mean absolute error minimization.",

keywords = "Bayesian Inference, Absorption Coefficient, Flow Resistivity, Room{\textquoteright}s Influence",

author = "Sang-Hyeon Choi and Ikjin Lee and Cheol-Ho Jeong",

year = "2016",

language = "English",

booktitle = "Winter meeting proceedings of the Korean Society of Mechanical Engineers",

note = "Winter Meeting proceedings of the Korean Society of Mechanical Engineers ; Conference date: 14-12-2016 Through 16-12-2016",

}

TY - GEN

T1 - First- and Second-level Bayesian Inference of Flow Resistivity of Sound Absorber and Room’s Influence

AU - Choi, Sang-Hyeon

AU - Lee, Ikjin

AU - Jeong, Cheol-Ho

PY - 2016

Y1 - 2016

N2 - Sabine absorption coefficient is a widely used one deduced from reverberation time measurements via the Sabine equation. First- and second-level Bayesian analysis are used to estimate the flow resistivity of a sound absorber and the influences of the test chambers from Sabine absorption coefficients measured in 13 different reverberation chambers. The first-level Bayesian analysis is more general than the second-level Bayesian analysis. Sharper posterior distribution can be acquired by the second-level Bayesian analysis than the one by the first-level Bayesian analysis because more data are used to set more reliable prior distribution. The estimated room’s influences by the first- and the second-level Bayesian analyses are similar to the estimated results by the mean absolute error minimization.

AB - Sabine absorption coefficient is a widely used one deduced from reverberation time measurements via the Sabine equation. First- and second-level Bayesian analysis are used to estimate the flow resistivity of a sound absorber and the influences of the test chambers from Sabine absorption coefficients measured in 13 different reverberation chambers. The first-level Bayesian analysis is more general than the second-level Bayesian analysis. Sharper posterior distribution can be acquired by the second-level Bayesian analysis than the one by the first-level Bayesian analysis because more data are used to set more reliable prior distribution. The estimated room’s influences by the first- and the second-level Bayesian analyses are similar to the estimated results by the mean absolute error minimization.

KW - Bayesian Inference

KW - Absorption Coefficient

KW - Flow Resistivity

KW - Room’s Influence

M3 - Article in proceedings

BT - Winter meeting proceedings of the Korean Society of Mechanical Engineers

T2 - Winter Meeting proceedings of the Korean Society of Mechanical Engineers

Y2 - 14 December 2016 through 16 December 2016

ER -

First- and Second-level Bayesian Inference of Flow Resistivity of Sound Absorber and Room’s Influence

Abstract

Conference

Keywords

Access to Document

OpenUrl availability

Fingerprint

Cite this