Related Experiment Videos
Blind deconvolution applied to acoustical systems identification with supporting experimental results
Michael J Roan1, Mark R Gramann, Josh G Erling
1Applied Research Laboratory, The Pennsylvania State University, P.O. Box 30, State College, Pennsylvania 16804, USA. mjr110@psu.edu
The Journal of the Acoustical Society of America
|November 1, 2003
Summary
This study experimentally verifies a blind deconvolution algorithm for acoustical system identification. The algorithm successfully identifies unknown acoustic filters and removes their effects, advancing noise control and signal processing applications.
Area of Science:
- Acoustics
- Signal Processing
- System Identification
Background:
- Acoustical applications often involve signals corrupted by unknown filtering functions.
- Blind deconvolution aims to recover original signals without prior knowledge of the filter, relying on signal statistics.
Purpose of the Study:
- To experimentally verify a blind deconvolution algorithm for acoustical system identification.
- To address the lack of experimental validation in previous blind deconvolution research for acoustics.
Main Methods:
- The study employed a blind deconvolution algorithm.
- Experiments were conducted on three classical acoustic systems: a driven pipe, a driven pipe with an open side branch, and a driven pipe with a Helmholtz resonator side branch.
Main Results:
- The blind deconvolution algorithm successfully learned the inverse impulse responses of the tested acoustic systems.
- Applying the learned inverse filters effectively removed the filtering effects from the observed signals.
Conclusions:
- Experimental results confirm the efficacy of the blind deconvolution algorithm in acoustical system identification.
- This work provides crucial experimental validation for blind deconvolution in acoustic applications, moving beyond simulated data.