Experimental method for the acoustical modelling of the echolocation process in bats
Experimental method for the acoustical modelling of the echolocation process in bats
The goal of this research is to establish an experimental technique that can accurately quantify and model the acoustics underlying the echolocation process in bats. The ultimate outcome in such a research direction would be the ability to determine the pair of pressure signals created at the positions of the bat’s ears due to the reflected portion of the energy carried by a given echolocation emission pulse (be it a frequency-modulated pulse, a constant-frequency pulse or a series of clicks). The ability to accurately determine these signals would provide a solid basis for the investigation and the modelling of the relevant physiological aspects of the bat’s auditory system.
Previous work in this direction includes that of Simmons and Chen (JASA 86(4) October 1989, pp. 1333-1350) in which an experimental arrangement was described for the measurement of the single-channel impulse response between a source and a receiver positioned close to each other and both facing a small object. Responses corresponding to reflections from a collection of different objects at various orientations were presented and studied in an effort to explain the ability of FM echolocating bats to discriminate between edible and non-edible targets presented to them. In that work, distinct features were identified in the time-domain and frequency-domain representations of the measured responses and they were correlated with the relative positioning of glints in the various objects and orientations.
The research presented here builds on the work by Simmons and Chen and consists of two parts. In the first part we present a set of measured results which we obtained with an apparatus that essentially replicates that of Simmons and Chen but scaled down in frequency and up in physical dimensions by a factor of 2.5. These results show good agreement with those of Simmons and Chen and verify the fact that our measurement procedure succeeds in identifying the same features in the measured responses as mentioned above. On the other hand, the results make apparent a basic shortcoming of the measurement procedure. More specifically, in order to be able to accurately position and rotate a small object in space, a substantial mechanical apparatus has to be constructed around the object. This inevitably results in reflections from the physical objects of this apparatus (including the source and receiver) which extend in the measured impulse responses so as to overlap with the features due to the object under investigation. Especially when very small objects are considered, the reflected energy from the object can be so low as to be completely buried under the extraneous reflections. A number of signal processing oriented methods for the isolation of the meaningful part of the measured responses from the part due to the extraneous reflections are presented and the restrictions of such methods are discussed.
In the second part of the paper we describe a reformulation of the imaging/demigration technique originally introduced by Kuster et al. (JASA 116(4) October 2004, pp. 2126-2137) in the context of the room impulse response modification problem. We propose this reformulation as an alternative means to address the problem of the unwanted reflections in the measured impulse responses described above and we discuss the required specifications of a measurement apparatus that could make use of such a technique in the frequency range of interest for the study of the acoustics of bat echolocation. Finally, the proposed experimental technique is expanded to cover the case where the pair of pressure signals at the bat ears is to be determined. We do this by means of a suitable adaptation of the binaural synthesis technique that is based on the concept of the Head Related Transfer Function (HRTF) and is widely used in the design of virtual acoustic imaging systems for humans. We describe the measurement apparatus that is needed for such an adaptation and we review the existing sources in the literature that describe the measurement or modelling of bat HRTFs.
8pp
Papadopoulos, T.
d1af8c5a-f58c-44c8-a4bb-6d2ad9695f95
Allen, R.
956a918f-278c-48ef-8e19-65aa463f199a
2007
Papadopoulos, T.
d1af8c5a-f58c-44c8-a4bb-6d2ad9695f95
Allen, R.
956a918f-278c-48ef-8e19-65aa463f199a
Papadopoulos, T. and Allen, R.
(2007)
Experimental method for the acoustical modelling of the echolocation process in bats.
In 4th International Conference on Bio-Acoustics.
vol. 29, pt.3,
Institute of Acoustics.
.
Record type:
Conference or Workshop Item
(Paper)
Abstract
The goal of this research is to establish an experimental technique that can accurately quantify and model the acoustics underlying the echolocation process in bats. The ultimate outcome in such a research direction would be the ability to determine the pair of pressure signals created at the positions of the bat’s ears due to the reflected portion of the energy carried by a given echolocation emission pulse (be it a frequency-modulated pulse, a constant-frequency pulse or a series of clicks). The ability to accurately determine these signals would provide a solid basis for the investigation and the modelling of the relevant physiological aspects of the bat’s auditory system.
Previous work in this direction includes that of Simmons and Chen (JASA 86(4) October 1989, pp. 1333-1350) in which an experimental arrangement was described for the measurement of the single-channel impulse response between a source and a receiver positioned close to each other and both facing a small object. Responses corresponding to reflections from a collection of different objects at various orientations were presented and studied in an effort to explain the ability of FM echolocating bats to discriminate between edible and non-edible targets presented to them. In that work, distinct features were identified in the time-domain and frequency-domain representations of the measured responses and they were correlated with the relative positioning of glints in the various objects and orientations.
The research presented here builds on the work by Simmons and Chen and consists of two parts. In the first part we present a set of measured results which we obtained with an apparatus that essentially replicates that of Simmons and Chen but scaled down in frequency and up in physical dimensions by a factor of 2.5. These results show good agreement with those of Simmons and Chen and verify the fact that our measurement procedure succeeds in identifying the same features in the measured responses as mentioned above. On the other hand, the results make apparent a basic shortcoming of the measurement procedure. More specifically, in order to be able to accurately position and rotate a small object in space, a substantial mechanical apparatus has to be constructed around the object. This inevitably results in reflections from the physical objects of this apparatus (including the source and receiver) which extend in the measured impulse responses so as to overlap with the features due to the object under investigation. Especially when very small objects are considered, the reflected energy from the object can be so low as to be completely buried under the extraneous reflections. A number of signal processing oriented methods for the isolation of the meaningful part of the measured responses from the part due to the extraneous reflections are presented and the restrictions of such methods are discussed.
In the second part of the paper we describe a reformulation of the imaging/demigration technique originally introduced by Kuster et al. (JASA 116(4) October 2004, pp. 2126-2137) in the context of the room impulse response modification problem. We propose this reformulation as an alternative means to address the problem of the unwanted reflections in the measured impulse responses described above and we discuss the required specifications of a measurement apparatus that could make use of such a technique in the frequency range of interest for the study of the acoustics of bat echolocation. Finally, the proposed experimental technique is expanded to cover the case where the pair of pressure signals at the bat ears is to be determined. We do this by means of a suitable adaptation of the binaural synthesis technique that is based on the concept of the Head Related Transfer Function (HRTF) and is widely used in the design of virtual acoustic imaging systems for humans. We describe the measurement apparatus that is needed for such an adaptation and we review the existing sources in the literature that describe the measurement or modelling of bat HRTFs.
This record has no associated files available for download.
More information
Published date: 2007
Venue - Dates:
4th International Conference on Bio-Acoustics, , Loughborough, United Kingdom, 2007-04-10 - 2007-04-12
Identifiers
Local EPrints ID: 46547
URI: http://eprints.soton.ac.uk/id/eprint/46547
PURE UUID: dfcb6332-cc86-4227-a09c-3305c35a049d
Catalogue record
Date deposited: 19 Jul 2007
Last modified: 20 Feb 2024 11:19
Export record
Contributors
Author:
T. Papadopoulos
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics