Continuous observation and archival of acoustic scenes usingwireless sensor networks

Gordon Wichern, Homin Kwon, Andreas Spanias, Alex Fink, Harvey Thornburg

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Acoustic scene analysis has proven an invaluable tool in diverse fields ranging from biology to defense and security. Wireless sensor networks present an attractive means of implementing an acoustic monitoring system due to their low cost and ability to be easily deployed in a wide range of areas. In this paper acoustic features are extracted at the sensor level, and then transmitted to the base station where acoustic events are segmented using a dynamic Bayseian network. Segmented events are then indexed with the time and location where they occurred, allowing users to link events in terms of time, place, and acoustic characteristics. Our experiments show that a feature set that allows for general characterization of diverse sound environments can be extracted at the sensor level, while an illustrative example shows the segmentation algorithm detecting footsteps in low SNR conditions.

Original languageEnglish (US)
Title of host publicationDSP 2009: 16th International Conference on Digital Signal Processing, Proceedings
DOIs
StatePublished - 2009
EventDSP 2009:16th International Conference on Digital Signal Processing - Santorini, Greece
Duration: Jul 5 2009Jul 7 2009

Other

OtherDSP 2009:16th International Conference on Digital Signal Processing
CountryGreece
CitySantorini
Period7/5/097/7/09

Fingerprint

Sensor networks
Acoustics
Sensors
Base stations
Wireless sensor networks
Acoustic waves
Monitoring
Costs
Experiments

Keywords

  • Acoustic scene analysis
  • Acoustic signal detection
  • Database query processing
  • Wireless sensor networks

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Wichern, G., Kwon, H., Spanias, A., Fink, A., & Thornburg, H. (2009). Continuous observation and archival of acoustic scenes usingwireless sensor networks. In DSP 2009: 16th International Conference on Digital Signal Processing, Proceedings [5201082] https://doi.org/10.1109/ICDSP.2009.5201082

Continuous observation and archival of acoustic scenes usingwireless sensor networks. / Wichern, Gordon; Kwon, Homin; Spanias, Andreas; Fink, Alex; Thornburg, Harvey.

DSP 2009: 16th International Conference on Digital Signal Processing, Proceedings. 2009. 5201082.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Wichern, G, Kwon, H, Spanias, A, Fink, A & Thornburg, H 2009, Continuous observation and archival of acoustic scenes usingwireless sensor networks. in DSP 2009: 16th International Conference on Digital Signal Processing, Proceedings., 5201082, DSP 2009:16th International Conference on Digital Signal Processing, Santorini, Greece, 7/5/09. https://doi.org/10.1109/ICDSP.2009.5201082
Wichern G, Kwon H, Spanias A, Fink A, Thornburg H. Continuous observation and archival of acoustic scenes usingwireless sensor networks. In DSP 2009: 16th International Conference on Digital Signal Processing, Proceedings. 2009. 5201082 https://doi.org/10.1109/ICDSP.2009.5201082
Wichern, Gordon ; Kwon, Homin ; Spanias, Andreas ; Fink, Alex ; Thornburg, Harvey. / Continuous observation and archival of acoustic scenes usingwireless sensor networks. DSP 2009: 16th International Conference on Digital Signal Processing, Proceedings. 2009.
@inproceedings{700e35ed1ff24097b0bd9649ab8e08db,
title = "Continuous observation and archival of acoustic scenes usingwireless sensor networks",
abstract = "Acoustic scene analysis has proven an invaluable tool in diverse fields ranging from biology to defense and security. Wireless sensor networks present an attractive means of implementing an acoustic monitoring system due to their low cost and ability to be easily deployed in a wide range of areas. In this paper acoustic features are extracted at the sensor level, and then transmitted to the base station where acoustic events are segmented using a dynamic Bayseian network. Segmented events are then indexed with the time and location where they occurred, allowing users to link events in terms of time, place, and acoustic characteristics. Our experiments show that a feature set that allows for general characterization of diverse sound environments can be extracted at the sensor level, while an illustrative example shows the segmentation algorithm detecting footsteps in low SNR conditions.",
keywords = "Acoustic scene analysis, Acoustic signal detection, Database query processing, Wireless sensor networks",
author = "Gordon Wichern and Homin Kwon and Andreas Spanias and Alex Fink and Harvey Thornburg",
year = "2009",
doi = "10.1109/ICDSP.2009.5201082",
language = "English (US)",
isbn = "9781424432981",
booktitle = "DSP 2009: 16th International Conference on Digital Signal Processing, Proceedings",

}

TY - GEN

T1 - Continuous observation and archival of acoustic scenes usingwireless sensor networks

AU - Wichern, Gordon

AU - Kwon, Homin

AU - Spanias, Andreas

AU - Fink, Alex

AU - Thornburg, Harvey

PY - 2009

Y1 - 2009

N2 - Acoustic scene analysis has proven an invaluable tool in diverse fields ranging from biology to defense and security. Wireless sensor networks present an attractive means of implementing an acoustic monitoring system due to their low cost and ability to be easily deployed in a wide range of areas. In this paper acoustic features are extracted at the sensor level, and then transmitted to the base station where acoustic events are segmented using a dynamic Bayseian network. Segmented events are then indexed with the time and location where they occurred, allowing users to link events in terms of time, place, and acoustic characteristics. Our experiments show that a feature set that allows for general characterization of diverse sound environments can be extracted at the sensor level, while an illustrative example shows the segmentation algorithm detecting footsteps in low SNR conditions.

AB - Acoustic scene analysis has proven an invaluable tool in diverse fields ranging from biology to defense and security. Wireless sensor networks present an attractive means of implementing an acoustic monitoring system due to their low cost and ability to be easily deployed in a wide range of areas. In this paper acoustic features are extracted at the sensor level, and then transmitted to the base station where acoustic events are segmented using a dynamic Bayseian network. Segmented events are then indexed with the time and location where they occurred, allowing users to link events in terms of time, place, and acoustic characteristics. Our experiments show that a feature set that allows for general characterization of diverse sound environments can be extracted at the sensor level, while an illustrative example shows the segmentation algorithm detecting footsteps in low SNR conditions.

KW - Acoustic scene analysis

KW - Acoustic signal detection

KW - Database query processing

KW - Wireless sensor networks

UR - http://www.scopus.com/inward/record.url?scp=70449589090&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70449589090&partnerID=8YFLogxK

U2 - 10.1109/ICDSP.2009.5201082

DO - 10.1109/ICDSP.2009.5201082

M3 - Conference contribution

AN - SCOPUS:70449589090

SN - 9781424432981

BT - DSP 2009: 16th International Conference on Digital Signal Processing, Proceedings

ER -