Abstract

Seafloor observatories are a rapidly maturing technological approach, enabling the monitoring of marine soundscapes over large areas (basin scale) and long timescales (10+ years). However, the time needed to process broadband measurements, especially over large periods, often acts as a bottleneck. This is particularly true when combining multi-resolution analyses with assessing the impacts of relatively short transients. We are using parallel processing to enable machine learning approaches. To accelerate the computation of spectrograms, we have implemented a parallel processing method that uses the FFT algorithm FFTW3 (http://www.fftw.org/fftw3.pdf), using MPI/C++ on the High Performance Computing facilities at the University of Bath, and compared with spectrogram calculations from well-established software PAMGuide (Merchant et al., Meth. Ecol. Evol., 2015), with Matlab’s Parallel Computing Toolbox. This approach was tested on 1 month of broadband (96 kHz) measurements from the NEPTUNE node at Folger Deep. One month of data can be processed in < 3 hours, to a dB accuracy even on short time segments, and that performance increases with the number of parallel processing units. Stability of the parallel approach has been tested with synthetic signals (e.g. chirps) and increasing signal-to-noise ratios. This enables much faster monitoring of long-term trends of important sound metrics.
Original languageEnglish
Pages1732-1732
Number of pages1
Publication statusPublished - Nov 2018
Event176th Meeting of the Acoustical Society of America and 2018 Acoustics Week in Canada - Victoria Conference Centre, Victoria, Canada
Duration: 5 Nov 20189 Nov 2018
Conference number: 176
https://acousticalsociety.org/176th-meeting-acoustical-society-of-america/

Conference

Conference176th Meeting of the Acoustical Society of America and 2018 Acoustics Week in Canada
Abbreviated titleASA-2018
CountryCanada
CityVictoria
Period5/11/189/11/18
Internet address

Keywords

  • underwater acoustics
  • ambient noise
  • parallel processing
  • seafloor observatories

ASJC Scopus subject areas

  • Acoustics and Ultrasonics
  • Mathematical Physics

Cite this

Parallel processing and machine learning for long-timescale ambient noise measurements – Illustration with data from the Neptune Ocean Observatory offshore British Columbia. / Klein, Amélie; Blondel, Philippe; Heine, Kari.

2018. 1732-1732 Abstract from 176th Meeting of the Acoustical Society of America and 2018 Acoustics Week in Canada, Victoria, Canada.

Research output: Contribution to conferenceAbstract

Klein, A, Blondel, P & Heine, K 2018, 'Parallel processing and machine learning for long-timescale ambient noise measurements – Illustration with data from the Neptune Ocean Observatory offshore British Columbia' 176th Meeting of the Acoustical Society of America and 2018 Acoustics Week in Canada, Victoria, Canada, 5/11/18 - 9/11/18, pp. 1732-1732.
@conference{f0cef27cfef44591b7506b80d61fa889,
title = "Parallel processing and machine learning for long-timescale ambient noise measurements – Illustration with data from the Neptune Ocean Observatory offshore British Columbia",
abstract = "Seafloor observatories are a rapidly maturing technological approach, enabling the monitoring of marine soundscapes over large areas (basin scale) and long timescales (10+ years). However, the time needed to process broadband measurements, especially over large periods, often acts as a bottleneck. This is particularly true when combining multi-resolution analyses with assessing the impacts of relatively short transients. We are using parallel processing to enable machine learning approaches. To accelerate the computation of spectrograms, we have implemented a parallel processing method that uses the FFT algorithm FFTW3 (http://www.fftw.org/fftw3.pdf), using MPI/C++ on the High Performance Computing facilities at the University of Bath, and compared with spectrogram calculations from well-established software PAMGuide (Merchant et al., Meth. Ecol. Evol., 2015), with Matlab’s Parallel Computing Toolbox. This approach was tested on 1 month of broadband (96 kHz) measurements from the NEPTUNE node at Folger Deep. One month of data can be processed in < 3 hours, to a dB accuracy even on short time segments, and that performance increases with the number of parallel processing units. Stability of the parallel approach has been tested with synthetic signals (e.g. chirps) and increasing signal-to-noise ratios. This enables much faster monitoring of long-term trends of important sound metrics.",
keywords = "underwater acoustics, ambient noise, parallel processing, seafloor observatories",
author = "Am{\'e}lie Klein and Philippe Blondel and Kari Heine",
year = "2018",
month = "11",
language = "English",
pages = "1732--1732",
note = "176th Meeting of the Acoustical Society of America and 2018 Acoustics Week in Canada, ASA-2018 ; Conference date: 05-11-2018 Through 09-11-2018",
url = "https://acousticalsociety.org/176th-meeting-acoustical-society-of-america/",

}

TY - CONF

T1 - Parallel processing and machine learning for long-timescale ambient noise measurements – Illustration with data from the Neptune Ocean Observatory offshore British Columbia

AU - Klein, Amélie

AU - Blondel, Philippe

AU - Heine, Kari

PY - 2018/11

Y1 - 2018/11

N2 - Seafloor observatories are a rapidly maturing technological approach, enabling the monitoring of marine soundscapes over large areas (basin scale) and long timescales (10+ years). However, the time needed to process broadband measurements, especially over large periods, often acts as a bottleneck. This is particularly true when combining multi-resolution analyses with assessing the impacts of relatively short transients. We are using parallel processing to enable machine learning approaches. To accelerate the computation of spectrograms, we have implemented a parallel processing method that uses the FFT algorithm FFTW3 (http://www.fftw.org/fftw3.pdf), using MPI/C++ on the High Performance Computing facilities at the University of Bath, and compared with spectrogram calculations from well-established software PAMGuide (Merchant et al., Meth. Ecol. Evol., 2015), with Matlab’s Parallel Computing Toolbox. This approach was tested on 1 month of broadband (96 kHz) measurements from the NEPTUNE node at Folger Deep. One month of data can be processed in < 3 hours, to a dB accuracy even on short time segments, and that performance increases with the number of parallel processing units. Stability of the parallel approach has been tested with synthetic signals (e.g. chirps) and increasing signal-to-noise ratios. This enables much faster monitoring of long-term trends of important sound metrics.

AB - Seafloor observatories are a rapidly maturing technological approach, enabling the monitoring of marine soundscapes over large areas (basin scale) and long timescales (10+ years). However, the time needed to process broadband measurements, especially over large periods, often acts as a bottleneck. This is particularly true when combining multi-resolution analyses with assessing the impacts of relatively short transients. We are using parallel processing to enable machine learning approaches. To accelerate the computation of spectrograms, we have implemented a parallel processing method that uses the FFT algorithm FFTW3 (http://www.fftw.org/fftw3.pdf), using MPI/C++ on the High Performance Computing facilities at the University of Bath, and compared with spectrogram calculations from well-established software PAMGuide (Merchant et al., Meth. Ecol. Evol., 2015), with Matlab’s Parallel Computing Toolbox. This approach was tested on 1 month of broadband (96 kHz) measurements from the NEPTUNE node at Folger Deep. One month of data can be processed in < 3 hours, to a dB accuracy even on short time segments, and that performance increases with the number of parallel processing units. Stability of the parallel approach has been tested with synthetic signals (e.g. chirps) and increasing signal-to-noise ratios. This enables much faster monitoring of long-term trends of important sound metrics.

KW - underwater acoustics

KW - ambient noise

KW - parallel processing

KW - seafloor observatories

M3 - Abstract

SP - 1732

EP - 1732

ER -