Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Dictionary learning of sound speed profiles.

Michael Bianco1, Peter Gerstoft1

  • 1Scripps Institution of Oceanography, University of California San Diego, La Jolla, California 92093-0238, USA.

The Journal of the Acoustical Society of America
|April 5, 2017
PubMed
Summary

Dictionary learning improves ocean sound speed profile (SSP) resolution by creating better shape function dictionaries than empirical orthogonal functions (EOFs), requiring fewer coefficients for accurate modeling.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

A Multi-Head Attention Transformer Model for Wearable in Situ Fall Detection.

IEEE access : practical innovations, open solutions·2026
Same author

Continuous forecasting of range-dependent ocean sound speed field: Diffusion model meets multi-output Gaussian process.

The Journal of the Acoustical Society of America·2026
Same author

Sensor beampattern and equivalent aperture in a distributed acoustic sensing system.

The Journal of the Acoustical Society of America·2026
Same author

Violence Exposure, Affect and ART Use Among Young Black and Latinx Men Who have Sex with Men: An Ecological Momentary Assessment Analysis.

AIDS and behavior·2026
Same author

Hankel-FNO: Fast underwater acoustic charting via physics-encoded Fourier neural operator.

The Journal of the Acoustical Society of America·2025
Same author

Evaluating Gaussian processes for matched-field processing localization using minimum mean squared error criterion.

JASA express letters·2025

Area of Science:

  • Oceanography
  • Applied Mathematics
  • Machine Learning

Background:

  • Ocean sound speed profiles (SSPs) are crucial for acoustic modeling.
  • Empirical orthogonal functions (EOFs) are commonly used for SSP regularization but often result in low-resolution estimates.
  • Existing methods struggle to balance resolution and accuracy in SSP inversion.

Purpose of the Study:

  • To introduce dictionary learning (DL) as a superior method for modeling SSPs.
  • To demonstrate DL's ability to enhance SSP resolution compared to traditional EOF methods.
  • To optimize SSP inversion by minimizing reconstruction error and coefficient count.

Main Methods:

  • Dictionary learning (DL), specifically the K-SVD algorithm, was employed to generate dictionaries of shape functions.
  • Learned Dictionaries (LDs) were generated using SSP observations from the HF-97 experiment and the South China Sea.
  • The performance of LDs was compared against traditional EOFs for SSP compression and reconstruction.

Main Results:

  • Learned dictionaries (LDs) effectively capture SSP variability using significantly fewer coefficients than EOFs.
  • A substantial portion of SSP variability was explained using just one coefficient with LDs.
  • The DL approach demonstrated improved SSP resolution with negligible computational overhead.

Conclusions:

  • Dictionary learning offers a powerful, data-driven approach to improve SSP estimation resolution.
  • LDs provide a more efficient and accurate method for SSP regularization compared to EOFs.
  • This machine learning technique holds promise for advancing ocean acoustic modeling and inversion.

Related Experiment Videos