A Learning-based Framework for Line-spectra Super-resolution

Gautier Izacard, Brett Bernstein, Carlos Fernandez-Granda

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

Abstract

We propose a learning-based approach for estimating the spectrum of a multisinusoidal signal from a finite number of samples. A neural-network is trained to approximate the spectra of such signals on simulated data. The proposed methodology is very flexible: adapting to different signal and noise models only requires modifying the training data accordingly. Numerical experiments show that the approach performs competitively with classical methods designed for additive Gaussian noise at a range of noise levels, and is also effective in the presence of impulsive noise.

Original languageEnglish (US)
Title of host publication2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages3632-3636
Number of pages5
ISBN (Electronic)9781479981311
DOIs
StatePublished - May 1 2019
Event44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Brighton, United Kingdom
Duration: May 12 2019May 17 2019

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume2019-May
ISSN (Print)1520-6149

Conference

Conference44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019
CountryUnited Kingdom
CityBrighton
Period5/12/195/17/19

Fingerprint

Impulse noise
Neural networks
Experiments

Keywords

  • deep learning
  • Line-spectra estimation
  • neural networks
  • sparse noise
  • spectral super-resolution

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Izacard, G., Bernstein, B., & Fernandez-Granda, C. (2019). A Learning-based Framework for Line-spectra Super-resolution. In 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings (pp. 3632-3636). [8682882] (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2019-May). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2019.8682882

A Learning-based Framework for Line-spectra Super-resolution. / Izacard, Gautier; Bernstein, Brett; Fernandez-Granda, Carlos.

2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. p. 3632-3636 8682882 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings; Vol. 2019-May).

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

Izacard, G, Bernstein, B & Fernandez-Granda, C 2019, A Learning-based Framework for Line-spectra Super-resolution. in 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings., 8682882, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, vol. 2019-May, Institute of Electrical and Electronics Engineers Inc., pp. 3632-3636, 44th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019, Brighton, United Kingdom, 5/12/19. https://doi.org/10.1109/ICASSP.2019.8682882
Izacard G, Bernstein B, Fernandez-Granda C. A Learning-based Framework for Line-spectra Super-resolution. In 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2019. p. 3632-3636. 8682882. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2019.8682882
Izacard, Gautier ; Bernstein, Brett ; Fernandez-Granda, Carlos. / A Learning-based Framework for Line-spectra Super-resolution. 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 3632-3636 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).
@inproceedings{273084424fdf4adfaeec554fd9152f9b,
title = "A Learning-based Framework for Line-spectra Super-resolution",
abstract = "We propose a learning-based approach for estimating the spectrum of a multisinusoidal signal from a finite number of samples. A neural-network is trained to approximate the spectra of such signals on simulated data. The proposed methodology is very flexible: adapting to different signal and noise models only requires modifying the training data accordingly. Numerical experiments show that the approach performs competitively with classical methods designed for additive Gaussian noise at a range of noise levels, and is also effective in the presence of impulsive noise.",
keywords = "deep learning, Line-spectra estimation, neural networks, sparse noise, spectral super-resolution",
author = "Gautier Izacard and Brett Bernstein and Carlos Fernandez-Granda",
year = "2019",
month = "5",
day = "1",
doi = "10.1109/ICASSP.2019.8682882",
language = "English (US)",
series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "3632--3636",
booktitle = "2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings",

}

TY - GEN

T1 - A Learning-based Framework for Line-spectra Super-resolution

AU - Izacard, Gautier

AU - Bernstein, Brett

AU - Fernandez-Granda, Carlos

PY - 2019/5/1

Y1 - 2019/5/1

N2 - We propose a learning-based approach for estimating the spectrum of a multisinusoidal signal from a finite number of samples. A neural-network is trained to approximate the spectra of such signals on simulated data. The proposed methodology is very flexible: adapting to different signal and noise models only requires modifying the training data accordingly. Numerical experiments show that the approach performs competitively with classical methods designed for additive Gaussian noise at a range of noise levels, and is also effective in the presence of impulsive noise.

AB - We propose a learning-based approach for estimating the spectrum of a multisinusoidal signal from a finite number of samples. A neural-network is trained to approximate the spectra of such signals on simulated data. The proposed methodology is very flexible: adapting to different signal and noise models only requires modifying the training data accordingly. Numerical experiments show that the approach performs competitively with classical methods designed for additive Gaussian noise at a range of noise levels, and is also effective in the presence of impulsive noise.

KW - deep learning

KW - Line-spectra estimation

KW - neural networks

KW - sparse noise

KW - spectral super-resolution

UR - http://www.scopus.com/inward/record.url?scp=85068972773&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068972773&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2019.8682882

DO - 10.1109/ICASSP.2019.8682882

M3 - Conference contribution

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 3632

EP - 3636

BT - 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -