Neural network for ocean bottom parameter estimation

Robert H. Baran, Hanseok Ko

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

Abstract

This concerns the estimation of physical parameters characterizing the ocean bottom from parallel time series representing (A) the instantaneous height of the water column above a section of bottom and (B) the vertical displacement of the bottom section from its long term average. Time series A, representing wind-generated waves, is modeled by a sinusoid with phase jitter. In the absence of both seismic background noise and any nonlinear behavior in the ocean bottom, time series B could be modeled by coupling the bottom through a spring and dashpot to a mass proportional to A. We created two- and three-layer adaptive networks in which series A and B (with lag) were inputs. Training consisted in subtracting the network output from the current value of series B and feeding back these errors in accordance with the appropriate formulas for gradient descent in squared error. The trained nets act as models of the way in which the bottom responds to changes at the surface.

Original languageEnglish
Title of host publicationProceedings of SPIE - The International Society for Optical Engineering
Place of PublicationBellingham, WA, United States
PublisherPubl by Int Soc for Optical Engineering
Pages52-63
Number of pages12
Volume1700
ISBN (Print)0819408654
Publication statusPublished - 1992 Dec 1
Externally publishedYes
EventAutomatic Object Recognition II - Orlando, FL, USA
Duration: 1992 Apr 221992 Apr 24

Other

OtherAutomatic Object Recognition II
CityOrlando, FL, USA
Period92/4/2292/4/24

Fingerprint

ocean bottom
Parameter estimation
Time series
Neural networks
sine waves
background noise
descent
Jitter
education
time lag
vibration
gradients
output
water
Water

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Condensed Matter Physics

Cite this

Baran, R. H., & Ko, H. (1992). Neural network for ocean bottom parameter estimation. In Proceedings of SPIE - The International Society for Optical Engineering (Vol. 1700, pp. 52-63). Bellingham, WA, United States: Publ by Int Soc for Optical Engineering.

Neural network for ocean bottom parameter estimation. / Baran, Robert H.; Ko, Hanseok.

Proceedings of SPIE - The International Society for Optical Engineering. Vol. 1700 Bellingham, WA, United States : Publ by Int Soc for Optical Engineering, 1992. p. 52-63.

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

Baran, RH & Ko, H 1992, Neural network for ocean bottom parameter estimation. in Proceedings of SPIE - The International Society for Optical Engineering. vol. 1700, Publ by Int Soc for Optical Engineering, Bellingham, WA, United States, pp. 52-63, Automatic Object Recognition II, Orlando, FL, USA, 92/4/22.
Baran RH, Ko H. Neural network for ocean bottom parameter estimation. In Proceedings of SPIE - The International Society for Optical Engineering. Vol. 1700. Bellingham, WA, United States: Publ by Int Soc for Optical Engineering. 1992. p. 52-63
Baran, Robert H. ; Ko, Hanseok. / Neural network for ocean bottom parameter estimation. Proceedings of SPIE - The International Society for Optical Engineering. Vol. 1700 Bellingham, WA, United States : Publ by Int Soc for Optical Engineering, 1992. pp. 52-63
@inproceedings{71ecf678ccf94e37b7c92a03c6b5e345,
title = "Neural network for ocean bottom parameter estimation",
abstract = "This concerns the estimation of physical parameters characterizing the ocean bottom from parallel time series representing (A) the instantaneous height of the water column above a section of bottom and (B) the vertical displacement of the bottom section from its long term average. Time series A, representing wind-generated waves, is modeled by a sinusoid with phase jitter. In the absence of both seismic background noise and any nonlinear behavior in the ocean bottom, time series B could be modeled by coupling the bottom through a spring and dashpot to a mass proportional to A. We created two- and three-layer adaptive networks in which series A and B (with lag) were inputs. Training consisted in subtracting the network output from the current value of series B and feeding back these errors in accordance with the appropriate formulas for gradient descent in squared error. The trained nets act as models of the way in which the bottom responds to changes at the surface.",
author = "Baran, {Robert H.} and Hanseok Ko",
year = "1992",
month = "12",
day = "1",
language = "English",
isbn = "0819408654",
volume = "1700",
pages = "52--63",
booktitle = "Proceedings of SPIE - The International Society for Optical Engineering",
publisher = "Publ by Int Soc for Optical Engineering",

}

TY - GEN

T1 - Neural network for ocean bottom parameter estimation

AU - Baran, Robert H.

AU - Ko, Hanseok

PY - 1992/12/1

Y1 - 1992/12/1

N2 - This concerns the estimation of physical parameters characterizing the ocean bottom from parallel time series representing (A) the instantaneous height of the water column above a section of bottom and (B) the vertical displacement of the bottom section from its long term average. Time series A, representing wind-generated waves, is modeled by a sinusoid with phase jitter. In the absence of both seismic background noise and any nonlinear behavior in the ocean bottom, time series B could be modeled by coupling the bottom through a spring and dashpot to a mass proportional to A. We created two- and three-layer adaptive networks in which series A and B (with lag) were inputs. Training consisted in subtracting the network output from the current value of series B and feeding back these errors in accordance with the appropriate formulas for gradient descent in squared error. The trained nets act as models of the way in which the bottom responds to changes at the surface.

AB - This concerns the estimation of physical parameters characterizing the ocean bottom from parallel time series representing (A) the instantaneous height of the water column above a section of bottom and (B) the vertical displacement of the bottom section from its long term average. Time series A, representing wind-generated waves, is modeled by a sinusoid with phase jitter. In the absence of both seismic background noise and any nonlinear behavior in the ocean bottom, time series B could be modeled by coupling the bottom through a spring and dashpot to a mass proportional to A. We created two- and three-layer adaptive networks in which series A and B (with lag) were inputs. Training consisted in subtracting the network output from the current value of series B and feeding back these errors in accordance with the appropriate formulas for gradient descent in squared error. The trained nets act as models of the way in which the bottom responds to changes at the surface.

UR - http://www.scopus.com/inward/record.url?scp=0027000517&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0027000517&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0027000517

SN - 0819408654

VL - 1700

SP - 52

EP - 63

BT - Proceedings of SPIE - The International Society for Optical Engineering

PB - Publ by Int Soc for Optical Engineering

CY - Bellingham, WA, United States

ER -