A multilayered ANN architecture for underwater target tracking

Author

Jing, Yuyang ; El-Hawary, Ferial

Author_Institution

Tech. Univ. Nova Scotia, Halifax, NS, Canada

fYear

1994

fDate

25-28 Sep 1994

Firstpage

785

Abstract

A multilayered artificial neural network (ANN) is proposed for tracking underwater targets. A method using a feedforward network is presented to obtain state estimates from the time series of measurements. We shifted the time series observations before presentation to the ANN input and the simulation results show that the ANN tracker achieved a satisfactory degree of accuracy and robustness in dealing with noise in the measurements

Keywords

backpropagation; feedforward neural nets; multilayer perceptrons; neural net architecture; noise; parameter estimation; sonar tracking; state estimation; time series; underwater sound; ANN input; ANN tracker; accuracy; backpropagation; feedforward network; measurements; multilayered ANN architecture; noise; passive sonar; robustness; simulation results; state estimates; time series observations; underwater target tracking; Backpropagation; Computer architecture; Feedforward neural networks; Multilayer perceptrons; Neural network applications; Noise; Parameter estimation; Sonar measurements; Sonar tracking; State estimation; Time series; Underwater acoustic measurements;

fLanguage

English

Publisher

ieee

Conference_Titel

Electrical and Computer Engineering, 1994. Conference Proceedings. 1994 Canadian Conference on

Conference_Location

Halifax, NS

Print_ISBN

0-7803-2416-1

Type

conf

DOI

10.1109/CCECE.1994.405869

Filename

405869