Comparison of multilayer perceptrons and maximum likelihood processors in the context of seafloor parameter estimation

Z. H. Michalopoulou, D. Alexandrou

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

Abstract

The objective of this work is acoustic seafloor characterization based on the statistical properties of bottom reverberation. Two seafloor roughness parameters are related to backscattering strength through the Helmholtz-Kirchhoff scattering model. These parameters are estimated through the application of multilayer perceptrons and Maximum Likelihood (ML) estimators to synthetic backscatter representing seafloors with different morphology. The multilayer perceptron is presented with synthetic sequences and is instructed to yield the parameters of interest in the output. ML estimation relies on the construction of a rectangular grid that maps scattering strength vectors to the two parameters under estimation through the Helmholtz-Kirchhoff model. The parameter estimates are the coordinates of the grid element at which the likelihood function - calculated for sets of simulated observations - obtains a maximum. The performance of both estimation processors is determined by the mean and variance of the obtained parameter estimates.

Original languageEnglish (US)
Title of host publicationProc Conf Oceans 93
Editors Anon
PublisherPubl by IEEE
Pages224-228
Number of pages5
ISBN (Print)0780313860
StatePublished - 1993
Externally publishedYes
EventProceedings of the Conference on Oceans '93. Part 3 (of 3) - Victoria, BC, Can
Duration: Oct 18 1993Oct 21 1993

Publication series

NameProc Conf Oceans 93

Other

OtherProceedings of the Conference on Oceans '93. Part 3 (of 3)
CityVictoria, BC, Can
Period10/18/9310/21/93

All Science Journal Classification (ASJC) codes

  • General Engineering

Fingerprint

Dive into the research topics of 'Comparison of multilayer perceptrons and maximum likelihood processors in the context of seafloor parameter estimation'. Together they form a unique fingerprint.

Cite this