Abstract
Original language | English |
---|---|
Place of Publication | Birmingham, UK |
Publisher | Aston University |
Number of pages | 58 |
ISBN (Print) | NCRG/99/005 |
Publication status | Published - 1999 |
Fingerprint
Keywords
- Satellite
- European Space Agency
- polar orbit
- scatterometer
- ripples
- local winds
- local inversion
- forward model
- mapping
- cost function
- density networks
- conditional probability density functions
- joint probability distribution
- kernels
- Gaussian mixture model
- geophysical knowledge
- ambiguity removal
- Bayesian framework
Cite this
}
First year qualifying report: neural networks for extracting wind vectors from satellite scatterometer data. / Evans, David J.
Birmingham, UK : Aston University, 1999.Research output: Working paper › Project report
TY - UNPB
T1 - First year qualifying report: neural networks for extracting wind vectors from satellite scatterometer data
AU - Evans, David J.
PY - 1999
Y1 - 1999
N2 - The ERS-1 Satellite was launched in July 1991 by the European Space Agency into a polar orbit at about km800, carrying a C-band scatterometer. A scatterometer measures the amount of radar back scatter generated by small ripples on the ocean surface induced by instantaneous local winds. Operational methods that extract wind vectors from satellite scatterometer data are based on the local inversion of a forward model, mapping scatterometer observations to wind vectors, by the minimisation of a cost function in the scatterometer measurement space.par This report uses mixture density networks, a principled method for modelling conditional probability density functions, to model the joint probability distribution of the wind vectors given the satellite scatterometer measurements in a single cell (the `inverse' problem). The complexity of the mapping and the structure of the conditional probability density function are investigated by varying the number of units in the hidden layer of the multi-layer perceptron and the number of kernels in the Gaussian mixture model of the mixture density network respectively. The optimal model for networks trained per trace has twenty hidden units and four kernels. Further investigation shows that models trained with incidence angle as an input have results comparable to those models trained by trace. A hybrid mixture density network that incorporates geophysical knowledge of the problem confirms other results that the conditional probability distribution is dominantly bimodal.par The wind retrieval results improve on previous work at Aston, but do not match other neural network techniques that use spatial information in the inputs, which is to be expected given the ambiguity of the inverse problem. Current work uses the local inverse model for autonomous ambiguity removal in a principled Bayesian framework. Future directions in which these models may be improved are given.
AB - The ERS-1 Satellite was launched in July 1991 by the European Space Agency into a polar orbit at about km800, carrying a C-band scatterometer. A scatterometer measures the amount of radar back scatter generated by small ripples on the ocean surface induced by instantaneous local winds. Operational methods that extract wind vectors from satellite scatterometer data are based on the local inversion of a forward model, mapping scatterometer observations to wind vectors, by the minimisation of a cost function in the scatterometer measurement space.par This report uses mixture density networks, a principled method for modelling conditional probability density functions, to model the joint probability distribution of the wind vectors given the satellite scatterometer measurements in a single cell (the `inverse' problem). The complexity of the mapping and the structure of the conditional probability density function are investigated by varying the number of units in the hidden layer of the multi-layer perceptron and the number of kernels in the Gaussian mixture model of the mixture density network respectively. The optimal model for networks trained per trace has twenty hidden units and four kernels. Further investigation shows that models trained with incidence angle as an input have results comparable to those models trained by trace. A hybrid mixture density network that incorporates geophysical knowledge of the problem confirms other results that the conditional probability distribution is dominantly bimodal.par The wind retrieval results improve on previous work at Aston, but do not match other neural network techniques that use spatial information in the inputs, which is to be expected given the ambiguity of the inverse problem. Current work uses the local inverse model for autonomous ambiguity removal in a principled Bayesian framework. Future directions in which these models may be improved are given.
KW - Satellite
KW - European Space Agency
KW - polar orbit
KW - scatterometer
KW - ripples
KW - local winds
KW - local inversion
KW - forward model
KW - mapping
KW - cost function
KW - density networks
KW - conditional probability density functions
KW - joint probability distribution
KW - kernels
KW - Gaussian mixture model
KW - geophysical knowledge
KW - ambiguity removal
KW - Bayesian framework
M3 - Project report
SN - NCRG/99/005
BT - First year qualifying report: neural networks for extracting wind vectors from satellite scatterometer data
PB - Aston University
CY - Birmingham, UK
ER -