A scatterometer neural network sensor model with input noise

Cornford, Dan, Ramage, Guillaume and Nabney, Ian T. (2000). A scatterometer neural network sensor model with input noise. Neurocomputing, 30 (1), pp. 13-21.


The ERS-1 satellite carries a scatterometer which measures the amount of radiation scattered back toward the satellite by the ocean's surface. These measurements can be used to infer wind vectors. The implementation of a neural network based forward model which maps wind vectors to radar backscatter is addressed. Input noise cannot be neglected. To account for this noise, a Bayesian framework is adopted. However, Markov Chain Monte Carlo sampling is too computationally expensive. Instead, gradient information is used with a non-linear optimisation algorithm to find the maximum em a posteriori probability values of the unknown variables. The resulting models are shown to compare well with the current operational model when visualised in the target space.

Publication DOI: https://doi.org/10.1016/S0925-2312(99)00137-X
Divisions: Engineering & Applied Sciences > Computer science
Engineering & Applied Sciences > Computer science research group
Engineering & Applied Sciences > Non-linearity and complexity research group
Additional Information: See http://eprints.aston.ac.uk/1411/
Uncontrolled Keywords: non-linear regression,input uncertainty,wind retrieval,scatterometer
Full Text Link: http://eprints.aston.ac.uk/1411/
Related URLs:
Published Date: 2000-01
Authors: Cornford, Dan ( 0000-0001-8787-6758)
Ramage, Guillaume
Nabney, Ian T. ( 0000-0003-1513-993X)


Item under embargo

Export / Share Citation


Additional statistics for this record