A neural network approach to correcting nonlinearity in optical interferometers

Real interferometers with phase quadrature detecting subsystems usually demonstrate nonlinearity with length measurements. Conventional nonlinearity correction techniques based on elliptical fittings are reviewed and their limitations are investigated using computer simulations. A new approach based...

Full description

Saved in:
Bibliographic Details
Published in:Measurement science & technology 2003-03, Vol.14 (3), p.376-381
Main Authors: Li, Zhi, Herrmann, Konrad, Pohlenz, Frank
Format: Article
Language:English
Subjects:
Exact sciences and technology
Instruments, apparatus, components and techniques common to several branches of physics and astronomy
Interferometers
Optical instruments, equipment and techniques
Physics
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Real interferometers with phase quadrature detecting subsystems usually demonstrate nonlinearity with length measurements. Conventional nonlinearity correction techniques based on elliptical fittings are reviewed and their limitations are investigated using computer simulations. A new approach based on neural networks (NNs) for correcting nonlinearity in optical interferometers for length and displacement measurements is introduced. The principle of the method is given, including the architecture of the NN and its training method. An experimental set-up was developed based on a differential plane mirror interferometer for testing the proposed method. The experimental results show that this new approach is successful when applied to real, noisy interferometer signals.
ISSN:0957-0233
1361-6501
DOI:10.1088/0957-0233/14/3/317