Next: Global
Modeling by Means
Appeared in the proceedings of the "International Conference on Unsolved Problems of Noise", Sept. 3-7 1996, Szeged, Hungary
1Institut für Theoretische Physik; Technische Universität
Graz
Petersgasse 16; A-8010 Graz; Austria
2Institute for Pure and Applied Physical Sciences
MS-0360, University of California at San Diego
9500 Gilman Drive; La Jolla, CA 92093; USA
Claudia S.M. Lainscsek1, Ferdinand Schürrer1, and James B. Kadtke2
The information contained in a scalar time series and its numerical derivatives is used to construct a global model for the underlying dynamical system, using a model transformation presented previously. Here, we present some modifications of this method to mitigate the effects of additive noise, test this method by reconstructing global models for known chaotic dynamical systems, and compare the dynamical properties of the re-constructed and original systems.
![]()
![]()