INVESTIGADORES
RESTREPO RINCKOAR Juan Felipe
artículos
Título:
Automatic estimation of attractor invariants
Autor/es:
RESTREPO, JUAN F.; SCHLOTTHAUER, GASTÓN
Revista:
NONLINEAR DYNAMICS
Editorial:
SPRINGER
Referencias:
Lugar: Berlin; Año: 2017
ISSN:
0924-090X
Resumen:
The invariants of an attractor have beenthe most used resource to characterize a nonlineardynamics. Their estimation is a challenging endeavorin short-time series and/or in presence of noise. In thisarticle, we present two new coarse-grained estimatorsfor the correlation dimension and for the correlationentropy. They can be easily estimated from the cal-culation of two U-correlation integrals. We have alsodeveloped an algorithm that is able to automaticallyobtain these invariants and the noise level in order toprocess large data sets. This method has been statis-tically tested through simulations in low-dimensionalsystems. The results show that it is robust in presence ofnoise and short data lengths. In comparison with similarapproaches, our algorithm outperforms the estimationof the correlation entropy.
