Dynamical entropy for systems with stochastic perturbation

Abstract

Dynamics of deterministic systems perturbed by random additive noise is characterized quantitatively. Since for such systems the KS-entropy diverges we analyse the difference between the total entropy of a noisy system and the entropy of the noise itself. We show that this quantity is non negative and in the weak noise limit is conjectured to tend to the KS-entropy of the deterministic system. In particular, we consider one-dimensional systems with noise described by a finite-dimensional kernel, for which the Frobenius-Perron operator can be represented by a finite matrix.

REVTeX 18 pages, 9 figures Revised section II, some minor improvements and corrections

References in corpus (4)

• Random perturbations of chaotic dynamical systems. Stability of the spectrum
• Entropy computing via integration over fractal measures
• Trace formulas for stochastic evolution operators: Weak noise perturbation theory
• Mean dynamical entropy of quantum system tends to infinity in the semiclassical limit

Cited by in corpus (6)

• Noise level estimation of time series using coarse grained entropy
• Entropy computing via integration over fractal measures
• Quantum Iterated Function Systems
• Coarse-Grained Probabilistic Automata Mimicking Chaotic Systems
• Quantization of Classical Maps with tunable Ruelle-Pollicott Resonances
• Spectrum of the Frobenius-Perron operator for systems with stochastic perturbation