Statistical recognition of the dynamical systems with chaotic perturbation

In this paper we consider a method of analyzing time series, generated by nonlinear dynamical systems with noise component in the form of perturbation. Noise filtering is carried out by evaluating the domain of density of the joint distribution function of values and its increments, depending on the fineness of the partitioning of the multidimensional histogram. Examples of different types of perturbations are given: additive noise, dimension perturbation, noise in the implementation of a dynamic system, random switching, injection of several processes.