Scaling features of intermittent dynamics: Differences of characterizing correlated and anti-correlated data sets

摘要

Using the detrended fluctuation analysis (DFA), we consider the effect of repetitive switching between different states in complex systems, including random processes, intermittent behavior of coupled chaotic oscillators, and the dynamics of blood pressure in rats. We address the problem of diagnosing the state of a system based on time series, when the latter includes data segments with distinct correlation properties, and show significant distinctions in the diagnostics of correlated and anti-correlated data sets. We demonstrate that anti-correlated dynamics are highly sensitive to switching between different states of the system, and the presence of several "alien" segments can provide a much stronger displacement of the scaling exponent unlike the case of correlated dynamics. (C) 2019 Elsevier B.V. All rights reserved.