Title :
A new approach for nonlinear time series characterization, “DivA”
Author :
Bucolo, Maide ; Grazia, Federica Di ; Sapuppo, Francesca ; Virzi, Maria C.
Author_Institution :
Dipt. di Ing. Elettr., Elettron. e dei Sist., Univ. degli Studi di Catania, Catania
Abstract :
DivA, acronym of divergence algorithm, a novel approach for the characterization of nonlinear time series has been developed. This new methodology is based on a numerical algorithm that calculates the divergence curve from which the maximum Lyapunov exponent and the asymptotic trajectories divergence can be extracted. The consistence of the algorithm, DivA, has been proved by making a comparison with another tool on chaotic systems. The novelty of the algorithms, DivA, lays on its lightness since it skips intermediate computational steps permitting its easy implementation on standalone devices for fast analysis of huge data like physiological signals.
Keywords :
Lyapunov methods; control nonlinearities; nonlinear control systems; time series; DivA; asymptotic trajectories divergence; chaotic systems; divergence algorithm; maximum Lyapunov exponent; nonlinear time series characterization; Automatic control; Automation; Chaos; Data analysis; Data mining; Differential equations; Embedded computing; Geophysical measurements; Signal analysis; Time series analysis;
Conference_Titel :
Control and Automation, 2008 16th Mediterranean Conference on
Conference_Location :
Ajaccio
Print_ISBN :
978-1-4244-2504-4
Electronic_ISBN :
978-1-4244-2505-1
DOI :
10.1109/MED.2008.4602067