An algorithm based on the data-adaptive filtering characteristics of singular spectrum analysis (SSA) is proposed to denoise chaotic data. Firstly, the empirical orthogonal functions (EOFs) and principal components (PCs) of the signal were calculated, reconstruct the signal using the EOFs and PCs, and choose the optimal reconstructing order based on sigular spectrum to obtain the denoised signal. The noise of the signal can influence the calculating precision of maximal Liapunov exponents. The proposed denoising algorithm was applied to the maximal Liapunov exponents calculations of two chaotic system, Henon map and Logistic map. Some numerical results show that this denoising algorithm could improve the calculating precision of maximal Liapunov exponent.
singular spectrum analysis denoising maximal Liapunov exponent chaotic system
Chinese Library Classification
2000 Mathematics Subject Classification
This is a preview of subscription content, log in to check access.
Kantz H, Schreiber T.Nonlinear Time Series Analysis[M] Cambridge University Press, New York, 1997.Google Scholar
Rosenstein Michael T, Collins James J, De Luca Carlo J. A practical method for calculating largest Liapunov exponents from small data sets[J].Physica D, 1993,65(1/2):117–133.MathSciNetCrossRefGoogle Scholar
Kantz Holger. A robust method to estimate the maximal Liapunov exponent of a time series[J].Physics Letters A, 1994,185(1):77–87.CrossRefGoogle Scholar