Research on the Phase Space Reconstruction Method of Chaotic Time Series

Abstract

The puzzle over the recognition of the quality of the Chaotic Dynamics based on single variable time series brings forward the new method of phase space reconstruction—identify the embedding dimension with the FNN after delay time is fixed through the method of autocorrelation function. By means of the numerical verification of a few typical examples of chaotic dynamic system, the result shows that this method can be able to efficiently reconstruct the phase space of the original system out of the time series and relatively completely reduce the dynamic characteristics of the original system and thus the validity of the method is testified on chaotic signal recognition.