Study on Parameter Selection of Phase Space Reconstruction for Chaotic Time Series

Based on the principle that the selection of delay time and embedding dimension has no correlation in the process of phase space reconstruction, this paper puts forward a new method of identifying the embedding dimension by Cao Liangyue Theory (CAO Theory) after the delay time is identified by Mutual Information Function. By means of the numerical verification of a few typical examples of chaotic dynamic system, the result shows that the method can identify the valid delay time of phase space reconstruction and the best embedding dimension. In addition, a new way of identifying the chaotic signal is provided and this method can effectively reconstruct the phase space of the original system from the time series.