中国物理B ›› 2010, Vol. 19 ›› Issue (6): 60514-060514.doi: 10.1088/1674-1056/19/6/060514
张淑清, 赵玉春, 贾健, 张立国, 上官寒露
Zhang Shu-Qing(张淑清)† , Zhao Yu-Chun(赵玉春), Jia Jian(贾健), Zhang Li-Guo(张立国), and Shangguan Han-Lu(上官寒露)
摘要: Phase space reconstruction is the first step to recognizing the chaos from observed time series. On the basis of differential entropy, this paper introduces an efficient method to estimate the embedding dimension and the time delay simultaneously. The differential entropy is used to characterize the disorder degree of the reconstructed attractor. The minimum value of the differential entropy corresponds to the optimum set of the reconstructed parameters. Simulated experiments show that the original phase space can be effectively reconstructed from time series, and the accuracy of the invariants in phase space reconstruction is greatly improved. It provides a new method for the identification of chaotic signals from time series.
中图分类号: (Time series analysis)