中国物理B ›› 2005, Vol. 14 ›› Issue (1): 49-54.doi: 10.1088/1009-1963/14/1/011
肖先赐1, 张家树2, 李恒超2
Zhang Jia-Shu (张家树)a, Li Heng-Chao (李恒超)a, Xiao Xian-Ci (肖先赐)b
摘要: In this paper a local discrete cosine transformation (DCT) domain Volterra prediction method is proposed to predict chaotic time series, where the DCT is used to lessen the complexity of solving the coefficient matrix. Numerical simulation results show that the proposed prediction method can effectively predict chaotic time series and improve the prediction accuracy compared with the traditional local linear prediction methods.
中图分类号: