基于局部Neymark分解的混沌信号与噪声的分离方法

从动力系统和几何的观点出发，提出了基于局部Ｎｅｙｍａｒｋ分解的混沌信号与噪声的分离方法。该方法在一个局部的轨道空间内进行Ｎｅｙｍａｒｋ分解，重构时消除由噪声引起的在某些方向上的错误分量，接着进行反嵌入，使得这一由于噪声影响而发生畸变的局部轨道得以纠正，达到混沌信号与噪声分离（以下简称混沌滤波）的目的，并对其中的反嵌入问题作了较为深入的讨论，仿真实验表明了本方法的有效性。

An Noise Reduction Method for Chaotic Signals Based on Local Neymark Decomposition

In view of dynamic system and geometry, this paper proposes a noise reduction method for chaotic signals based on local Neymark decomposition. The essence of this method is that it calculates Neymark decomposition in a local state space, reconstruct the embedding vectors to eliminate the erroneous components engendered by noise, and then by disembedding correct the distorted points due to noise. The disembedology is discussed in detail and the effectiveness of this method is proved by simulating tests.