数学物理学报 ›› 2016, Vol. 36 ›› Issue (4): 783-794.

• 论文 • 上一篇    下一篇

基于EMD及非平稳性度量的趋势噪声分解方法

谭秋衡1, 吴量2,3, 李波4   

  1. 1. 国信证券博士后工作站 广东深圳 518001;
    2. 中国科学院武汉物理与数学研究所 武汉 430071;
    3. 中国科学院大学 北京 100049;
    4. 华中师范大学数学与统计学学院 武汉 430079
  • 收稿日期:2015-12-07 修回日期:2016-05-27 出版日期:2016-08-26 发布日期:2016-08-26
  • 作者简介:吴量,wuliangshine@gmail.com
  • 基金资助:

    国家自然科学基金青年基金(11201165)与国家自然科学基金(11275259,91330113)资助

Decomposition of Noise and Trend Based on EMD and Non-Stationarity Measure

Tan Qiuheng1, Wu Liang2,3, Li Bo4   

  1. 1. Postdoctoral Workstation in Guosen Securities, Guangdong Shenzhen 518001;
    2. Wuhan Institute of Physics and Mathematics, Chinese Academy of Sciences, Wuhan 430071;
    3. University of Chinese Academy of Sciences, Beijing 100049;
    4. School of Mathematics and Statistics, Central China Normal University, Wuhan 430079
  • Received:2015-12-07 Revised:2016-05-27 Online:2016-08-26 Published:2016-08-26
  • Supported by:

    Support by the National Natural Science Foundation for the Youth(11201165) and the NSFC(11275259,91330113)

摘要:

该文结合非平稳性度量,研究利用经验模态分解算法进行趋势噪声分解,提出基于非平稳性度量的准则来判定舍弃的本性模态函数的数目. 通过数值模拟证明了该准则克服了连续均方误差准则的缺陷,在不同噪声强度和复杂趋势下,都能够达到很好的去噪效果.

关键词: 趋势噪声分解, 经验模态分解, 非平稳性度量

Abstract:

In this paper, we study the decomposition of trend and noise based on empirical mode decomposition algorithm and non-stationarity measure, and propose a criterion to choose intrinsic mode function for trend. Numerical simulation results show that the proposed criterion can overcome drawback of continuous mean square error criterion, achieve a good effect in different noise intensity and complex trend.

Key words: Decomposition of noise and trend, Empirical mode decomposition, Non-stationarity measure

中图分类号: 

  • O211.64