经验模态分解中包络线算法  

Analysis and Study of Envelope Algorithm in EMD

作　　者：朱赛[1] 尚伟[1] 

机构地区：[1]军械工程学院,石家庄050003

出　　处：《火力与指挥控制》2012年第9期125-128,共4页Fire Control & Command Control

摘　　要：经验模态分解(EMD)是一种先进的信号处理方法,对非线性、非平稳信号具有独特的分析能力。它的包络线算法存在着过冲/欠冲和端点问题,影响信号分解质量。分析了过冲/欠冲和端点问题产生的原因,提出了采用保形分段3次插值方法作为EMD分解过程中的包络算法,解决了拟合包络线的过冲/欠冲问题;采用端点处包络线位置预测,对端点处的包络线位置加以约束,与极值点对称延拓的方法结合使用,充分利用信号本身信息,一定程度上抑制了端点问题。最后,用一个仿真实验验证了该方法的有效性。The Empirical Mode Decomposition（EMD） is an advanced method for signal analysis,and it has its own ability for processing nonlinear and non-stationary signals. But there are overshoots/ undershoots and end issue in the course of getting two envelops of the data, and the problems affect the quality of the decomposing. Based on the analyzing the reason caused the overshoots/undershoots and end issue,the shape-preserving piecewise cubic interpolation is presented as a substitute for the cubic spline interpolating. In order to improve the end issue, the paper gives the way that estimating the envelops position at the end used with symmetrical extrema extension. Finally, the result of the simulating experiment shows that the proposed method has a good result.

关 键 词：经验模态分解(EMD) 包络线算法 端点问题 极值点对称延拓 

分 类 号：TN911[电子电信—通信与信息系统]