摘要： 针对经验模态分解的希尔波特-黄变换在分解过程中会产生端点效应使信号两端出现失真的问题，提出包络线性延拓法。将其与其他3种端点效应处理方法进行对比研究，利用能量、分解信号与原信号相关系数以及算法的运行时间作为端点处理方法的评价标准。仿真结果表明，该方法可以使端点效应得到有效抑制。

关键词: 希尔波特-黄变换, 经验模态分解, 边界效应, 评价标准

Abstract: Aiming at the problem that end effect produced by the edge effect Hilbert-Huang Transform(HHT) in the decomposition process of Empirical Mode Decomposition(EMD) appearing distortion for double signal, this paper proposes envelop linearity extension method. Comparison and research are made among three kinds of processing method for end effect. It develops several indexes for evaluating the end effect, such as energy, the correlation coefficient of decomposed signal and original signal, and operation time. Simulation results indicate that the method can restrain the end effects effectively.

Key words: Hilbert-Huang Transform(HHT), Empirical Mode Decomposition(EMD), edge effect, evaluation criterion

中图分类号: 

• TP393