Improving the empirical mode decomposition method

Abstract

In recent years, the analysis of non-stationary signals has taken on great importance since the introduction of empirical mode decomposition (EMD) by Huang in 1998. The algorithm is based entirely on a discrete algorithm, meaning that, for now, no clear analytical interpretation exists. In this article, we study several aspects of the EMD algorithm in order to improve the decomposition. On the one hand, the sampling frequency must be optimized so as to maximize the similarity between the discrete and continuous signals, minimizing the computational cost required to apply analysis methods for non-linear and non-stationary signals. On the other hand, a solution to border effect which gives good results for signals of approximately constant, growing or decreasing amplitude near the borders is provided. Moreover, the stopping criteria must be modified to limit the amplitudes allowed to IMF. Some examples are shown at the end.

Keywords:

• empirical mode decomposition
• nonlinear process
• nonstationary

AMS Subject Classifications::

• 42C99
• 94A12
• 92C55