Fast Empirical Mode Decomposition Based on Gaussian Noises

Abstract

Mode-mixing, boundary effects and necessary extrema lacking and etc. are the main problems involved in empirical mode decomposition (EMD). The paper presents an improved empirical mode decomposition based on assisted signals: Gaussian noises. Firstly, the given 1D Gaussian noise and its negative counterpart are added to the original respectively to construct the two s to be decomposed. Secondly, the decomposed IMFs from the two signals are added together to get the IMFs, in which the added noises are canceled out with less mode-mixing and boundary effects. Lastly, the efficiency and performance of the method are given through theoretical analysis and experiments.