A New Approach to Empirical Mode Decomposition Based on Akima Spline Interpolation Technique

Abstract

The objective of this research work is to extend the scope of the empirical mode decomposition (EMD) algorithm, as an efficient tool to decompose the nonlinear and non-stationary time series. For EMD to be widely applicable, the extension utilizes both clean and noisy data sets. When constructing upper and lower envelopes, the proposed algorithm utilizes the Akima spline interpolation technique rather than a cubic spline. The proposed EMD is called Akima-EMD, which is used to identify non-informative fluctuations in the signal, such as noise, outliers, and ultra-high frequency components, and to break down the clean and chaotic data into various components to avoid distortion. It has been shown through synthetic as well as real-world time series data analysis that the proposed method successfully extracts noise in the form of the first IMF from the data.