Empirical Mode Decomposition as a tool for data analysis

The recently introduced Empirical Mode Decomposition (EMD) is a powerful data analysis tool that can deal with the non-stationary and non-linear characteristics of natural phenomena data. A comparison with two popular data analysis methods, i.e., the discrete Fourier transform and the discrete wavelet transform, has been conducted through simulation experiments. It can be seen that EMD is able to capture distinct features of the non-stationary data that the other methods can not, making EMD a valuable tool that can be applied to signal analysis, modelling and denoising.