Empirical Mode Decomposition: Theory and Applications in Underwater Acoustics

Empirical mode decomposition (EMD) is a signal processing method that produces data-driven time-frequency representation suited to characterize time-varying and nonlinear phenomena. In EMD, intrinsic functions (IMF) are sequentially estimated from the of interest represent different oscilation modes produce an orthogonal original information. Different algorithms have been proposed for EMD estimation deal with limitations such as mode-mixing noise sensitivity. To obtain frequency-domain representation, usually associated Hilbert transform, in this case, referred Hilbert-Huang transform (HHT). This paper presents theoretical review fundamental aspects both HHT, IMF procedure orthogonality. Variations algorithm also presented. Both simulated experimental underwater acoustic signals used illustrate efficiency EMD/HHT revealing relevant characteristics