Empirical Mode Decomposition (EMD) is a general purpose signal processing algorithm with no analytic counterpart.
It aims to decompose a signal into elemental ones, called Intrinsic Mode Functions, IMFs. Each IMF can be seen as an AMFM band pass signal.
EMD can be used to find modes in bio signals, in mechanical equipment, in communications and also as a preprocessing stage for characteristics detection in pattern recognition.
In this illustrative example, EMD is used to estimate the true analog signal, after quantification, in order to improve the signal to noise ratio.

An innovative EMD algorithm has been implemented.