Problem of insufficient sampling rate in EMD

摘要

The objective of this paper is to investigate the empirical mode decomposition (EMD) error caused by insufficient sampling rate (Signal sampling obeys the law of Nyquist sampling, but the sampling rate is not high enough) and indicate a noise assisted data to cancel the EMD error. When the intrinsic mode functions are extracted from a signal mixed noise at a certain level on the sifting algorithm, an extraordinary phenomenon, where noise submerges the EMD error, is discovered. Thus, noise assisted data is proposed to disturb the EMD error in the sifting process. In order to realize the purpose, the ensemble EMD is employed to signal processing. As a result, the intrinsic mode functions have no error caused by insufficient sampling rate. An EMD example of ball bearing vibration is presented to illustrate the validity of the approach. The investigation will be significance in the EMD decomposition on the vibration and acoustic signals with broad band.