MEMD decomposition of the trivariate dada, consisting of three noise time series: the white noise (left top), the noise with positive long-range dependence (left middle) and the noise with negative long-range dependence (left bottom).

2014-06-16T03:31:01Z (GMT) by Meng Hu Hualou Liang

The left panels show the power spectra of IMFs from the each noise (the sampling rate of 1000 Hz), in which the number denotes the order of IMF components. The right panel shows the inverse CDFs of IMF 3 from three different noise time series. Thanks to the rank-order statistics used in the estimation, there is an excellent match among three inverse CDFs, indicating that our approach is robust to the different types of noises.