Methods for detection and characterization of signals in noisy data with the Hilbert-Huang Transform

摘要

The Hilbert-Huang Transform is a novel, adaptive approach to time seriesanalysis that does not make assumptions about the data form. Its adaptive,local character allows the decomposition of non-stationary signals withhightime-frequency resolution but also renders it susceptible to degradationfrom noise. We show that complementing the HHT with techniques such aszero-phase filtering, kernel density estimation and Fourier analysis allows itto be used effectively to detect and characterize signals with low signal tonoise ratio.