A method for extracting human gait series from accelerometer signals based on the ensemble empirical mode decomposition

In this paper, the ensemble empirical mode decomposition({EEMD}) is applied to analyse accelerometer signals collectedduring human normal walking. First, the self-adaptive feature of{EEMD} is utilised to decompose the accelerometer signals, thussifting out several intrinsic mode functions {(IMFs}) at disparatescales. Then, gait series can be extracted through peak detectionfrom the eigen {rm IMF} that best represents gait rhythmicity.Compared with the method based on the empirical mode decomposition({EMD}), the {EEMD}-based method has following advantages: itremarkably improves the detection rate of peak values hidden in theoriginal accelerometer signal, even when the signal is severelycontaminated by the intermittent noises; this method effectivelyprevents the phenomenon of mode mixing found in the process of{EMD}. And a reasonable selection of parameters for thestop-filtering criteria can improve the calculation speed of the{EEMD}-based method. Meanwhile, the endpoint effect can besuppressed by using the {auto regressive and moving average} modelto extend a short-time series in dual directions. The resultssuggest that {EEMD} is a powerful tool for extraction of gaitrhythmicity and it also provides valuable clues for extracting eigenrhythm of other physiological signals.