Research on long range characteristics of human gait cycle based on electrostatic signals

A variety researches have been done on gait signal in recent years, which contains abundant information of human physical condition and health status. We measured electrostatic gait signal of 6 young men by a novel electrostatic detector instead of accelerometer. The noise was filtered out using self-correlation algorithm to sink the same phase point in every gait and to obtain accurate gait cycle series. The gait cycle series were decomposed into two component series, magnitude series (absolute value of gait cycle increment) and sign series. The detrended fluctuation analysis algorithm was applied to analyze the original gait cycle series and two component series. It is found that the gait cycle series appears positive long range correlation. For the two component series, the magnitude series show strongly positive long range correlation, while the sign series show obviously long range anti-correlation characteristics for all the tested objects.