Classification of acceleration waveform in a continuous walking record

We attempted to distinguish walking on level ground from walking on a stairway using a waist acceleration signal. A triaxial accelerometer was fixed to the waist and the three acceleration signals were recorded by a portable data logger at a sampling rate of 256 Hz. Twenty healthy male subjects were asked to walk through a corridor, or up and down a stairway continuously without any instruction. The data were analyzed using discrete wavelet transform. Walking patterns were classified according to two steps. At the first step, the times when walking pattern changed were detected using the low-frequency component of the anterio-posterior acceleration (LF_A) and of the vertical acceleration (LF_V). At the second step, three types of walking patterns were distinguished by comparing powers of wavelet coefficients in the vertical direction (P_WCV) and in the anterio-posterior direction (P_WCA). Changes in walking patterns could be detected by using both LF_A and LF_V . Walking down stairs could be distinguished from the other types of walking by the largest value in P_WCV and walking up stairs could be discriminated from level walking by P_WCA. Level and stairway walking could be distinguished by waist acceleration