Method and apparatus to detect ambulation
Low frequency harmonics detected by an accelerometer in a wrist-mounted device are used to infer ambulation. The fundamental rhythm or pitch is extracted from gathered accelerometer data and the determined pitch and associated properties are used to infer whether a subject is walking or not walking.
1. An apparatus comprising:
an accelerometer in close proximity to a wrist; and
an ambulation detector to receive a signal having low-frequency harmonics from the accelerometer and to extract a measure of pitch variance and goodness of a pitch estimate of the received signal in order to infer ambulation.
2. The apparatus of claim 1 , wherein the ambulation detector is configured to sample the signal to provide a plurality of frames of sample magnitude values.
3. The apparatus of claim 2 , wherein the ambulation detector is configured to extract the measure of pitch variance and average goodness of pitch estimate from the plurality of frames of sample magnitude values.
4. The apparatus of claim 1 , wherein the ambulation detector is configured to infer ambulation based on comparing the extracted measure of pitch variance and average goodness of pitch estimate with learned values of pitch variance and goodness of pitch estimate stored in a decision tree classifier.
5. The apparatus of claim 1 , wherein the accelerometer has three axis and the ambulation detector is configured to receive a plurality of signals having low-frequency harmonics, each of the plurality of signals received from a different axis, and the ambulation detector to compute a magnitude signal by combining the three axis from which to extract the measure of pitch variance and average goodness of pitch estimate.