Continuous output periodicity detector from accelerometer data in activity monitoring
Gergő Sántha, Gyula Hermann · 2014
Autocorrelation is a widely used statistical tool for determining periodicity and the fundamental period of a time-series. Calculating the autocorrelation function Rxx(k), we cross-correlate the signal with itself in order to detect non-randomness or to find repeating patterns. A continuous output, autocorrelation algorithm is proposed here which is able to detect periodicity in motor behaviour of free-roaming animals. The data stream is coming from a 3-axis accelerometer mounted on a collar.