Skip to main content
Fig. 6 | Animal Biotelemetry

Fig. 6

From: Simulation-based validation of activity logger data for animal behavior studies

Fig. 6

ADXL362 Activity Detection. A single axis of acceleration is shown for simplicity. (a) Initially, the activity detector is in the Inactive state, continually monitoring the accelerometer for movement that exceeds the Active Threshold (upper and lower horizontal red lines), relative to an initial “reference” acceleration level (center red horizontal lines). On the three-axis ADXL362, reference accelerations are independently determined for each axis to account for biases due to static acceleration. (b) When the active threshold is exceeded in any axis, the activity detector transitions to the Active state and will monitor the accelerometer for movement that does not exceed the Inactive Threshold (upper and lower horizontal blue lines), whose baselines (center horizontal blue lines) are determined per-axis by the first sample in the Active state. If the Inactive Threshold is exceeded in any axis, its reference is set to the previous sample’s value. (c) In this example, the Activity Detector is configured to require three consecutive samples to not exceed the Inactive Threshold in any axis before returning to the Inactive state. Thus, the Activity Detector does not yet transition, as it has only encountered two samples since re-referencing that do not exceed the threshold. (d) When three consecutive samples do not exceed the Inactive threshold in any axis, the Activity Detector transitions to Inactive once again. (e) The references for the Active Threshold are then set to the value of the first sample in the Inactive state

Back to article page