Energy benefits of on-board behaviour classification for animal-borne sensor applications

The ability to study animal behaviour is important in many fields of science, including behavioural ecology, conservation and precision farming. These studies typically employ biotelemetry tags attached to animals to collect raw sensor data from tri-axial accelerometers. The lifespan of such tags is constrained by their power usage and is often a limiting factor when performing behavioural studies for extended periods of time. This study considers the effect on power requirements when performing statistical behaviour classification on the tag itself, as opposed to at a later stage, after raw data transmission. Such animal-borne classification is particularly attractive when live behavioural updates are required. Experiments using specially-designed low-power biotelemetry sensors demonstrated a 27fold reduction in energy consumption when classification was performed on the tag, as opposed to conventional post-processing techniques. By performing on-board statistical behaviour classification, the power requirements are drastically reduced, thereby prolonging the lifespan of the tag.