Remote sensing in agricultural livestock welfare monitoring: practical considerations

Intensive monitoring of domestic cattle has many economic and welfare benefits, but to realize these, timely notifications of changes in animal condition are required. Many welfare threatening conditions and physiological events can advance at a rate for which daily updates would provide insufficient warning of onset. Wireless sensor networks can provide a platform for various monitoring devices to convey and process measurements pertaining to behavior and physiology but are placed under higher cost and robustness constraints than in applications typical in other industry sectors. A wireless sensor attached to livestock must not only have lengthy battery life, thus minimizing costly maintenance, it may also have to transmit data over long ranges. While ad-hoc
wireless sensor networks using a multi-hop strategy are a potential solution, it is
not clear whether they would provide a reliable path for early stage welfare
alerts; given the variability associated with animal behavior, this is a scenario
that cannot reliably be simulated. In this paper, the connectivity of a wireless
sensor network formed from a herd of free-ranging cows wearing collar
mounted sensors is modeled using sets of GPS fixes gathered from on-animal
telemetry, thus removing the need to simulate placement. Working on the basic
assumption that a single base station would be sufficient for a 6 hectare field,
the direct and multi-hop connectivities of a small herd with collars comprising
wireless sensors are compared and the practicalities and suitability of each
approach for small to medium farms discussed.