Cattle Movement Monitoring and Location Prediction System Using Markov Decision Process and IoT Sensors

Abstract

Though many research contributions have been made on cattle health monitoring, those techniques can only apply to specific geographical ranges. Another big challenge in the sensor networks is the limited amount of battery power in the sensor nodes. This paper proposes a cattle movement monitoring and location prediction system using the Markov decision process (LPS-MDP). In this system, the cattle's location is tracked by monitoring its movement along a geographical boundary. The current location and movement velocity are gathered through the cows' collars, from which their usual mobility pattern is determined. The probability of a cattle moving outside its boundary is determined based on the counted movement pattern using the Markov decision process. When the probability exceeds a specified threshold, the farmer is notified. Experimental results have proven that LPS-MD minimizes the prediction cost and delay and increases the prediction accuracy.