Pasturemanagermobility: modelo de mobilidade de animais em pastagem para redes de sensores sem fio utilizando modelo Markoviano de percurso aleatório e estados finitos de cadeias de Markov

Abstract

The use of collars containing sensors for the monitoring and tracking of grazing ani-mals, especially for cattle herds, has been gaining support in Brazil and worldwide. The monitoring of the animals is very important in terms of optimization of the gains, health care and quality of products obtained throughout the production chain. However, the real-time monitoring, made possible by the transmission of the information collected by the collars with communication stations, is even more effective, since it allows for faster decision making, before possible problems happen or opportunities are lost. However, testing and deploying these devices in the real world are extremely costly and complex. Because of this, the use of wireless sensor network simulators becomes essential for the development of this type of technology. The present work describes and implements a configurable mobility model that approximates the mobility performed by animals that organize themselves in herds, using finite states in Markov chains. This model was imple-mented in the Wireless Sensor Networks simulator called Castalia, due to its reliability with the measurement of spent energy, one of the major problems faced by the WSN, and its ease of implementation of medium access and routing protocols, showing thus a great environment for the realization of the desired tests.