Deployment and mobility for animal social life monitoring based on preferential attachment

The effort and time spent in wild life monitoring increases with the area and the number of objects to observe. Deployment of wireless sensor nodes enables a scalable sampling method and fine granularity of data collection. When we include resource rich actor nodes the data collection and evaluation are further optimized. However realistic mobility data is missing for various animal species to develop prototypes of wireless network based monitoring systems. In this paper, the problem of the absence of realistic data is considered from a social network perspective. Node deployment and mobility algorithms are provided to model a complete system of an animal swarm to be used for animal social life monitoring. A novel spatial cut-off preferential attachment model and center of mass concept are used and extended for the models according to the characteristics of the animal swarms. In the application scenario of a gorilla swarm, each gorilla is equipped with a sensor node for the monitoring of the social system. The local interaction patterns among the mobile nodes are also monitored throughout the network life time to observe the social interactions among animals and to determine the role of each animal in the society. The performance of the monitoring protocol and the applicability of the deployment and mobility models are presented through extensive simulations.