Node Selection for Target Tracking using Bearing Measurements from Unattended Ground Sensors

摘要

Efficient resource management is necessary to optimize the lifetime of a network of wireless sensor networks. In this paper, we consider a collection of acoustical unattended ground sensors collecting bearing measurements and tracking a target. To conserve power, we describe and test different node selection methods. These selection techniques use the predicted target state from a tracker to determine a subset of nodes in the network that provide a good viewing geometry to localize the target at the present snapshot time. The selected nodes actively sense and communicate bearing measurement for the current snapshot. We compare the different selection approaches by integrating them into an extended Kalman filter and evaluating the respective localization accuracy through extensive simulations.