|
Due to the limited energy and bandwidth resources of Underwater Wireless Sensor Networks (UWSNs), only a small part of nodes are allowed to track underwater target at each time step. As far as we know, almost all the existing node selection schemes assume the true locations of nodes are known. However, such assumption is hard to satisfy due to the mobility of the nodes. This paper presents a new node selection scheme which only depends on the estimated locations of nodes rather than the true ones. Firstly, the uncertainty of node's location is approximated the additional measurement noise via the first-order Taylor series expansion. Then, in order to select nodes under node's location uncertainty, the relation between posterior Cramer-Rao lower bound (PCRLB) and estimated locations of nodes is derived. Then, our tracking scheme which consists of multi-sensor particle filter (PF) and the new node selection scheme is designed to track the underwater target under the uncertainty of node's location. Finally, a simulation is presented to illustrate the tracking improvements obtained by estimating target's state using our scheme. |
|
Keywords:Node selection, node's location uncertainty, target tracking, posterior Cramer-Rao lower bound, underwater wireless sensor networks. |
|