Constrained Optimal Placements of Heterogeneous Range/Bearing/RSS Sensor Networks for Source Localization with Distance-Dependent Noise

摘要

Plenty of practical applications require accurate emitting or reflective source localization using wireless sensor networks (WSNs). The geometric placement of WSNs is known to significantly affect the source localization results. In this letter, optimal placements of heterogeneous sensor networks (HSNs) comprised of range/bearing/received signal strength sensors are studied. The measurement noise of the sensors is set to be distance dependent. The maximization of the determinant of the Fisher information matrix is chosen as the optimality criterion. First, the sensors are assumed to be deployable in an annulus as usually done in the literature, and it is found that the bearing deployment range can be significantly reduced (even to 0) if bearing sensors are involved. Then, the HSN placements in constrained regions, specifically in a segmental arch, on a straight line, or in an external closed region, are investigated. Plentiful optimal placements are observed under the conditions of HSN, distance-dependent noise, and constrained regions.