Strategies to overcome border area effects of coarse grained localization

Localization of sensor nodes is one of the key issues in wireless sensor networks. Next to the ability, to assign a phenomenon to a position, localization is a precondition for sensor network algorithms like geographic clustering and routing. A simple approach for coarse grained localization is centroid localization (CL), which was firstly presented by Bulusu et al. and assumes regularly arranged beacons. The localization accuracy was improved by various centroid-based algorithms, which use approximate distances to improve location estimation through weighting beacons in range, e.g. weighted centroid localization (WCL). Nevertheless, all these approaches have in common an increased localization error near network borders. In this work, we investigate this error and present two strategies to reduce the localization error of border area nodes.