ABSTRACT

Nodes in a sensor network are often randomly distributed. To assign measurements to locations, each node has to determine its own position. Due to limited resources of the nodes, resource-aware positioning algorithms are required.

In this paper, we present the novel “Weighted Centroid Localization” algorithm (WCL). Compared to other approximative algorithms, WCL achieves a remarkable average positioning error below 6%. Moreover, the algorithm features simple implementation and scales well in large sensor networks. In contrast to exact algorithms using analytical equations and complex matrix operations, WCL requires only a very small memory footprint.