Experimental localization with MICA2 Motes

As research on Wireless Sensor Networks (WSNs) matures and the number of real life deployments increases, locationing (the problem of finding the physical placing of sensors without prior knowledge of their position) is attracting considerable interest. In almost any WSN application, localization knowledge within the network is a necessity: A sensor's data are only useful if put in context of where it came from - this is particularly the case in field sensing applications (i.e, the sensing of a phenomonen over an area or volume). The accuracy of localization therefore directly affects the accuracy of the information derived from the network. This paper describes the characterization of the accuracy achievable when localization is based on range data derived from acoustic time-of-flight measurements, and the consequent accuracy of the derived sensing surface.