Location Discovery in Wireless Sensor Networks (LD)

The Location Discovery (LD) is one of the fundamental problems for Wireless Sensor Networks (WSN). When a WSN is deployed in an uncontrolled fashion (e.g.: when nodes are dropped from an aircraft), an automatic locationing mechanism becomes a necessity.

A popular technique consists in equipping a subset of the sensor nodes (called beacon nodes) with GPS positioning hardware, and use some distance measurement technique (received signal strength, time difference of arrival, etc.) to determine a set of node-to-node distances. A sistem can be then defined by merging the information obtained from the beacons and the node-to-node distance measurements. Obtaining the node locations that best fit this system is known as Location Discovery.

There generally exists some error in both the beacon locations and the distance measurements. These errors result in location errors after LD is performed. When the errors in the available information produce errors in the node locations that exceed a given threshold the whole process becomes uneffective; in this scenario it is interesting to determine an error model for the initial information (beacons and internode distances) that enables for a higher accuracy LD solving.