Geometric and decentralized approach for localization in wireless sensor network

The main function of a sensor node is to collect data from its environment and forward it to base station. In the absence of further information concerning their locations, those data will be unnecessary. Hence, developing algorithms for localizing all nodes of wireless sensor network is extremely important. We present in this paper, a new approach to determine geographical coordinates of unknown nodes, by using mobile anchor. The mobile anchor adopts a spiral trajectory, and diffuses its position periodically during its travel. The proposed approach uses Received Signal Strength Indicator to estimate distance with all broadcast messages received from mobile anchor. To calculate position, our approach determines a cloud of points that surround the solution; these points are selected from the set of intersection points of all beacons received by unknown node, by considering some constraints. The estimated position of unknown node represents the geometric center of this cloud. The behavior of our algorithm was studied by varying some metrics; the average error was minimized compared to those proposed in literature.