Abstract: Localization gets more interest because it is very important for many applications of a wireless sensor networks (WSNs). This paper presents three contributions according to centroid localization algorithm (CLA). The first one is improving the localization accuracy of sensor nodes using the nearest three anchors technique; this proposed algorithm is called enhanced centroid localization algorithm (ECLA). The second contribution which is called centroid localization algorithm using reference nodes (CLAR), aims to increase the percentage of localized sensor nodes (i.e. Coverage) by using reference nodes (localized sensor nodes) broadcasting their estimated position to all unknown sensor nodes (i.e. non-localized nodes). But from simulation results CLAR has higher localization error than CLA, so that the objective is to have low localization error and high coverage which is done by applying the nearest three anchors and/or reference nodes. The final proposed algorithm is called enhanced centroid localization algorithm using reference nodes (ECLAR). The work done in this paper is based on NS-2 simulation by applying different scenarios then measuring the performance of the proposed algorithms. Simulation results show that enhanced centroid localization algorithm using reference nodes (ECLAR) has a better performance (lower localization error and higher percentage of localized sensor nodes) than the conventional centroid localization algorithm (CLA).
Keywords: Centroid, Localization, NS-2, WSNs.