Comparative node selection‐based localization technique for wireless sensor networks: A bilateration approach

• Published in: International journal of communication systems Vol. 33; no. 15
• Main Authors: Tabassum, Nikhath, Devanagavi, Geetha D., Biradar, Rajashekhar C., Lazarescu, Mihai T.
• Format: Journal Article
• Language: English
• Published: Chichester Wiley Subscription Services, Inc 01.10.2020
• Subjects: Accuracy; Anchors; bilateration localization; cluster formation; Clustering; Computer simulation; Environmental monitoring; Localization; Nodes; outdoor localization; received signal strength indicator; Sensors; Structural health monitoring; trilateration; Urban planning; Wireless networks; wireless sensor network; Wireless sensor networks
• ISSN: 1074-5351; 1099-1131
• DOI: 10.1002/dac.4559

Summary Wireless sensor networks find extensive applications, such as environmental and smart city monitoring, structural health, and target location. To be useful, most sensor data must be localized. We propose a node localization technique based on bilateration comparison (BACL) for dense networks, which considers two reference nodes to determine the unknown position of a third node. The mirror positions resulted from bilateration are resolved by comparing their coordinates with the coordinates of the reference nodes. Additionally, we use network clustering to further refine the location of the nodes. We show that BACL has several advantages over Energy Aware Co‐operative Localization (EACL) and Underwater Recursive Position Estimation (URPE): (1) BACL uses bilateration (needs only two reference nodes) instead of trilateration (that needs three reference nodes), (2) BACL needs reference (anchor) nodes only on the field periphery, and (3) BACL needs substantially less communication and computation. Through simulation, we show that BACL localization accuracy, as root mean square error, improves by 53% that of URPE and by 40% that of EACL. We also explore the BACL localization error when the anchor nodes are placed on one or multiple sides of a rectangular field, as a trade‐off between localization accuracy and network deployment effort. Best accuracy is achieved using anchors on all field sides, but we show that localization refinement using node clustering and anchor nodes only on one side of the field has comparable localization accuracy with anchor nodes on two sides but without clustering. The proposed bilateration method uses two reference nodes to locate an unknown sensor node by circle to circle intersection. The mirror node is eliminated by comparing its coordinates with the coordinates of the reference nodes. The proposed method needs anchor nodes only on the edge of the network to localize all the nodes.