Iterative Virtual Force Localization Based on Anchor Selection for Three-Dimensional Wireless Sensor Networks

Abstract

Network localization is an emerging paradigm that enables high-accuracy location awareness in global positioning system (GPS)-challenged environments. The existing localization methods suffer from low accuracy, inflexible node distribution, and high hardware cost, especially in three-dimensional (3D) environment. To develop an efficient 3D localization method, this paper proposes the iterative virtual force (IVF) localization based on anchor selection. To eliminate the distance estimation error of range-free measurement, the K-means clustering (KMC) was introduced to exclude anchors with distance outliers, forming a set of selected anchors, which is enclosed as a virtual space. Based on the centroid of the virtual space, a drifting coefficient was defined, and the balance point of virtual force was deduced. The centroid was drifted to the balance point, and used to replace the anchor with the farthest distance to agent, producing a new set of selected anchors. In this way, the IVF method iteratively localizes the agent. In addition, the authors configured the threshold of change rate, and proved the condition of faster convergence. Simulation results show that the IVF method excels in localization accuracy, hardware cost, and computing complexity. Our research widens the application scope and improves the robustness of IVF localization in complex 3D WSNs.

Keywords

• iterative virtual force
• K-means clustering (KMC)
• network localization
• three-dimensional (3D) wireless sensor networks (WSNs)