期刊
IEEE INTERNET OF THINGS JOURNAL
卷 5, 期 6, 页码 4728-4752出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/JIOT.2018.2811741
关键词
Fuzzy logic; localization; virtual node; weighted centroid (WC); wireless sensor network (WSN)
资金
- Thailand Research Fund through the Royal Golden Jubilee Ph.D. [PHD/0214/2558]
- Department of Computer Science, Faculty of Science, Khon Kaen University
Due to their low cost, various range-free localization techniques are widely applied to estimate device locations in wireless sensor networks (WSNs), especially where other communication signals, such as those from a global positioning system, are absent. Among range-free techniques, the centroid algorithm has gained popularity because of its simplicity and low computational cost, rendering it suitable for power-sensitive sensor nodes or nodes. WSN topologies are unpredictable because sensor nodes are often deployed at arbitrary locations, making it difficult for sufficient numbers of known (anchor) nodes to cover all unknown nodes. Consequently, both centroid algorithm and its enhanced versions [e.g., weighted centroid (WC) algorithms] yield relatively high localization inaccuracy. In this paper, to successfully form anchor-node triangles, we propose a low-cost technique to determine the number of virtual anchor nodes or virtual nodes together with their positions. Specifically, we develop and provide proof for accurate approximate unknown node sides. We also improve localization accuracy by adding virtual nodes that collaborate with physical anchor nodes to provide the necessary coverage of the unknown nodes. We address estimations of unknown node locations by applying a fuzzy-based centroid localization method to prioritize anchor nodes by assigning different fine-tuned weighted factors. The results show that the proposed algorithm outperforms state-of-the-art fuzzy-based localization techniques for WC algorithms.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据