|
ナカモト ユキカズ
Yukikazu Nakamoto
中本 幸一 所属 ノートルダム清心女子大学 情報デザイン学部 情報デザイン学科 職種 教授 |
|
| 言語種別 | 英語 |
| 発行・発表の年月 | 2017/06 |
| 形態種別 | 研究論文（国際会議プロシーディングス） |
| 査読 | 査読あり |
| 標題 | Faulty Sensor Data Detection in Wireless Sensor Networks Using Logistical Regression |
| 執筆形態 | 共著 |
| 掲載誌名 | Proceedings - IEEE 37th International Conference on Distributed Computing Systems Workshops, ICDCSW 2017 |
| 出版社・発行元 | Institute of Electrical and Electronics Engineers Inc. |
| 巻・号・頁 | pp.13-18 |
| 著者・共著者 | Tianyu Zhang,Qian Zhao,Yukikazu Nakamoto |
| 概要 | Wireless sensor networks (WSNs) are commonly used to monitor changes in an environment and prevent disasters such as structural instability, forest fires, and tsunami. WSNs should rapidly respond to changes, and must process and analyze sensor data in a distributed way to minimize battery consumption. On the other hand, machine learning (ML) algorithms are a powerful tool for data analyzing. However, ML algorithms are so complex that cannot be executed on resource constrained sensor nodes. Another challenge of using ML algorithms in WSNs is that ML algorithms are difficult to be distributed on every sensor node. Because ML algorithms are based on statistics' methods that need collecting amount of data to approach accuracy. In this paper, we propose a method that divides a logistical regression ML method into two steps, then distributes the two steps into sink nodes and sensor nodes to detect faulty sensor data. |
| DOI | 10.1109/ICDCSW.2017.37 |