Initial Attempt on Wi-Fi CSI Based Vibration Sensing for Factory Equipment Fault Detection

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Wi-Fi signal based detection is widely implemented in indoor action detection because of its low-cost and easy implementation. But it is still rarely used in equipment vibration detection. Moreover, it is hard to detect multiple targets where we need to monitor multiple equipments' vibration state such as in the factory environment. In this paper, we propose a wireless based vibration sensing method using Wi-Fi for factory equipment fault detection. First, we use CSI amplitude data to distinguish sensing target equipments. Then, we apply an anomaly detection method to detect faulty machine operation. We conducted initial experiments to validate the feasibility of our proposed fault detection method. The experimental results show that our method detected abnormal situations with an accuracy of 100%, while 10% of normal situations were mistakenly recognized as abnormal.

Original languageEnglish
Title of host publicationICDCN 2021 - Adjunct Proceedings of the 2021 International Conference on Distributed Computing and Networking
PublisherAssociation for Computing Machinery
Pages163-168
Number of pages6
ISBN (Electronic)9781450381840
DOIs
Publication statusPublished - Jan 5 2021
Event22nd International Conference on Distributed Computing and Networking, ICDCN 2021 - Virtual, Online, Japan
Duration: Jan 5 2021Jan 8 2021

Publication series

NameACM International Conference Proceeding Series

Conference

Conference22nd International Conference on Distributed Computing and Networking, ICDCN 2021
CountryJapan
CityVirtual, Online
Period1/5/211/8/21

All Science Journal Classification (ASJC) codes

  • Human-Computer Interaction
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Software

Fingerprint Dive into the research topics of 'Initial Attempt on Wi-Fi CSI Based Vibration Sensing for Factory Equipment Fault Detection'. Together they form a unique fingerprint.

Cite this