
Research Article
FaultBit: Generic and Efficient Wireless Fault Detection Using the Internet of Things
@INPROCEEDINGS{10.1007/978-3-031-63989-0_5, author={Koustabh Dolui and Ashok Samraj Thangarajan and Sergii Morshchavka and Zhaoyi Liu and Sam Michiels and Danny Hughes}, title={FaultBit: Generic and Efficient Wireless Fault Detection Using the Internet of Things}, proceedings={Mobile and Ubiquitous Systems: Computing, Networking and Services. 20th EAI International Conference, MobiQuitous 2023, Melbourne, VIC, Australia, November 14--17, 2023, Proceedings, Part I}, proceedings_a={MOBIQUITOUS}, year={2024}, month={7}, keywords={Internet of Things Wireless Sensor Networks Machine Learning Signal Processing Energy awareness}, doi={10.1007/978-3-031-63989-0_5} }
- Koustabh Dolui
Ashok Samraj Thangarajan
Sergii Morshchavka
Zhaoyi Liu
Sam Michiels
Danny Hughes
Year: 2024
FaultBit: Generic and Efficient Wireless Fault Detection Using the Internet of Things
MOBIQUITOUS
Springer
DOI: 10.1007/978-3-031-63989-0_5
Abstract
The timely monitoring and maintenance of industrial machines is critical to prevent expensive disruptions and down-time. The Internet of Things (IoT) offers a solution to instrument production processes at lower cost and complexity. However, wireless IoT networks are bandwidth constrained, which precludes the transmission of high frequency signals such as vibration or electrical current. Prior approaches to tackling this problem require a high degree of application-specific re-engineering. In this paper, we argue for a new approach to fault detection that is accurate, efficient and applicable to a large class of fault detection problems. We propose FaultBit, an application independent toolkit for fault classification that can gather, compress and classify data using IoT networks. We evaluate FaultBit in two representative scenarios using current and vibration data. In both cases, FaultBit offers classification accuracy 99%, which is very close to application-specific classifiers, while requiring 512(\times )less bandwidth, enabling a battery life of several years.