Nonthreshold-based event detection for 3d environment monitoring in sensor networks
- Hong Kong University of Science & Technology, Kowloon (China)
Event detection is a crucial task for wireless sensor network applications, especially environment monitoring. Existing approaches for event detection are mainly based on some predefined threshold values and, thus, are often inaccurate and incapable of capturing complex events. For example, in coal mine monitoring scenarios, gas leakage or water osmosis can hardly be described by the overrun of specified attribute thresholds but some complex pattern in the full-scale view of the environmental data. To address this issue, we propose a nonthreshold-based approach for the real 3D sensor monitoring environment. We employ energy-efficient methods to collect a time series of data maps from the sensor network and detect complex events through matching the gathered data to spatiotemporal data patterns. Finally, we conduct trace-driven simulations to prove the efficacy and efficiency of this approach on detecting events of complex phenomena from real-life records.
- OSTI ID:
- 21116283
- Journal Information:
- IEEE Transactions of Knowledge and Data Engineering, Journal Name: IEEE Transactions of Knowledge and Data Engineering Journal Issue: 12 Vol. 20; ISSN 1041-4347
- Country of Publication:
- United States
- Language:
- English
Similar Records
Distributed feature extraction for event identification.
Wireless battery management control and monitoring system