Issue No. 12 - December (2008 vol. 20)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TKDE.2008.114
Mo Li , Hong Kong University of Science and Technology, Hong Kong
Lei Chen , Hong Kong University of Science and Technology, Hong Kong
Yunhao Liu , Hong Kong University of Science and Technology, Hong Kong
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 non-threshold 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 spatio-temporal 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.
Distributed databases, Distributed networks
Mo Li, Lei Chen, Yunhao Liu, "Nonthreshold-Based Event Detection for 3D Environment Monitoring in Sensor Networks", IEEE Transactions on Knowledge & Data Engineering, vol. 20, no. , pp. 1699-1711, December 2008, doi:10.1109/TKDE.2008.114