Sciweavers

Share
ICDCS
2007
IEEE

Non-Threshold based Event Detection for 3D Environment Monitoring in Sensor Networks

9 years 8 months ago
Non-Threshold based Event Detection for 3D Environment Monitoring in Sensor Networks
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.
Mo Li, Yunhao Liu, Lei Chen 0002
Added 03 Jun 2010
Updated 03 Jun 2010
Type Conference
Year 2007
Where ICDCS
Authors Mo Li, Yunhao Liu, Lei Chen 0002
Comments (0)
books