Receiver-oriented design of Bloom filters for data-centric routing
Computer Networks: The International Journal of Computer and Telecommunications Networking
Level the buffer wall: Fair channel assignment in wireless sensor networks
Computer Communications
Long-term large-scale sensing in the forest: recent advances and future directions of GreenOrbs
Frontiers of Computer Science in China
Approaching the optimal schedule for data aggregation in wireless sensor networks
WASA'10 Proceedings of the 5th international conference on Wireless algorithms, systems, and applications
Event recognition via energy efficient voting for wireless sensor networks
ruSMART/NEW2AN'10 Proceedings of the Third conference on Smart Spaces and next generation wired, and 10th international conference on Wireless networking
Journal of Parallel and Distributed Computing
Optimizing event detection in low duty-cycled sensor networks
Wireless Networks
Proceedings of the 15th ACM international conference on Modeling, analysis and simulation of wireless and mobile systems
Adaptive edge detection with distributed behaviour-based agents in WSNs
International Journal of Sensor Networks
| Hi-index | 0.00 |
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.