Time Series Analysis and Its Applications (Springer Texts in Statistics)
Time Series Analysis and Its Applications (Springer Texts in Statistics)
Proceedings of the 3rd international conference on Embedded networked sensor systems
Approximate Data Collection in Sensor Networks using Probabilistic Models
ICDE '06 Proceedings of the 22nd International Conference on Data Engineering
Introduction to Probability Models, Ninth Edition
Introduction to Probability Models, Ninth Edition
ASAP: An Adaptive Sampling Approach to Data Collection in Sensor Networks
IEEE Transactions on Parallel and Distributed Systems
Model-driven data acquisition in sensor networks
VLDB '04 Proceedings of the Thirtieth international conference on Very large data bases - Volume 30
Sensor network data fault types
ACM Transactions on Sensor Networks (TOSN)
Sensor faults: Detection methods and prevalence in real-world datasets
ACM Transactions on Sensor Networks (TOSN)
Sensor network data fault detection with maximum a posteriori selection and bayesian modeling
ACM Transactions on Sensor Networks (TOSN)
Packet-level attestation (PLA): A framework for in-network sensor data reliability
ACM Transactions on Sensor Networks (TOSN)
Hi-index | 0.00 |
Various studies have shown that a substantial portion of the data gathered in real-world sensing applications is faulty. Most existing fault-detection approaches are off-line, centralised, and rely heavily on expert domain knowledge which may not always be available. The stochastic nature of physical phenomenon means that expert knowledge or historical models that reflect the physical world at some point may become stale later and give rise to a large rate of false alarms. We propose a data collection method with in-network, hierarchical, Demand-based, Adaptive Fault Detectors (DAFD). By applying a two-tiered error detection technique, the approach adapts itself to the changing physical environment. Preliminary real world implementation was done to show its feasibility for resource-restricted sensors. We demonstrate good detection accuracy in simulation while keeping the false alarm rate low.