Templates for the solution of algebraic eigenvalue problems: a practical guide
Templates for the solution of algebraic eigenvalue problems: a practical guide
Proceedings of the 3rd international conference on Embedded networked sensor systems
Telos: enabling ultra-low power wireless research
IPSN '05 Proceedings of the 4th international symposium on Information processing in sensor networks
The design and evaluation of a hybrid sensor network for Cane-Toad monitoring
IPSN '05 Proceedings of the 4th international symposium on Information processing in sensor networks
Fidelity and yield in a volcano monitoring sensor network
OSDI '06 Proceedings of the 7th USENIX Symposium on Operating Systems Design and Implementation - Volume 7
LUSTER: wireless sensor network for environmental research
Proceedings of the 5th international conference on Embedded networked sensor systems
Using Echo State Networks for Anomaly Detection in Underground Coal Mines
IPSN '08 Proceedings of the 7th international conference on Information processing in sensor networks
Spatiotemporal anomaly detection in gas monitoring sensor networks
EWSN'08 Proceedings of the 5th European conference on Wireless sensor networks
Spatio-temporal outlier detection in precipitation data
Sensor-KDD'08 Proceedings of the Second international conference on Knowledge Discovery from Sensor Data
An adaptive sensor mining framework for pervasive computing applications
Sensor-KDD'08 Proceedings of the Second international conference on Knowledge Discovery from Sensor Data
Meeting ecologists' requirements with adaptive data acquisition
Proceedings of the 8th ACM Conference on Embedded Networked Sensor Systems
| Hi-index | 0.00 |
Sensor networks deployed for scientific data acquisition must inspect measurements for faults and events of interest. Doing so is crucial to ensure the relevance and correctness of the collected data. In this work we unify fault and event detection under a general anomaly detection framework. We use machine learning techniques to classify measurements that resemble a training set as normal and measurements that significantly deviate from that set as anomalies . Furthermore, we aim at an anomaly detection framework that can be implemented on motes, thereby allowing them to continue collecting scientifically-relevant data even in the absence of network connectivity. The general consensus thus far has been that learning-based techniques are too resource intensive to be implemented on mote-class devices. In this paper, we challenge this belief. We implement an anomaly detection algorithm using Echo State Networks (ESN), a family of sparse neural networks, on a mote-class device and show that its accuracy is comparable to a PC-based implementation. Furthermore, we show that ESNs detect more faults and have fewer false positives than rule-based fault detection mechanisms. More importantly, while rule-based fault detection algorithms generate false negatives and misclassify events as faults, ESNs are general , correctly identifying a wide variety of anomalies.