Download now Free registration required
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 paper the authors unify fault and event detection under a general anomaly detection framework. They 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, they 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.
- Format: PDF
- Size: 284.3 KB