Simultaneous Sporadic Sensor Anomaly Detection for Smart Homes

What is it about?

Sensor anomalies hinder deploying smart home applications, such as energy management, security, and healthcare. We propose a method to detect sporadic simultaneous sensor anomalies in smart home sensor networks. We leverage Hypersphere Classification (HCS), the state-of-the-art anomaly detection technique based on Deep Neural Network (DNN).

Featured Image