Anomaly Detection Service for Sensor Stream Data based on Lag-correlation Analysis

The sensor streams influence and correlate with each other, and the hidden correlation can be used to identify and explain abnormal problems. This paper proposed one kind of anomaly detection service based on lag-correlation analysis. It first constructs correlation graph model based on lag-correlation analysis of multiple sensor streams. Then sensor streams groups are constructed according to the correlation degree in the graph model, and are encapsulated as corresponding services to realize the anomaly detection within and out of stream groups in real time. Experimental results on a real industrial sensor data set show that the proposed method is effective in anomaly detection tasks in multiple sensor stream data.