Proceedings of the 2013 International Conference on Advanced Computer Science and Electronics Information (ICACSEI 2013)

An Anomaly Detection Method for Spacecraft Using ICA Technology

Authors
Bing lin Xu, Zhan huai Li
Corresponding author
Xu
Keywords
Data Process, Spacecraft Heath Monitoring, Independent Component Analysis, Fault Diagnose
Abstract
Anomalies, outlier and fault means that the received data do not fit the common pattern, and may often be occur in spacecraft missions because of the high complexity and rugged operating environment of spacecraft. In order to reveal the potential faults or anomalies many methods and technology have been researched and achieved fruitful results. In this paper, we proposed an anomaly detection method for Spacecraft based on ICA technology. The independent component analysis (ICA) is recently developed in order to extract possible independent component and reveal the hidden anomaly pattern and can be used in the anomaly detection for spacecraft. By Applying ICA technology on spacecraft telemetry data, we can reveal the hidden anomalies in telemetry data which can not be accomplished by the ordinary method. The fastICA algorithm is used when doing ICA on real spacecraft telemetry. The simulation result and the application on real telemetry data prove the effectiveness of the method.
Download article (PDF)