Satellite Telemetry Data Anomaly Detection with Hybrid Similarity Measures

Abstract

Anomaly detection based on telemetry data can improve the operating safety for spacecrafts. Most of the anomaly detection methods in this domain are based on Euclidean distance for similarity measure of monitoring parameters. However, the Euclidean distance has many limitations on telemetry data similarity measure and may affect the detecting performance. Therefore, improved distance measures and combined distance measures are applied in telemetry data analysis. An improved anomaly detection framework with different similarity measures are presented for multiple monitoring parameters of satellite in this paper. Then, the proposed anomaly detection approach based on the k-Nearest Neighbor (KNN) classification with improved similarity measures are applied into the actual satellite telemetry data. Experimental results show that the presented anomaly detection method can achieve satisfied performance on the actual satellite telemetry data sets.