Research on an Artificial Intelligence Based Diagnosis Algorithm of BDS Telemetry Anomaly

The traditional satellite telemetry data anomaly detection method based on threshold judgement and expert knowledge base has the characteristics of a wide diagnostic threshold range and large constraint by experience, and the accuracy of the detection results cannot fully meet the application requirements of high reliability of navigation satellites. Therefore, it is urgent to develop an automatic system for satellite anomaly detection. In this work, the Auto-Encoder algorithm (AE) and Prophet algorithm are applied to BDS telemetry anomaly detection and prediction for the first time, and typical satellite anomaly telemetry mode is selected for analysis and verification. The result shows that the satellite's abnormal fluctuation trend predicted by the Prophet algorithm is consistent with the actual data abnormal distribution. This algorithm can support accurate satellite anomaly detection, it can correctly predict the changing trend of satellite telemetry data and ensure the effectiveness of anomaly detection. It will be of great significance for the timely and effective diagnosis of satellite system telemetry data anomalies in the future.