Development of a methodology for identifying anomalies in telemetry data of small spacecraft using the ensemble method

Northern (Arctic) Federal University, Northern Dvina emb., 17, Arkhangelsk, 163000, Russia

^* Corresponding author: r.aleshko@gmail.com

Abstract

This study includes work related to the analysis of satellite telemetry data. Prior to that, the focus is on investigating various ways of finding anomalies in telemetry data. Both traditional statistical approaches, such as outlier analysis and process control, and modern machine learning techniques, such as deep learning algorithms and anomaly detection techniques, are covered in this. It is seeking to decide on a machine learning model that can identify anomalies in telemetry data. This requires the analysis of several machine learning algorithms, their comparison, and evaluation using a small spacecraft telemetry dataset. This work needs to identify the most suitable and effective methods for detecting anomalies in satellite telemetry data.