Toward Deployable Satellite Anomaly Detection: A Benchmark Study on Large-Scale ESA-ADB Telemetry

2026-07-08

AI Summary

Satellite telemetry data present significant challenges for anomaly detection due to their high dimensionality, irregular structure, and class imbalance. This study presents the first systematic evaluation of both supervised methods—such as Multiscale CNN, Graph Convolutional Networks (GCN), and Graph Attention Networks (GAT)—and unsupervised approaches—including Elliptic Envelope and ECOD—on the large-scale, real-world ESA-ADB telemetry dataset across multiple temporal scales. The evaluation assesses detection performance, computational overhead, and scalability. Results demonstrate that supervised models generally achieve higher accuracy, whereas unsupervised methods maintain competitive performance while substantially reducing computational costs. These findings highlight a clear trade-off between precision and efficiency, offering empirical insights and practical guidance for selecting and deploying anomaly detection strategies in real-world satellite health monitoring systems.

Abstract

Satellite anomaly detection is essential for maintaining mission reliability and spacecraft health, yet remains challenging due to the high-dimensional, irregular, and imbalanced nature of spacecraft telemetry data. This paper presents a systematic benchmark study evaluating supervised and unsupervised anomaly detection approaches on the large-scale ESA-ADB dataset across two mission settings of varying temporal scales. Supervised models, including Multiscale Convolutional Neural Networks (Multiscale CNN), Graph Convolutional Networks (GCN), and Graph Attention Networks (GAT), are compared against unsupervised methods, namely Elliptic Envelope (EE) and Empirical Cumulative Distribution Function-based Outlier Detection (ECOD). Beyond detection performance, we rigorously analyze computational runtime and scalability, which are critical for practical deployment in spacecraft operations. Results show that supervised models achieve stronger overall performance, while unsupervised methods offer competitive precision with significantly lower computational overhead. These findings underscore a fundamental trade-off between detection capacity and operational efficiency, offering practical guidance for mission engineers designing scalable satellite health monitoring systems.

Problem

• satellite anomaly detection
• telemetry data
• computational scalability
• deployment
• imbalanced data

Innovation

• satellite anomaly detection
• benchmark study
• computational scalability
• ESA-ADB dataset
• deployable monitoring