Contrastive Learning with Prototype-Based Negative Mixing for Satellite Telemetry Anomaly Detection

Telemetry data are the most important basis for ground operators to assess the status of satellites in orbit, and telemetry data-based anomaly detection has become a key tool to improve the reliability and safety of spacecrafts. Recent research on anomaly detection focuses on constructing a normal p...

Full description

Bibliographic Details
Main Authors: Guohang Guo, Tai Hu, Taichun Zhou, Hu Li, Yurong Liu
Format: Article
Language:English
Published: MDPI AG 2023-05-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/23/10/4723
_version_ 1827740093353295872
author Guohang Guo
Tai Hu
Taichun Zhou
Hu Li
Yurong Liu
author_facet Guohang Guo
Tai Hu
Taichun Zhou
Hu Li
Yurong Liu
author_sort Guohang Guo
collection DOAJ
description Telemetry data are the most important basis for ground operators to assess the status of satellites in orbit, and telemetry data-based anomaly detection has become a key tool to improve the reliability and safety of spacecrafts. Recent research on anomaly detection focuses on constructing a normal profile of telemetry data using deep learning methods. However, these methods cannot effectively capture the complex correlations between the various dimensions of telemetry data, and thus cannot accurately model the normal profile of telemetry data, resulting in poor anomaly detection performance. This paper presents CLPNM-AD, contrastive learning with prototype-based negative mixing for correlation anomaly detection. The CLPNM-AD framework first employs an augmentation process with random feature corruption to generate augmented samples. Following that, a consistency strategy is employed to capture the prototype of samples, and then prototype-based negative mixing contrastive learning is used to build a normal profile. Finally, a prototype-based anomaly score function is proposed for anomaly decision-making. Experimental results on public datasets and datasets from the actual scientific satellite mission show that CLPNM-AD outperforms the baseline methods, achieves up to 11.5% improvement based on the standard <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>F</mi><mn>1</mn></msub></semantics></math></inline-formula> score and is more robust against noise.
first_indexed 2024-03-11T03:20:11Z
format Article
id doaj.art-30f3bcaa52054839856c532edd61e3ef
institution Directory Open Access Journal
issn 1424-8220
language English
last_indexed 2024-03-11T03:20:11Z
publishDate 2023-05-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj.art-30f3bcaa52054839856c532edd61e3ef2023-11-18T03:11:35ZengMDPI AGSensors1424-82202023-05-012310472310.3390/s23104723Contrastive Learning with Prototype-Based Negative Mixing for Satellite Telemetry Anomaly DetectionGuohang Guo0Tai Hu1Taichun Zhou2Hu Li3Yurong Liu4National Space Science Center, Chinese Academy of Sciences, Beijing 101499, ChinaNational Space Science Center, Chinese Academy of Sciences, Beijing 101499, ChinaNational Space Science Center, Chinese Academy of Sciences, Beijing 101499, ChinaNational Space Science Center, Chinese Academy of Sciences, Beijing 101499, ChinaNational Space Science Center, Chinese Academy of Sciences, Beijing 101499, ChinaTelemetry data are the most important basis for ground operators to assess the status of satellites in orbit, and telemetry data-based anomaly detection has become a key tool to improve the reliability and safety of spacecrafts. Recent research on anomaly detection focuses on constructing a normal profile of telemetry data using deep learning methods. However, these methods cannot effectively capture the complex correlations between the various dimensions of telemetry data, and thus cannot accurately model the normal profile of telemetry data, resulting in poor anomaly detection performance. This paper presents CLPNM-AD, contrastive learning with prototype-based negative mixing for correlation anomaly detection. The CLPNM-AD framework first employs an augmentation process with random feature corruption to generate augmented samples. Following that, a consistency strategy is employed to capture the prototype of samples, and then prototype-based negative mixing contrastive learning is used to build a normal profile. Finally, a prototype-based anomaly score function is proposed for anomaly decision-making. Experimental results on public datasets and datasets from the actual scientific satellite mission show that CLPNM-AD outperforms the baseline methods, achieves up to 11.5% improvement based on the standard <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>F</mi><mn>1</mn></msub></semantics></math></inline-formula> score and is more robust against noise.https://www.mdpi.com/1424-8220/23/10/4723telemetry dataanomaly detectioncontrastive learningnegative mixing
spellingShingle Guohang Guo
Tai Hu
Taichun Zhou
Hu Li
Yurong Liu
Contrastive Learning with Prototype-Based Negative Mixing for Satellite Telemetry Anomaly Detection
Sensors
telemetry data
anomaly detection
contrastive learning
negative mixing
title Contrastive Learning with Prototype-Based Negative Mixing for Satellite Telemetry Anomaly Detection
title_full Contrastive Learning with Prototype-Based Negative Mixing for Satellite Telemetry Anomaly Detection
title_fullStr Contrastive Learning with Prototype-Based Negative Mixing for Satellite Telemetry Anomaly Detection
title_full_unstemmed Contrastive Learning with Prototype-Based Negative Mixing for Satellite Telemetry Anomaly Detection
title_short Contrastive Learning with Prototype-Based Negative Mixing for Satellite Telemetry Anomaly Detection
title_sort contrastive learning with prototype based negative mixing for satellite telemetry anomaly detection
topic telemetry data
anomaly detection
contrastive learning
negative mixing
url https://www.mdpi.com/1424-8220/23/10/4723
work_keys_str_mv AT guohangguo contrastivelearningwithprototypebasednegativemixingforsatellitetelemetryanomalydetection
AT taihu contrastivelearningwithprototypebasednegativemixingforsatellitetelemetryanomalydetection
AT taichunzhou contrastivelearningwithprototypebasednegativemixingforsatellitetelemetryanomalydetection
AT huli contrastivelearningwithprototypebasednegativemixingforsatellitetelemetryanomalydetection
AT yurongliu contrastivelearningwithprototypebasednegativemixingforsatellitetelemetryanomalydetection