Anomaly Detection Framework for Rotational Mechanisms in Satellite Laser Communication Based on Periodic Dilated Convolution and Channel Attention Networks

A high-precision attitude control system is essential for ensuring the connectivity and stability of satellite laser communication systems, directly influencing the reliability of communication links and the consistency of data transmission. This system relies on precise rotational mechanisms that continuously adjust satellite orientation to maintain alignment with laser communication links. This paper presents an innovative anomaly detection framework that integrates Periodic Dilated Convolutional Networks (PDCN) and an Enhanced Efficient Channel Attention Network (EECA) to identify potential anomalies in these critical rotational mechanisms within the attitude control system. The proposed method capitalizes on the periodic characteristics of attitude control data by extracting key periodic components and combining them with periodic dilated convolution, thus enhancing the model's ability to capture essential time-series features and improving its accuracy in detecting fluctuations that may impact communication stability. Additionally, the inclusion of EECA further amplifies the model's focus on key anomaly signals, particularly those that could disrupt the connectivity and stability of laser communication links. This enhanced attention mechanism enables the model to highlight critical data features, significantly improving predictive accuracy and anomaly detection capabilities. Experimental results on real satellite telemetry datasets demonstrate that, compared to conventional neural network models, the proposed method excels in identifying subtle fault signatures and providing early warnings, achieving improvements of several percentage points in both prediction accuracy and anomaly detection rates. Furthermore, this approach has extensive applicability for detecting anomalies in other systems with periodic variation characteristics, providing robust support for maintaining the reliability of satellite communication systems.

Authors

Institutions

Publication Details

Journal
IEICE Transactions on Communications
Published
2025-05-02
DOI
https://doi.org/10.23919/transcom.2025ebp3008
Citations
1
Primary Topic
Fractal and DNA sequence analysis
Type
article
Field-Weighted Citation Impact
0.81
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Anomaly Detection Framework for Rotational Mechanisms in Satellite Laser Communication Based on Periodic Dilated Convolution and Channel Attention Networks

Ye Zhu, Qing Lan, Baojun Lin, Y.H. Zuo et al.
1 citations
IEICE Transactions on Communications
Fractal and DNA sequence analysis
0.81
article

Anomaly Detection Framework for Rotational Mechanisms in Satellite Laser Communication Based on Periodic Dilated Convolution and Channel Attention Networks

Ye Zhu, Qing Lan, Baojun Lin, Y.H. Zuo, Yi Lai
article en
1 citations

Abstract

A high-precision attitude control system is essential for ensuring the connectivity and stability of satellite laser communication systems, directly influencing the reliability of communication links and the consistency of data transmission. This system relies on precise rotational mechanisms that continuously adjust satellite orientation to maintain alignment with laser communication links. This paper presents an innovative anomaly detection framework that integrates Periodic Dilated Convolutional Networks (PDCN) and an Enhanced Efficient Channel Attention Network (EECA) to identify potential anomalies in these critical rotational mechanisms within the attitude control system. The proposed method capitalizes on the periodic characteristics of attitude control data by extracting key periodic components and combining them with periodic dilated convolution, thus enhancing the model's ability to capture essential time-series features and improving its accuracy in detecting fluctuations that may impact communication stability. Additionally, the inclusion of EECA further amplifies the model's focus on key anomaly signals, particularly those that could disrupt the connectivity and stability of laser communication links. This enhanced attention mechanism enables the model to highlight critical data features, significantly improving predictive accuracy and anomaly detection capabilities. Experimental results on real satellite telemetry datasets demonstrate that, compared to conventional neural network models, the proposed method excels in identifying subtle fault signatures and providing early warnings, achieving improvements of several percentage points in both prediction accuracy and anomaly detection rates. Furthermore, this approach has extensive applicability for detecting anomalies in other systems with periodic variation characteristics, providing robust support for maintaining the reliability of satellite communication systems.

IEICE Transactions on CommunicationsVol. E108-B(11)
Shanghai Micro Satellite Engineering Center (CN), University of Chinese Academy of Sciences (CN), Innovation Academy for Microsatellites of Chinese Academy of Sciences
Openalex Percentile: Top 28%
Fractal and DNA sequence analysis
0.81
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.