Lightweight Attention Networks for Onboard Satellite Anomaly Detection Under Memory Constraints
Frontiers in Agriculture · 01.04.2026 · 3 (1) · 15-26
DOI: 10.71465/fa769 ↗Kurz gefasst
Fachartikel in der Zeitschrift Frontiers in Agriculture, online erschienen am 01.04.2026. Verfasst von Yuchen Fang. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The rapid expansion of low-Earth orbit satellite constellations has created urgent demand for autonomous onboard fault detection systems capable of operating within the severe memory and computational budgets imposed by space-grade embedded processors. This paper proposes a Lightweight Attention Network (LAN) architecture that integrates depthwise separable convolution (DSC) layers with a squeeze-and-excitation (SE) channel attention module to enable real-time anomaly detection in multivariate satellite telemetry streams under a quantized model footprint of 113 KB. Experiments on the NASA SMAP and MSL spacecraft telemetry benchmarks demonstrate that LAN achieves an F1 score of 0.891, outperforming LSTM-based baselines by 6.3 percentage points while reducing inference memory consumption by 78%. Deployment simulation on a Cortex-M7 processor confirms real-time feasibility at 14.2 ms per inference window. These results establish LAN as a practical solution for next-generation autonomous satellite health monitoring under strict resource constraints.
Zitieren
APA
Fang, Y. (2026). Lightweight Attention Networks for Onboard Satellite Anomaly Detection Under Memory Constraints. Frontiers in Agriculture, 3(1), 15-26. https://doi.org/10.71465/fa769
Vancouver
Fang Y. Lightweight Attention Networks for Onboard Satellite Anomaly Detection Under Memory Constraints. Frontiers in Agriculture. 2026;3(1):15-26. doi:10.71465/fa769
BibTeX
@article{Fang2026,
author = {Fang, Yuchen},
title = {Lightweight Attention Networks for Onboard Satellite Anomaly Detection Under Memory Constraints},
journal = {Frontiers in Agriculture},
year = {2026},
volume = {3},
number = {1},
pages = {15--26},
doi = {10.71465/fa769}
}
EUVIMED