GDP-UAV: A Gaussian Differential Privacy-Empowered Multi-UAV Task Offloading in Low-Altitude Economy Networks
IEEE Transactions on Cognitive Communications and Networking · 2026 · 12 · 5001-5013
DOI: 10.1109/tccn.2025.3638768 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Cognitive Communications and Networking, 2026. Verfasst von Miao Du und 5 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 64 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Jiangsu Province, NSF.
Abstract (Originalsprache)
The rapid growth of the Low-Altitude Economy (LAE) drives widespread deployment of multi-unmanned aerial vehicle (UAV) systems for latency-sensitive applications. Collaborative task offloading to edge servers addresses computational limitations but exposes sensitive operational patterns to adversaries. Existing offloading strategies primarily optimize performance metrics such as latency and energy consumption while overlooking critical privacy concerns. To address this gap, we propose GDP-UAV, a novel privacy-preserving offloading framework based on differential privacy. Unlike traditional approaches, GDP-UAV mechanism injects dynamically calibrated Gaussian noise into both action selection and policy updates during offloading, effectively obfuscating decision patterns while maintaining learning stability. Specifically, we first formulate a privacy-constrained optimization problem to minimize noise-perturbed computational costs. Moreover, we establish formal differential privacy guarantees through bounded strategy sensitivity with detailed proven convergence and privacy under persistent noise. Extensive simulations validate the framework, demonstrating 11.6% lower privacy leakage and under 13.2% task completion overhead compared to non-private baselines across LAE scenarios.
Förderung
- National Natural Science Foundation of China
- Natural Science Foundation of Jiangsu Province
- NSF
- Amazon and Japan Science and Technology Agency (JST) Adopting Sustainable Partnerships for Innovative Research Ecosystem
Zitieren
APA
Du, M., Liu, Y., Zhao, C., Tian, W., Shi, H., & Han, Z. (2026). GDP-UAV: A Gaussian Differential Privacy-Empowered Multi-UAV Task Offloading in Low-Altitude Economy Networks. IEEE Transactions on Cognitive Communications and Networking, 12, 5001-5013. https://doi.org/10.1109/tccn.2025.3638768
Vancouver
Du M, Liu Y, Zhao C, Tian W, Shi H, Han Z. GDP-UAV: A Gaussian Differential Privacy-Empowered Multi-UAV Task Offloading in Low-Altitude Economy Networks. IEEE Transactions on Cognitive Communications and Networking. 2026;12:5001-5013. doi:10.1109/tccn.2025.3638768
BibTeX
@article{Du2026,
author = {Du, Miao and Liu, Yinqiu and Zhao, Changyuan and Tian, Wen and Shi, Huaifeng and Han, Zhu},
title = {GDP-UAV: A Gaussian Differential Privacy-Empowered Multi-UAV Task Offloading in Low-Altitude Economy Networks},
journal = {IEEE Transactions on Cognitive Communications and Networking},
year = {2026},
volume = {12},
pages = {5001--5013},
doi = {10.1109/tccn.2025.3638768}
}
EUVIMED