无人机联邦学习场景下动态选择性同态加密隐私保护研究

仇建斌; 章祖葳; 郑宇辉

doi:10.11959/j.issn.1000-0801.2025224

您当前的位置：

首页 >

文章列表页 >

无人机联邦学习场景下动态选择性同态加密隐私保护研究

研究与开发 | 更新时间：2026-01-08

- 无人机联邦学习场景下动态选择性同态加密隐私保护研究
- Research on privacy protection of dynamic selective homomorphic encryption in the context of UAV federated learning
- 电信科学 2025年41卷第12期页码：116-127
- 作者机构：
  
  1.南京工业大学信息管理中心，江苏南京 211816
  2.南京工业大学计算机与信息工程学院（人工智能学院），江苏南京 211816
- 作者简介：
  
  [ "仇建斌（1988- ），女，南京工业大学信息管理中心工程师，主要研究方向为数据安全、高校信息化。" ]
  [ "章祖葳（2002- ），男，南京工业大学计算机与信息工程学院（人工智能学院）硕士生，主要研究方向为网络调度、密码学。" ]
  [ "郑宇辉（2002- ），男，南京工业大学计算机与信息工程学院（人工智能学院）在读，主要研究方向为计算机网络、网络安全。" ]
- 基金信息：
- DOI：10.11959/j.issn.1000-0801.2025224
  中图分类号： TP18;TN918
- 收稿：2025-06-16，
  
  修回：2025-07-03，
  
  录用：2025-07-10，
  
  纸质出版：2025-12-20
- 稿件说明：
移动端阅览
仇建斌,章祖葳,郑宇辉.无人机联邦学习场景下动态选择性同态加密隐私保护研究[J].电信科学,2025,41(12):116-127.

QIU Jianbin,ZHANG Zuwei,ZHENG Yuhui.Research on privacy protection of dynamic selective homomorphic encryption in the context of UAV federated learning[J].Telecommunications Science,2025,41(12):116-127.
仇建斌,章祖葳,郑宇辉.无人机联邦学习场景下动态选择性同态加密隐私保护研究[J].电信科学,2025,41(12):116-127. DOI： 10.11959/j.issn.1000-0801.2025224.

QIU Jianbin,ZHANG Zuwei,ZHENG Yuhui.Research on privacy protection of dynamic selective homomorphic encryption in the context of UAV federated learning[J].Telecommunications Science,2025,41(12):116-127. DOI： 10.11959/j.issn.1000-0801.2025224.

摘要

随着低空经济的快速发展，无人机（unmanned aerial vehicle，UAV）在环境监测、应急救援和物流配送等领域得到广泛应用，并在数据采集与处理过程中面临日益突出的隐私保护需求。为了应对这一挑战，联邦学习结合同态加密被引入以提升数据安全性，但在计算与通信资源受限的无人机场景下，其高昂的资源开销成为实际应用的主要瓶颈。为此，提出了一种面向无人机场景的选择性同态加密隐私保护方案。在每轮本地训练后，客户端基于梯度敏感度评估参数重要性，并结合通信与能耗预算，通过启发式贪心算法动态选择“隐私收益”最大的参数子集进行加密。该方案在联邦学习框架下实现，并采用CKKS同态加密库进行模拟实验。基于CIFAR-10数据集和SimpleCNN模型，对比了5种方案：无加密、全加密、MaskCrypt固定比例加密、DP-AvgFed方案及提出的动态预算方案。实验结果表明，所提方法在实现与MaskCrypt相当安全性的同时，资源开销降低约10%，在保障隐私的同时有效控制了资源消耗，验证了其在资源受限的无人机场景中的可行性和优越性。

Abstract

With the rapid development of the low-altitude economy

unmanned aerial vehicle (UAV) has been widely used in areas such as environmental monitoring

emergency rescue

and logistics delivery. As UAV increasingly engage in data collection and processing

the demand for privacy protection in such scenarios has become increasingly prominent. To address this issue

federated learning combined with homomorphic encryption has been adopted to enhance data security. However

under the constraints of limited computational and communication resources on UAV

the high resource overhead of such approaches becomes a major bottleneck for practical deployment. To this end

a selective homomorphic encryption scheme for privacy protection tailored to UAV scenarios was proposed. After each round of local training

the client evaluates the importance of model parameters based on gradient sensitivity was evaluated by client

and within the given communication and energy budget

dynamically a subset of parameters was selected with the highest “privacy gain” for encryption via a heuristic greedy algorithm. The scheme was implemented within a federated learning framework and simulated using the CKKS homomorphic encryption library. Experiments were conducted on the CIFAR-10 dataset using the SimpleCNN model

comparing five schemes: no encryption

full encryption

fixed-ratio MaskCrypt

DP-AvgFed scheme and the proposed dynamic budget scheme. Results show that the proposed method achieves a balanced trade-off between resource consumption and security. Compared to MaskCrypt

it achieves comparable privacy protection with approximately 10% lower resource overhead

demonstrating its feasibility and effectiveness in resource-constrained UAV scenarios.

关键词

Keywords

references

BASHIR A K , VICTOR N , BHATTACHARYA S , et al . Federated learning for the healthcare metaverse: concepts, applications, challenges, and future directions [J ] . IEEE Internet of Things Journal , 2023 , 10 ( 24 ): 21873 - 21891 .

ZHANG S Y , LI J , SHI L , et al . Federated learning in intelligent transportation systems: recent applications and open problems [J ] . IEEE Transactions on Intelligent Transportation Systems , 2024 , 25 ( 5 ): 3259 - 3285 .

ZHAO Y , ZHAO J , JIANG L S , et al . Privacy-preserving blockchain-based federated learning for IoT devices [J ] . IEEE Internet of Things Journal , 2021 , 8 ( 3 ): 1817 - 1829 .

KURUNATHAN H , HUANG H L , LI K , et al . Machine learning-aided operations and communications of unmanned aerial vehicles: a contemporary survey [J ] . IEEE Communications Surveys & Tutorials , 2024 , 26 ( 1 ): 496 - 533 .

汤凌韬 , 陈左宁 , 张鲁飞 , 等 . 联邦学习中的隐私问题研究进展 [J ] . 软件学报 , 2023 , 34 ( 1 ): 197 - 229 .

TANG L T , CHEN Z N , ZHANG L F , et al . Research progress of privacy issues in federated learning [J ] . Journal of Software , 2023 , 34 ( 1 ): 197 - 229 .

HITAJ B , ATENIESE G , PEREZ-CRUZ F . Deep models under the GAN: information leakage from collaborative deep learning [C ] // Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security . New York : ACM Press , 2017 : 603 - 618 .

PHONG L T , AONO Y , HAYASHI T , et al . Privacy-preserving deep learning via additively homomorphic encryption [J ] . IEEE Transactions on Information Forensics and Security , 2018 , 13 ( 5 ): 1333 - 1345 .

JIANG Z F , WANG W , LIU Y . FLASHE: additively symmetric homomorphic encryption for cross-silo federated learning [EB ] . 2021 .

李晓东 , 李慧 , 赵炽野 , 等 . 基于模分量同态加密的隐私数据联邦学习研究 [J ] . 信息安全研究 , 2025 , 11 ( 3 ): 198 - 204 .

LI X D , LI H , ZHAO C Y , et al . Privacy-preserving federated learning research based on confused modulo projection homomorphic encryption [J ] . Journal of Information Security Research , 2025 , 11 ( 3 ): 198 - 204 .

LIU X Y , LI H W , XU G W , et al . Privacy-enhanced federated learning against poisoning adversaries [J ] . IEEE Transactions on Information Forensics and Security , 2021 , 16 : 4574 - 4588 .

ZHANG X L , FU A M , WANG H Q , et al . A privacy-preserving and verifiable federated learning scheme [C ] // Proceedings of the 2020 IEEE International Conference on Communications (ICC) . Piscataway : IEEE Press , 2020 : 1 - 6 .

余晟兴 , 陈钟 . 基于同态加密的高效安全联邦学习聚合框架 [J ] . 通信学报 , 2023 , 44 ( 1 ): 14 - 28 .

YU S X , CHEN Z . Efficient secure federated learning aggregation framework based on homomorphic encryption [J ] . Journal on Communications , 2023 , 44 ( 1 ): 14 - 28 .

MA J , NAAS S A , SIGG S , et al . Privacy-preserving federated learning based on multi-key homomorphic encryption [J ] . International Journal of Intelligent Systems , 2022 , 37 ( 9 ): 5880 - 5901 .

XU G W , LI H W , LIU S , et al . VerifyNet: secure and verifiable federated learning [J ] . IEEE Transactions on Information Forensics and Security , 2019 , 15 : 911 - 926 .

郭显 , 王典冬 , 冯涛 , 等 . 基于同态加密的可验证隐私保护联邦学习方案 [J ] . 电子与信息学报 , 2025 , 47 ( 4 ): 1113 - 1125 .

GUO X , WANG D D , FENG T , et al . A verifiable privacy protection federated learning scheme based on homomorphic encryption [J ] . Journal of Electronics & Information Technology , 2025 , 47 ( 4 ): 1113 - 1125 .

HIJAZI N M , ALOQAILY M , GUIZANI M , et al . Secure federated learning with fully homomorphic encryption for IoT communications [J ] . IEEE Internet of Things Journal , 2024 , 11 ( 3 ): 4289 - 4300 .

卢为党 , 冯凯 , 丁雨 , 等 . 基于无人机辅助联邦边缘学习通信系统的安全隐私能效研究 [J ] . 电子与信息学报 , 2025 , 47 ( 5 ): 1322 - 1331 .

LU W D , FENG K , DING Y , et al . Research on security, privacy, and energy efficiency in unmanned aerial vehicle-assisted federal edge learning communication systems [J ] . Journal of Electronics & Information Technology , 2025 , 47 ( 5 ): 1322 - 1331 .

卢彦丰 , 吴韬 , 刘春生 , 等 . 无人机辅助的高能效边缘联邦学习综述 [J ] . 计算机科学 , 2024 , 51 ( 4 ): 270 - 279 .

LU Y F , WU T , LIU C S , et al . Survey of UAV-assisted energy-efficient edge federated learning [J ] . Computer Science , 2024 , 51 ( 4 ): 270 - 279 .

HU C H , LI B C . MaskCrypt: federated learning with selective homomorphic encryption [J ] . IEEE Transactions on Dependable and Secure Computing , 2025 , 22 ( 1 ): 221 - 233 .

MCMAHAN H B , RAMAGE D , TALWAR K , et al . Learning differentially private recurrent language models [C ] // Proceedings of the 6th International Conference on Learning Representations (ICLR) . Vancouver : ICLR , 2018 .

浏览量

348

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

Least core：联邦学习中高效稳定的参与者贡献评价机制

异构数据分布下的跨域联邦入侵检测模型

面向低空经济应用场景的无人机路径规划技术研究综述

SWTA-Shapley：一种高效的联邦学习贡献评估方法

基于模型分割的联邦学习数据隐私保护方法