结合自适应数据增强和集成学习的机载网络入侵检测研究

周茂辉; 刘文琪; 李艳军; 宫艺姝

doi:10.11959/j.issn.1000-0801.2026050

您当前的位置：

首页 >

文章列表页 >

结合自适应数据增强和集成学习的机载网络入侵检测研究

更新时间：2026-01-21

- 结合自适应数据增强和集成学习的机载网络入侵检测研究
- Research on airborne network intrusion detection combining adaptive data augmentation and ensemble learning
- 电信科学 2026年页码：1-11
- 作者机构：
  
  1.南京航空航天大学民航学院，江苏南京 210016
  2.新疆通达低空科技有限公司，新疆石河子市邮编 832000
- 作者简介：
  
  [ "周茂辉（1996-），男，南京航空航天大学交通运输专业博士研究生，主要研究方向是航空器机电系统测试性设计与验证、航空器故障智能诊断、航空安全等。" ]
  [ "刘文琪（1997-），女，南京航空航天大学电子信息专业博士研究生，主要研究方向是航空器机载网络入侵检测，航空安全等。" ]
  [ "李艳军（1969-），男，南京航空航天大学教授，博导，主要研究方向为航空器故障诊断与健康监测，适航审定及验证技术，飞机改装设计及适航审定，航空维修工程及管理。" ]
  [ "宮艺姝（1993-），女，疆通达低空科技有限公司科创副总经理，主要研究方向为低空起降安全、低空气象探测与预警。" ]
- 基金信息：
  
  江苏省研究生科研与实践创新计划项目(KYCX25_0620);新疆生产建设兵团科技计划资助(2025AB070.2025AB068)
- DOI：10.11959/j.issn.1000-0801.2026050
  中图分类号： TN915.08;TP309
- 收稿：2025-07-23，
  
  修回：2025-11-25，
  
  录用：2025-11-25，
  
  网络出版：2026-01-06，
- 稿件说明：
移动端阅览
周茂辉,刘文琪,李艳军等.结合自适应数据增强和集成学习的机载网络入侵检测研究[J].电信科学,

ZHOU Maohui,LIU Wenqi,LI Yanjun,et al.Research on airborne network intrusion detection combining adaptive data augmentation and ensemble learning[J].Telecommunications Science,
周茂辉,刘文琪,李艳军等.结合自适应数据增强和集成学习的机载网络入侵检测研究[J].电信科学, DOI：10.11959/j.issn.1000−0801.2026050.

ZHOU Maohui,LIU Wenqi,LI Yanjun,et al.Research on airborne network intrusion detection combining adaptive data augmentation and ensemble learning[J].Telecommunications Science, DOI：10.11959/j.issn.1000−0801.2026050.

摘要

机载网络入侵检测可能面临异常样本稀缺和数据分布不平衡的双重挑战，传统方法难以同时保证检测精度和泛化能力。为此，结合多视图对比稀疏自编码器（multi-view contrastive sparse autoencoder，MCSAE）的数据增强方法，提出一种改进分层抽样集成学习的联合优化方法。首先，针对异常样本缺失问题，设计MCSAE，通过多视图数据增强和对比学习策略，在稀疏自编码器框架下学习更具判别性的潜在表示，并利用重输入对比机制优化异常样本生成质量，有效缓解数据稀疏性带来的模型偏差。其次，针对类别不平衡问题，提出改进分层抽样策略，在传统分层抽样基础上引入全局特征保留机制，避免局部采样导致多数类分布失真，确保分类器能够学习数据的完整统计特性。最后，结合

1分数自适应加权集成学习，融合随机森林、长短期记忆（long short-term memory，LSTM）网络等多样化基分类器，动态调整模型权重，进一步提升对少数类攻击的检测能力。实验结果表明，

相较于现有方法，所提方法在机载网络数据集上的召回率提升5.2%，

1提升3.7%，为复杂网络环境下的入侵检测提供了可靠性解决方案。

Abstract

Airborne network intrusion detection may face the dual challenges of scarce abnormal samples and unbalanced data distribution. Traditional methods are difficult to simultaneously guarantee detection accuracy and generalization ability. Therefore

combined with the data augmentation method of multi-view contrastive sparse autoencoder (MCSAE)

a joint optimization method with improved hierarchical sampling ensemble learning was proposed. Firstly

for the problem of missing abnormal samples

a MCSAE was designed. Through multi-view data augmentation and contrastive learning strategies

a more discriminative latent representation was learned under the framework of sparse autoencoder

and the quality of abnormal sample generation was optimized by the re-input contrast mechanism

effectively alleviating the model deviation caused by data sparsity. Secondly

for the problem of class imbalance

an improved hierarchical sampling strategy was proposed. On the basis of traditional hierarchical sampling

a global feature retention mechanism was introduced to avoid the distortion of the majority class distribution caused by local sampling

ensuring that the classifier can learn the complete statistical characteristics of the data. Finally

combined with

1 score adaptive weighted ensemble learning

diverse base classifiers such as random forest and LSTM were integrated to dynamically adjust the model weights

further improving the detection ability for minority class attacks. The experimental results showed that

compared with the existing methods on the airborne network dataset

the proposed method had a 5.2% increase in recall rate and a 3.7% increase in

1 score. This provides a reliable solution for intrusion detection in complex network environments.

关键词

Keywords

references

DE SANTO D , MALAVENDA C S , ROMANO S P , et al . Exploiting the MIL-STD-1553 avionic data bus with an active cyber device [J ] . Computers & Security , 2021 , 100 : 102097 .

HAYAEIAN SHIRVAN M , MOATTAR M H , HOSSEINZADEH M . Deep generative approaches for oversampling in imbalanced data classification problems: a comprehensive review and comparative analysis [J ] . Applied Soft Computing , 2025 , 170 : 112677 .

MUJAHID M , KıNA E , RUSTAM F , et al . Data oversampling and imbalanced datasets: an investigation of performance for machine learning and feature engineering [J ] . Journal of Big Data , 2024 , 11 ( 1 ): 87 .

YANG Y Q , ZHENG K F , WU C H , et al . Improving the classification effectiveness of intrusion detection by using improved conditional variational autoencoder and deep neural network [J ] . Sensors , 2019 , 19 ( 11 ): 2528 .

JO W , KIM D . OBGAN: Minority oversampling near borderline with generative adversarial networks [J ] . Expert Systems with Applications , 2022 , 197 : 116694 .

DAYAN O , WOLF L , WANG F , et al . Optimizing AI for mobile malware detection by self-built-dataset GAN oversampling and LGBM [C ] // Proceedings of the 2023 IEEE International Conference on Cyber Security and Resilience (CSR) . Piscataway : IEEE Press , 2023 : 60 - 65 .

ZHANG Y L , LIU Y C , WANG Y , et al . An ensemble oversampling method for imbalanced classification with prior knowledge via generative adversarial network [J ] . Chemometrics and Intelligent Laboratory Systems , 2023 , 235 : 104775 .

KIM J , KIM J , KIM H , et al . CNN-based network intrusion detection against denial-of-service attacks [J ] . Electronics , 2020 , 9 ( 6 ): 916 .

IMRANA Y , XIANG Y P , ALI L , et al . A bidirectional LSTM deep learning approach for intrusion detection [J ] . Expert Systems with Applications , 2021 , 185 : 115524 .

WALI S , ALI FARRUKH Y , KHAN I . Explainable AI and random forest based reliable intrusion detection system [J ] . Computers & Security , 2025 , 157 : 104542 .

LI H , GE H J , SANG Y Q , et al . An optimized multi-layer ensemble model for airborne networks intrusion detection [J ] . Applied Soft Computing , 2024 , 167 : 112282 .

LE T T H , SHIN Y , KIM M , et al . Towards unbalanced multiclass intrusion detection with hybrid sampling methods and ensemble classification [J ] . Applied Soft Computing , 2024 , 157 : 111517 .

ELSAYED M A , WRANA M , MANSOUR Z , et al . AdaptIDS: adaptive intrusion detection for mission-critical aerospace vehicles [J ] . IEEE Transactions on Intelligent Transportation Systems , 2022 , 23 ( 12 ): 23459 - 23473 .

LI H , SANG Y Q , GE H J , et al . Anomaly detection of aviation data bus based on SAE and IMD [J ] . Computers & Security , 2024 , 137 : 103619 .

张重生 , 陈杰 , 李岐龙 , 等 . 深度对比学习综述 [J ] . 自动化学报 , 2023 , 49 ( 1 ): 15 - 39 .

ZHANG C S , CHEN J , LI Q L , et al . Deep contrastive learning: a survey [J ] . Acta Automatica Sinica , 2023 , 49 ( 1 ): 15 - 39 .

AHMED H , TRAORE I , QUINAN P , et al . A collection of datasets for intrusion detection in MIL-STD-1553 platforms [M ] // Artificial Intelligence for Cyber-Physical Systems Hardening . Cham : Springer International Publishing , 2023 : : 81 - 100 .

GÉNÉREUX S J J , LAI A K H , FOWLES C O , et al . MAIDENS: MIL-STD-1553 anomaly-based intrusion detection system using time-based histogram comparison [J ] . IEEE Transactions on Aerospace and Electronic Systems , 2020 , 56 ( 1 ): 276 - 284 .

刘文琪 , 胡涛 , 闫洁 , 等 . 基于DeepInsight和迁移学习的入侵检测技术 [J ] . 工程科学学报 , 2024 , 46 ( 12 ): 2238 - 2245 .

LIU W Q , HU T , YAN J , et al . Network intrusion detection technology based on DeepInsight and transfer learning [J ] . Chinese Journal of Engineering , 2024 , 46 ( 12 ): 2238 - 2245 .

ZHOU M H , LI Y J , CAO Y Y , et al . Physics-informed spatio-temporal hybrid neural networks for predicting remaining useful life in aircraft engine [J ] . Reliability Engineering & System Safety , 2025 , 256 : 110685 .

HE D J , LIU X X , ZHENG J J , et al . A lightweight and intelligent intrusion detection system for integrated electronic systems [J ] . IEEE Network , 2020 , 34 ( 4 ): 173 - 179 .

LIU W Q , LI S J , GAO C , et al . An adaptive graph neural network-based intrusion detection system for airborne network [J ] . Engineering Applications of Artificial Intelligence , 2025 , 152 : 110851 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于Powershap和混合采样的动态集成入侵检测模型

基于深度学习和SVM-RFE的网络入侵检测模型

面向车联网的基于卷积神经网络的入侵检测模型

演化森林哈希：一种无监督的在线哈希学习算法