一种融合联邦学习的混合神经网络入侵检测模型

A Hybrid Neural Network Intrusion Detection Model Integrating Federated Learning

摘要

针对网络入侵检测在分布式环境下隐私保护与识别精度难以兼顾的问题,提出一种融合联邦学习架构的混合神经网络模型.该模型通过整合残差结构、门控循环单元和通道注意力机制,解决了流量时序特征挖掘不深和权重感知不足的问题;同时利用合成少数类过采样技术和标签平滑交叉熵损失算法,有效克服了由于节点数据分布异构而导致的模型性能退化难题.在数据集CICIDS2017上的实验结果表明,该模型在准确率、精确率、召回率等核心指标上均达约99.80%,性能显著优于现有的联邦学习及经典检测算法.该结果证明了在不共享原始敏感数据的前提下实现超高精度协同防御的可行性,不仅有效缓解了网络安全领域的数据孤岛难题,也为构建隐私保护型智能化防御体系提供了可靠的技术依据.

Abstract

Addressing the challenge that privacy protection and recognition accuracy are difficult to balance in distributed environments for network intrusion detection, this paper proposes a hybrid neural network model integrating a federated learning architecture. By integrating residual structures, gated recurrent units, and the squeeze-and-excitation channel attention mechanism, the model addresses the problems of insufficient temporal feature mining and lack of weight perception in network traffic. Meanwhile, the synthetic minority over-sampling technique and the label smoothing cross-entropy loss algorithm are utilized to effectively overcome the problem of model performance degradation caused by heterogeneous data distribution across nodes. Experimental results on the CICIDS2017 dataset demonstrate that the model achieves 99.80% in key metrics such as accuracy, precision, and recall, significantly outperforming existing federated learning and classical detection algorithms. These results validate the feasibility of achieving ultra-high precision collaborative defense without sharing raw sensitive data, which not only effectively alleviates the data island dilemma in the field of network security but also provides a reliable technical basis for constructing privacy-preserving intelligent defense systems.

关键词

入侵检测 / 联邦学习 / 通道注意力机制 / 门控循环单元 / 残差网络

Key words

intrusion detection / federated learning / channel attention mechanism / gated recurrent unit / residual network

参考文献

[1] Kasongo S M. A Deep Learning Technique for Intrusion Detection System Using a Recurrent Neural Networks Based Framework[J]. Computer Communications, 2023, 199: 113-125.

[2] Sharma B, Sharma L, Lal C, et al. Explainable Artificial Intelligence for Intrusion Detection in IoT Networks: A Deep Learning Based Approach[J]. Expert Systems with Applications, 2024, 238: 121751-1-121751-15.

[3] Latif S, Boulila W, Koubaa A, et al. DTL-IDS: An Optimized Intrusion Detection Framework Using Deep Transfer Learning and Genetic Algorithm[J]. Journal of Network and Computer Applications, 2024, 221: 103784-1-103784-9.

[4] Sun Y, Ochiai H, Esaki H. Intrusion Detection with Segmented Federated Learning for Large-Scale Multiple Lans[C]// 2020 International Joint Conference on Neural Networks. Piscataway, NJ: IEEE, 2020: 1-8.

[5] Dhakal R, Raza W, Tummala V, et al. Enhancing Intrusion Detection in IoT Networks through Federated Learning[J]. IEEE Access, 2024, 12: 167168-167182.

[6] 罗文华, 张晓龙. 基于联邦学习与卷积神经网络的入侵检测模型[J]. 信息安全研究, 2024, 10(7):642-648.

[7] (Luo W H, Zhang X L. Intrusion Detection Model Based on Federated Learning and Convolutional Neural Networks[J]. Journal of Information Security Research, 2024, 10(7): 642-648.)

[8] Chaurasia N, Ram M, Verma P, et al. A Federated Learning Approach to Network Intrusion Detection Using Residual Networks in Industrial IoT Networks[J]. The Journal of Supercomputing, 2024, 80(13): 18325-18346.

[9] Chen Z, Lv N, Liu P, et al. Intrusion Detection for Wireless Edge Networks Based on Federated Learning[J]. IEEE Access, 2020, 8: 217463-217472.

[10] Shen J, Yang W, Chu Z, et al. Effective Intrusion Detection in Heterogeneous Internet-of-Things Networks via Ensemble Knowledge Distillation-Based Federated Learning[C]// ICC 2024-IEEE International Conference on Communications. Piscataway, NJ: IEEE, 2024: 2034-2039.

[11] Mahadik S S, Pawar P M, Muthalagu R. Edge-Federated Learning-Based Intelligent Intrusion Detection System for Heterogeneous Internet of Things[J]. IEEE Access, 2024, 12: 81736-81757.

[12] 王军, 王华琳, 黄博文, 等. 基于联邦学习和自注意力的工业物联网入侵检测[J]. 吉林大学学报(工学版), 2023, 53(11):3229-3237.

[13] (Wang J, Wang H L, Huang B W, et al. Industrial Internet of Things Intrusion Detection Based on Federated Learning and Self-attention Mechanism[J]. Journal of Jilin University (Engineering and Technology Edition), 2023, 53(11): 3229-3237.)

[14] 康海燕, 张聪明. 基于联邦学习的自适应网络攻击分析方法研究[J]. 信息安全研究, 2024, 10(12):1091-1099.

[15] (Kang H Y, Zhang C M. Research on Adaptive Network Attack Analysis Method Based on Federated Learning[J]. Journal of Information Security Research, 2024, 10(12): 1091-1099.)

[16] Zhang J, Luo C, Carpenter M, et al. Federated Learning for Distributed IoT Intrusion Detection Using Transfer Approaches[J]. IEEE Transactions on Industrial Informatics, 2022, 19(7): 8159-8169.

[17] Javeed D, Saeed M S, Adil M, et al. A Federated Learning-Based Zero Trust Intrusion Detection System for Internet of Things[J]. Ad Hoc Networks, 2024, 162: 103540-1-103540-12.

[18] 周璨, 杨栋, 魏松杰. 融合GRU和CNN的轻量级网络入侵检测模型[J]. 计算机系统应用, 2023, 32(8):162-170.

[19] (Zhou C, Yang D, Wei S J. Lightweight Network Intrusion Detection Model Based on the Fusion of GRU and CNN[J]. Computer System Applications, 2023, 32(8): 162-170.)

[20] Alhasawi Y, Alghamdi S. Federated Learning for Decentralized DDoS Attack Detection in IoT Networks[J]. IEEE Access, 2024, 12: 42357-42368.

[21] Lazzarini R, Tianfield H, Charissis V. Federated Learning for IoT Intrusion Detection[J]. AI, 2023, 4(3): 509-530.

基金资助

四川省科技计划项目(2025NSFJQ0053)

先进微处理器国家工程实验室(共建)开放课题资助项目(CUIT-SICSC-2025-06)