基于平衡生成对抗网络的海洋气象传感网入侵检测研究

doi:10.11959/j.issn.1000-436x.2023078

通信学报 ›› 2023, Vol. 44 ›› Issue (4): 124-136.doi: 10.11959/j.issn.1000-436x.2023078

基于平衡生成对抗网络的海洋气象传感网入侵检测研究

苏新¹, 张桂福¹, 行鸿彦², Zenghui Wang³

¹ 河海大学物联网工程学院，江苏常州 231022
² 南京信息工程大学电子与信息工程学院，江苏南京 210044
³ 南非大学电气工程系，约翰内斯堡 1710

修回日期:2023-02-20 出版日期:2023-04-25 发布日期:2023-04-01
作者简介:苏新（1986- ），男，河北霸州人，博士，河海大学教授，主要研究方向为移动通信、边缘/雾计算、智慧海洋等
张桂福（1999- ），男，江西赣州人，河海大学硕士生，主要研究方向为入侵检测、边缘/雾计算、智慧海洋等
行鸿彦（1962- ），男，山西新绛人，博士，南京信息工程大学教授，主要研究方向为气象仪器设计与计量、信号检测与处理等
Zenghui Wang（1979- ），男，博士，南非大学教授，主要研究方向为人工智能及其应用、自动控制等
基金资助:
国家重点研发计划基金资助项目(2021YFE0105500)

Research on intrusion detection for maritime meteorological sensor network based on balancing generative adversarial network

Xin SUN¹, Guifu ZHANG¹, Hongyan XING², Wang Zenghui³

¹ The College of IoT Engineering, Hohai University, Changzhou 213022, China
² School of Electronics ＆Information Engineering, Nanjing University of Information Science ＆Technology, Nanjing 210044, China
³ Department of Electrical Engineering, University of South Africa, Johannesburg 1710, South Africa

Revised:2023-02-20 Online:2023-04-25 Published:2023-04-01
Supported by:
The National Key Research and Development Program of China(2021YFE0105500)

摘要/Abstract

摘要：

针对海洋气象传感网（MMSN）环境下海洋移动终端资源受限和网络流量不平衡导致网络入侵难以被准确检测的问题，提出了一种基于移动边缘计算的MMSN物理架构和一种基于平衡生成对抗网络的入侵检测模型。首先，利用改进的平衡生成对抗网络对不平衡数据进行数据增强。其次，利用基于分组卷积的轻量级网络对入侵数据进行分类。最后，通过计算机仿真证明了所提模型较传统数据增强模型具有更高识别各类攻击的能力，尤其是针对MMSN的少数类样本攻击。

关键词: 海洋气象传感网, 入侵检测, 移动边缘计算, 平衡生成对抗网络, 分组卷积

Abstract:

Aiming at the problem that the resources of maritime mobile terminals were limited and the network traffic was imbalanced in the MMSN (maritime meteorological sensor network) environment, which made it difficult to detect network intrusion accurately, a mobile edge computing based physical architecture of MMSN was proposed, and an intrusion detection model based on balancing generative adversarial network was proposed.First, an advanced balancing generative adversarial network was adopted to augment the imbalanced data.Then, a lightweight network based on group convolution was applied to intrusion data classification.Finally, compared with conventional data augmentation models, the computer simulation proves that the proposed model has a higher ability to recognize various attacks, especially minority class attacks on MMSN.

Key words: maritime meteorological sensor network, intrusion detection, mobile edge computing, balancing generative adversarial network, group convolution

中图分类号:

TN929.52

苏新, 张桂福, 行鸿彦, Zenghui Wang. 基于平衡生成对抗网络的海洋气象传感网入侵检测研究[J]. 通信学报, 2023, 44(4): 124-136.

Xin SUN, Guifu ZHANG, Hongyan XING, Wang Zenghui. Research on intrusion detection for maritime meteorological sensor network based on balancing generative adversarial network[J]. Journal on Communications, 2023, 44(4): 124-136.

图/表 18

图1

图2

图3

图4

表1

表2

表3

表4

图5

图6

表5

表6

图7

图8

图9

表7

表8

表9

参考文献 38

[22]	VU L , NGUYEN Q U . Handling imbalanced data in intrusion detection systems using generative adversarial networks[J]. Journal on Information Technologies ＆ Communications, 2020,2020(1): 1-13.
[23]	YUAN D N , OTA K , DONG M X ,et al. Intrusion detection for smart home security based on data augmentation with edge computing[C]// Proceedings of 2020 IEEE International Conference on Communications (ICC). Piscataway:IEEE Press, 2020: 1-6.
[24]	ANDRESINI G , APPICE A , ROSE L D ,et al. GAN augmentation to deal with imbalance in imaging-based intrusion detection[J]. Future Generation Computer Systems, 2021,123: 108-127.
[25]	MARIANI G , SCHEIDEGGER F , ISTRATE R ,et al. BAGAN:data augmentation with balancing GAN[J]. arXiv Preprint,arXiv:1803.09655, 2018.
[26]	ZHU J X , MENG L L , WU W X ,et al. Generative adversarial network-based atmospheric scattering model for image dehazing[J]. Digital Communications and Networks, 2021,7(2): 178-186.
[27]	BROPHY E , WANG Z W , SHE Q ,et al. Generative adversarial networks in time series:a survey and taxonomy[J]. arXiv Preprint,arXiv:2107.11098, 2021.
[28]	ZHAO X , HUANG G Q , JIANG J ,et al. Task offloading of cooperative intrusion detection system based on deep Q network in mobile edge computing[J]. Expert Systems with Applications, 2022,206:117860.
[29]	HE Y , CHENG J G . User location privacy protection mechanism for location-based services[J]. Digital Communications and Networks, 2021,7(2): 264-276.
[30]	MINAHIL , AYUB M F , MAHMOOD K ,et al. Lightweight authentication protocol for e-health clouds in IoT-based applications through 5G technology[J]. Digital Communications and Networks, 2021,7(2): 235-244.
[31]	GURUMURTHY S , SARVADEVABHATLA R K , BABU R V . DELIGAN:generative adversarial networks for diverse and limited data[C]// Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway:IEEE Press, 2017: 4941-4949.
[32]	HUANG G F , JAFARI A H . Enhanced balancing GAN:minority-class image generation[J]. Neural Computing and Applications, 2023,35(7): 5145-5154.
[33]	GULRAJANI I , AHMED F , ARJOVSKY M ,et al. Improved training of Wasserstein GANs[C]// Proceedings of the 31st International Conference on Neural Information Processing Systems. Massachusetts:MIT Press, 2017: 5769-5779.
[34]	KODALI N , ABERNETHY J , HAYS J ,et al. On convergence and stability of GANs[J]. arXiv Preprint,arXiv:1705.07215, 2017.
[35]	MAKHZANI A , SHLENS J , JAITLY N ,et al. Adversarial autoencoders[J]. arXiv Preprint,arXiv:1511.05644, 2015.
[36]	KRIZHEVSKY A , SUTSKEVER I , HINTON G E . ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017,60(6): 84-90.
[37]	PANIGRAHI R , BORAH S . A detailed analysis of CICIDS2017 dataset for designing intrusion detection systems[J]. International Journal of Engineering ＆ Technology, 2018,7(3): 479-482.
[1]	XIA T T , WANG M M , ZHANG J J ,et al. Maritime Internet of things:challenges and solutions[J]. IEEE Wireless Communications, 2020,27(2): 188-196.
[2]	YANG X , XING H Y . A data complementary method for thunderstorm point charge localization based on atmospheric electric field apparatus array group[J]. Digital Communications and Networks, 2021,7(2): 170-177.
[3]	LIU R W , NIE J T , GARG S ,et al. Data-driven trajectory quality improvement for promoting intelligent vessel traffic services in 6G-enabled maritime IoT systems[J]. IEEE Internet of Things Journal, 2021,8(7): 5374-5385.
[4]	苏新, 孟蕾蕾, 周一青 ,等. 基于深度强化学习的海洋移动边缘计算卸载方法[J]. 通信学报, 2022,43(10): 133-145.
	SU X , MENG L L , ZHOU Y Q ,et al. Maritime mobile edge computing offloading method based on deep reinforcement learning[J]. Journal on Communications, 2022,43(10): 133-145.
[5]	ASHRAF I , PARK Y , HUR S ,et al. A survey on cyber security threats in IoT-enabled maritime industry[J]. IEEE Transactions on Intelligent Transportation Systems, 2023,24(2): 2677-2690.
[6]	SU X , WANG H , FU X ,et al. Substring-searchable attribute-based encryption and its application for IoT devices[J]. Digital Communications and Networks, 2021,7(2): 277-283.
[7]	ANDERSON J P . Computer security threat monitoring and surveillance[R]. 1980.
[8]	LI W J , WANG Y , JIN Z P ,et al. Challenge-based collaborative intrusion detection in software-defined networking:an evaluation[J]. Digital Communications and Networks, 2021,7(2): 257-263.
[9]	AMBUSAIDI M A , HE X J , NANDA P ,et al. Building an intrusion detection system using a filter-based feature selection algorithm[J]. IEEE Transactions on Computers, 2016,65(10): 2986-2998.
[10]	ALAZZAM H , SHARIEH A , SABRI K E . A feature selection algorithm for intrusion detection system based on pigeon inspired optimizer[J]. Expert Systems with Applications, 2020,148:113249.
[11]	DERHAB A , BELAOUED M , MOHIUDDIN I ,et al. Histogram-based intrusion detection and filtering framework for secure and safe in-vehicle networks[J]. IEEE Transactions on Intelligent Transportation Systems, 2021,23(3): 2366-2379.
[12]	VU L , CAO V L , NGUYEN Q U ,et al. Learning latent representation for IoT anomaly detection[J]. IEEE Transactions on Cybernetics, 2022,52(5): 3769-3782.
[13]	ZHAO R J , YIN J , XUE Z ,et al. An efficient intrusion detection method based on dynamic autoencoder[J]. IEEE Wireless Communications Letters, 2021,10(8): 1707-1711.
[14]	CHEN Y , LIN Q Z , WEI W H ,et al. Intrusion detection using multi-objective evolutionary convolutional neural network for Internet of things in fog computing[J]. Knowledge-Based Systems, 2022,244:108505.
[15]	LEI S W , XIA C H , LI Z ,et al. HNN:a novel model to study the intrusion detection based on multi-feature correlation and temporal-spatial analysis[J]. IEEE Transactions on Network Science and Engineering, 2021,8(4): 3257-3274.
[16]	ZHANG X Q , YANG F , HU Y ,et al. RANet:network intrusion detection with group-gating convolutional neural network[J]. Journal of Network and Computer Applications, 2022,198:103266.
[17]	KANNA P R , SANTHI P . Hybrid intrusion detection using MapReduce based black widow optimized convolutional long short-term memory neural networks[J]. Expert Systems with Applications, 2022,194:116545.
[18]	ANDRESINI G , APPICE A , CAFORIO F P ,et al. ROULETTE:a neural attention multi-output model for explainable network intrusion detection[J]. Expert Systems with Applications, 2022,201:117144.
[19]	REHMAN S U , KHALIQ M , IMTIAZ S I ,et al. DIDDOS:an approach for detection and identification of distributed denial of service (DDoS) cyberattacks using gated recurrent units (GRU)[J]. Future Generation Computer Systems, 2021,118: 453-466.
[20]	BAGUI S , LI K Q . Resampling imbalanced data for network intrusion detection datasets[J]. Journal of Big Data, 2021,8(1): 1-41.
[21]	HUANG S K , LEI K . IGAN-IDS:an imbalanced generative adversarial network towards intrusion detection system in ad-hoc networks[J]. Ad Hoc Networks, 2020,105:102177.
[38]	XU X , LI J , YANG Y ,et al. Toward effective intrusion detection using log-cosh conditional variational autoencoder[J]. IEEE Internet of Things Journal, 2021,8(8): 6187-6196.

数据集	Normal/条	DoS/条	Probe/条	U2R/条	R2L/条	总计/条
KDDTrain+	67 343	45 927	11 656	52	995	125 973
KDDTest+	9 711	7 459	2 421	200	2 754	22 544

数据集	Benign/条	DoS/条	Port Scan/条	暴力破解/条	Web Attack/条	Bot/条	Infiltration/条	总计/条
训练集	127 193	30 380	12 704	11 065	1 744	1 565	29	184 680
测试集	31 799	7 595	3 176	2 767	436	391	7	46 171

参数设置	NSL-KDD	CIC-IDS2017
编码器En	122-256-128-64-32	78-85-128-64-32-15
解码器De	37-64-128-256-122	22-32-64-128-85-78
训练周期/轮	30, 300	40, 500
批次大小/条	128, 128	128, 128
学习率	0.000 1, 0.000 2	0.01, 0.000 2

参数	NSL-KDD	CIC-IDS2017
普通卷积层	[1,3]×8, poolsize:2	[1,3]×8, poolsize:2
分组卷积层	[1,2]×3,[1,2]×8	[1,2]×12,[1,2]×18
逐点卷积层	[1,1]×1	[1,1]×1
分类层	160-80-5, Dropout:0.5	128-64-7, Dropout:0.5
训练周期/轮	100	200
批次大小/条	256	1 024
学习率	0.000 5	0.001

模型	DoS		Probe		U2R		R2L
模型	MED	MMD	MED	MMD	MED	MMD	MED	MMD
ADASYN	5.482 5	2.531 2	0.915 6	0.501 9	0.163 8	0.262 5	0.109 0	0.224 2
CWGAN-GP	0.178 2	0.117 3	0.006 1	0.005 1	0.098 2	0.189 9	0.034 1	0.064 4
BAGAN	0.546 9	0.673 6	1.037 0	0.561 9	0.340 7	0.505 0	0.052 3	0.133 4
本文模型	0.003 7	0.009 7	0.022 7	0.014 1	0.062 7	0.256 2	0.009 7	0.019 8

基于平衡生成对抗网络的海洋气象传感网入侵检测研究

Research on intrusion detection for maritime meteorological sensor network based on balancing generative adversarial network

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 18

参考文献 38

相关文章 15

Metrics

推荐阅读 0

模型	DoS		Port Scan		暴力破解		Web Attack		Bot		Infiltration
模型	MED	MMD	MED	MMD	MED	MMD	MED	MMD	MED	MMD	MED	MMD
ADASYN	1.234 6	1.073 2	0.044 2	1.093 8	0.260 6	0.626 0	0.378 0	1.721 7	0.403 3	0.826 6	0.012 9	0.039 2
CWGAN-GP	0.001 8	0.003 7	0.017 4	0.267 4	0.078 7	0.138 5	0.143 3	2.373 5	0.007 9	0.060 0	4.078 1	4.374 4
BAGAN	0.130 5	0.157 3	0.018 8	0.683 8	0.012 9	0.029 4	0.017 0	0.101 3	0.028 1	0.063 7	0.263 9	0.792 7
本文模型	0.148 4	0.134 4	0.001 0	0.035 2	0.001 4	0.004 0	0.025 6	0.140 5	0.016 3	0.029 1	0.511 9	0.759 8

算法	Precision	Recall	F1值
RANet	0.819 2	0.832 3	0.825 7
ROULETTE	0.828 5	0.804 3	0.779 4
IGAN	0.848 5	0.844 5	0.841 7
LCVAE	0.976 1	0.685 3	0.807 9
本文模型	0.876 2	0.864 2	0.870 2

数据集	训练参数数量/个	FLOP/次	模型大小/KB
NSL-KDD	22 940	34 688	92
CIC-IDS2017	14 632	54 576	62

[1]	余雪勇, 邱礼翔, 宋家宁, 朱洪波. 无人机辅助边缘计算中安全通信与能效优化策略[J]. 通信学报, 2023, 44(3): 45-54.
[2]	范伟, 彭诚, 朱大立, 王雨晴. 移动边缘计算网络下基于静态贝叶斯博弈的入侵响应策略研究[J]. 通信学报, 2023, 44(2): 70-81.
[3]	赵庶旭, 韦萍, 王小龙. 多任务并发边缘计算环境中最优联盟结构生成策略[J]. 通信学报, 2023, 44(2): 172-184.
[4]	龙隆, 刘子辰, 陆在旺, 张玉成, 李蕾. 移动边缘网络下服务缓存与资源分配联合优化策略[J]. 通信学报, 2023, 44(1): 64-74.
[5]	王一丰, 郭渊博, 陈庆礼, 方晨, 林韧昊. 基于对比学习的细粒度未知恶意流量分类方法[J]. 通信学报, 2022, 43(10): 12-25.
[6]	李斌, 刘文帅, 谢万城, 费泽松. 智能反射面赋能无人机边缘网络计算卸载方案[J]. 通信学报, 2022, 43(10): 223-233.
[7]	黄永明, 郑冲, 张征明, 尤肖虎. 大规模无线通信网络移动边缘计算和缓存研究[J]. 通信学报, 2021, 42(4): 44-61.
[8]	李方伟, 张海波, 王子心. 车联网中基于MEC的V2X协同缓存和资源分配[J]. 通信学报, 2021, 42(2): 26-36.
[9]	刘奇旭, 王君楠, 尹捷, 陈艳辉, 刘嘉熹. 对抗机器学习在网络入侵检测领域的应用[J]. 通信学报, 2021, 42(11): 1-12.
[10]	王娜,郑坤,付俊松,李剑. 基于分块的移动边缘计算密文检索方法[J]. 通信学报, 2020, 41(7): 95-102.
[11]	田有亮,吴雨龙,李秋贤. 基于信息论的入侵检测最佳响应方案[J]. 通信学报, 2020, 41(7): 121-130.
[12]	亓晋,孙海蓉,巩锟,许斌,张顺颐,孙雁飞. 移动边缘计算中基于信誉值的智能计算卸载模型研究[J]. 通信学报, 2020, 41(7): 141-151.
[13]	吴大鹏,吕吉,李职杜,王汝言. 移动性感知的边缘服务迁移策略[J]. 通信学报, 2020, 41(4): 1-13.
[14]	张海君,张资政,隆克平. 基于移动边缘计算的NOMA异构网络资源分配[J]. 通信学报, 2020, 41(4): 27-33.
[15]	赵临东,庄文芹,陈建新,周亮. 异构蜂窝网络中分层任务卸载：建模与优化[J]. 通信学报, 2020, 41(4): 34-44.

算法	Precision	Recall	F1值
LR	0.896 6	0.906 7	0.901 6
SVM	0.935 1	0.931 3	0.933 2
MLP	0.978 2	0.976 7	0.977 4
CNN	0.976 8	0.975 8	0.976 3
LSTM	0.979 6	0.978 7	0.979 1
本文模型	0.989 8	0.989 4	0.989 6