鼠标行为HHT变换的工业互联网用户身份认证

doi:10.11959/j.issn.2096-3750.2022.00268

物联网学报 ›› 2022, Vol. 6 ›› Issue (2): 77-87.doi: 10.11959/j.issn.2096-3750.2022.00268

鼠标行为HHT变换的工业互联网用户身份认证

张一弓¹^,², 易茜¹, 李剑², 李聪波¹, 尹爱军¹, 易树平¹

¹ 重庆大学机械传动国家重点实验室，重庆 400044
² 重庆大学输配电装备及系统安全与新技术国家重点实验室，重庆 400044

修回日期:2022-04-24 出版日期:2022-06-30 发布日期:2022-06-01
作者简介:张一弓（1996− ），男，重庆大学电气学院博士生，主要研究方向为网络用户行为模式与可信交互、电力物联网
易茜（1986−），女，博士，重庆大学讲师、硕士生导师，主要研究方向为网络用户行为模式与可信交互、智能制造系统、绿色制造等
李剑（1971− ），男，博士，重庆大学教授、博士生导师，国家杰出青年科学基金获得者，主要研究方向为电工绝缘新材料、电力装备智能化、物联网等
李聪波（1981− ），男，博士，重庆大学教授、博士生导师，主要研究方向为绿色制造、智能制造系统、制造系统工程等
尹爱军（1978− ），男，博士，重庆大学教授、博士生导师，主要研究方向为智能测试仪器、工业大数据智能运维系统、设备故障诊断与预测、高端装备等
易树平（1960− ），男，博士，重庆大学教授、博士生导师，主要研究方向为工业工程理论与技术、数字化背景下的人因工程、智能制造等
基金资助:
国家自然科学基金资助项目(71671020);中央高校基本科研业务费资助项目(2021CDJKYJH022)

User authentication of industrial internet based on HHT transform of mouse behavior

Yigong ZHANG¹^,², Qian YI¹, Jian LI², Congbo LI¹, Aijun YIN¹, Shuping YI¹

¹ State Key Laboratory of Mechanical Transmission, Chongqing University, Chongqing 400044, China
² State Key Laboratory of Power Transmission Equipment and System Security and New Technology, Chongqing University, Chongqing 400044, China

Revised:2022-04-24 Online:2022-06-30 Published:2022-06-01
Supported by:
The National Natural Science Foundation of China(71671020);The Fundamental Research Funds for the Central Universities(2021CDJKYJH022)

摘要/Abstract

摘要：

工业互联网的快速发展引发了对网络安全的广泛关注，终端用户身份认证技术成为研究热点。根据工业互联网人机交互特点，设计了实验网站，收集了该网站 24 名用户两年半的非受控环境下鼠标行为数据作实例，采用希尔伯特黄变换（HHT, Hilbert-Huang transform）提取鼠标行为信号频域特征，结合时域特征，形成163维时频域联合特征矩阵，用于表征用户鼠标行为模式特征。使用 Bagged tree、支持向量机（SVM, support vector machine）、Boost tree和K最邻近（KNN, K-nearest neighbor）算法构建网络用户身份认证模型，对比数据测试结果表明，Bagged tree算法在本案例中内部检测效果最佳，平均错误接受率（FAR, false acceptance rate）为0.12%、平均错误拒绝率（FRR, false rejection rate）为0.28%；外部检测中，平均FAR为1.47%。相较于传统鼠标动力学方法，使用HHT提取鼠标行为频域信息能更好地实现终端用户身份认证，为保障工业互联网安全提供有效的技术支撑。

关键词: 工业互联网, 身份认证, 鼠标行为, 希尔伯特黄变换, Baggedtree

Abstract:

The rapid development of the industrial internet had caused widespread concern about the network security, and the end-user authentication technology was considered a research hotspot.According to the characteristics of human-computer interaction in industrial internet, an experimental website was designed.24 users' mouse behavior data in an uncontrolled environment were collected within 2.5 years to conduct case studies.Hilbert-Huang transform (HHT) was used to extract frequency domain features of mouse behavior signals, combined with time domain features to form a time-frequency joint domain feature matrix of 163-dimensional to characterize user mouse behavior patterns.Bagged tree, support vector machine (SVM), Boost tree and K-nearest neighbor (KNN) were used to build a user authentication model， and the comparison result showed that the Bagged tree had the best internal detection effect in this case, with an average false acceptance rate (FAR) of 0.12% and an average false rejection rate (FRR) of 0.28%.In external detection, the FAR was 1.47%.Compared with the traditional mouse dynamics method, the frequency domain information of mouse behavior extracted by HHT can better realize the user authentication, and provide technical support the security of the industrial internet.

Key words: industrial internet, identity authentication, mouse behavior, Hilbert-Huang transform, Bagged tree

中图分类号:

TP273

张一弓, 易茜, 李剑, 李聪波, 尹爱军, 易树平. 鼠标行为HHT变换的工业互联网用户身份认证[J]. 物联网学报, 2022, 6(2): 77-87.

Yigong ZHANG, Qian YI, Jian LI, Congbo LI, Aijun YIN, Shuping YI. User authentication of industrial internet based on HHT transform of mouse behavior[J]. Chinese Journal on Internet of Things, 2022, 6(2): 77-87.

图/表 10

图1

表1

鼠标运动行为时域特征"

时域特征	计算方法
X轴坐标	X
Y轴坐标	Y
X轴速度	$v_{x} = δ x / δ t$
Y轴速度	$v_{y} = δ y / δ t$
切向速度	$v = \sqrt{v_{x}^{2} + v_{y}^{2}}$
切向加速度	$a = δ v / δ t$
速度二阶导数	$j = δ a / δ t$
速度三阶导数	$s = δ j / δ t$
速度四阶导数	$d = δ s / δ t$
角度	$θ = \arctan (δ y / δ x)$
角度变化率	$v_{θ} = δ θ / δ t$
曲率	$c = δ θ / δ s$
曲率变化率	$v_{cur} = δ c / δ t$
曲率距离^[11]	cd

表1

图2

表2

表3

图4

表4

表5

表6

表7

参考文献 37

[1]	中国国务院. 国务院关于深化“互联网+先进制造业”发展工业互联网的指导意见[EB]. 2017.
	The Chinese State Council. Guiding opinions on deepening the "Internet plus advanced manufacturing industry" to develop industrial Internet[EB]. 2017.
[2]	人民政协报. 深入实施工业互联网创新发展战略为建设制造强国和现代化经济体系提供有力支撑:全国政协“加快推进工业互联网建设”双周协商座谈会发言摘登(上)[N]. 人民政协报, 2020-05-07(4).
	People's Political Consultative Conference. In-depth implementation of the Industrial Internet innovation development strategy provides strong support for the construction of a strong manufacturing country and a modern economic system-The CPPCC National Committee "accelerate the promotion of industrial Internet construction" bi-weekly consultation forum to replace the excerpt[N]. CPPCC Daily, 2020.
[3]	工业互联网产业联盟. 工业互联网安全总体要求[EB]. 2018.
	Alliance of Industrial Internet. General requirements for industrial Internet security[EB]. 2018.
[4]	互联网产业联盟. 中国工业互联网安全态势报告[EB]. 2019.
	Alliance of Industry Internet. China industrial internet security situation report（2018）[EB]. 2019.
[5]	KUMAR R , GOYAL R . On cloud security requirements,threats,vulnerabilities and countermeasures:a survey[J]. Computer Science Review, 2019(33): 1-48.
[6]	易树平, 李嘉佳, 易茜 . 基于行为流图的可信交互检测方法[J]. 控制与决策, 2020,35(11): 2715-2722.
	YI S P , LI J J , YI Q . Trustworthy interaction detection method based onuser behavior flow diagram[J]. Control and Decision, 2020,35(11): 2715-2722.
[7]	MALATHI R , JEBERSON RETNARAJ R . An integrated approach of physical biometric authentication system[J]. Procedia Computer Science, 2016(85): 820-826.
[8]	PUSARA M , BRODLEY C E . User re-authentication via mouse movements[C]// Proceedings of the 2004 ACM Workshop on Visualization and Data Mining for Computer Security-VizSEC/DMSEC '04. NewYork:ACM Press, 2004: 1-8.
[9]	AHMED A A E , TRAORE I . A new biometric technology based on mouse dynamics[J]. IEEE Transactions on Dependable and Secure Computing, 2007,4(3): 165-179.
[10]	FEHER C , ELOVICI Y , MOSKOVITCH R ,et al. User identity verification via mouse dynamics[J]. Information Sciences, 2012,201: 19-36.
[11]	ZHENG N , PALOSKI A , WANG H N . An efficient user verification system using angle-based mouse movement biometrics[J]. ACM Transactions on Information and System Security, 2016,18(3): 1-27.
[12]	沈超, 蔡忠闽, 管晓宏 ,等. 基于鼠标行为特征的用户身份认证与监控[J]. 通信学报, 2010,31(7): 68-75.
	SHEN C , CAI Z M , GUAN X H ,et al. User authentication and monitoring based on mouse behavioral features[J]. Journal on Communications, 2010,31(7): 68-75.
[13]	徐剑, 李明洁, 周福才 ,等. 基于用户鼠标行为的身份认证方法[J]. 计算机科学, 2016,43(2): 148-154.
	XU J , LI M J , ZHOU F C ,et al. Identity authentication method based on user's mouse behavior[J]. Computer Science, 2016,43(2): 148-154.
[14]	YI Q , XIONG S Q , WANG B ,et al. Identification of trusted interactive behavior based on mouse behavior considering Web user's emotions[J]. International Journal of Industrial Ergonomics, 2020(76):102903.
[15]	CHONG P , ELOVICI Y , BINDER A . User authentication based on mouse dynamics using deep neural networks:a comprehensive study[J]. IEEE Transactions on Information Forensics and Security, 2020(15): 1086-1101.
[16]	NOY L , ALON U , FRIEDMAN J . Corrective jitter motion shows similar individual frequencies for the arm and the finger[J]. Experimental Brain Research, 2015,233(4): 1307-1320.
[17]	ALPAR O . Frequency spectrograms for biometric keystroke authentication using neural network based classifier[J]. Knowledge-Based Systems, 2017(116): 163-171.
[18]	ALPAR O . TAPSTROKE:a novel intelligent authentication system using tap frequencies[J]. Expert Systems With Applications, 2019(136): 426-438.
[19]	易茜, 黎伟, 易树平 ,等. 基于鼠标行为时频联合分析的用户可信认证[J]. 北京邮电大学学报, 2021,44(4): 121-128.
	YI Q , LI W , YI S P ,et al. Trustworthy identity authentication based on joint time-frequency analysis of mouse behavior[J]. Journal of Beijing University of Posts and Telecommunications, 2021,44(4): 121-128.
[20]	BREIMAN L . Bagging predictors[J]. Machine Learning, 1996,24(2): 123-140.
[21]	MISHRA P K , YADAV A , PAZOKI M . A novel fault classification scheme for series capacitor compensated transmission line based on bagged tree ensemble classifier[J]. IEEE Access, 2018(6): 27373-27382.
[22]	LE TT H , KANG H , KIM H . Household appliance classification using lower odd-numbered harmonics and the bagging decision tree[J]. IEEE Access, 2020(8): 55937-55952.
[23]	工业互联网产业联盟. 工业互联网平台白皮书（2017）[EB]. 2017.
	Alliance of Industrial Internet. Industrial internet platform white paper (2017)[EB]. 2017.
[24]	MORASSO P , MUSSA IVALDI F A . Trajectory formation and handwriting:a computational model[J]. Biological Cybernetics, 1982,45(2): 131-142.
[25]	UNO Y , KAWATO M , SUZUKI R . Formation and control of optimal trajectory in human multijoint arm movement[J]. Biological Cybernetics, 1989,61(2): 89-101.
[26]	LEE D , PORT N L , GEORGOPOULOS A P . Manual interception of moving targets.II.On-line control of overlapping submovements[J]. Experimental Brain Research, 1997,116(3): 421-433.
[27]	NOVAK K E , MILLER L E , HOUK J C . The use of overlapping submovements in the control of rapid hand movements[J]. Experimental Brain Research, 2002,144(3): 351-364.
[28]	郇战, 陈学杰, 吕士云 ,等. 基于多分类器融合的步态识别方法[J]. 计算机应用, 2019,39(3): 712-718.
	HUAN Z , CHEN X J , LYU S Y ,et al. Gait recognition method based on multiple classifier fusion[J]. Journal of Computer Applications, 2019,39(3): 712-718.
[29]	HUANG N E , SHEN Z , LONG S R ,et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proceedings of the Royal Society of London Series A:Mathematical,Physical and Engineering Sciences, 1998,454(1971): 903-995.
[30]	AYENU-PRAH A , ATTOH-OKINE N , . A criterion for selecting relevant intrinsic mode functions in empirical mode decomposition[J]. Advances in Adaptive Data Analysis, 2010,2(1): 1-24.
[31]	MANN K A , WERNERE F W , PALMER A K . Frequency spectrum analysis of wrist motion for activities of daily living[J]. Journal of Orthopaedic Research, 1989,7(2): 304-306.
[32]	HARILAL A , TOFFALINI F , CASTELLANOS J ,et al. TWOS:adataset of malicious insider threat behavior based on a gamified competition[C]// Proceedings of the 2017 International Workshop on Managing Insider Security Threats. New York:ACM, 2017: 45-56.
[33]	HAN H , WANG W Y , MAO B H . Borderline-SMOTE:anew over-sampling method in imbalanced data sets learning[C]// Advances in Intelligent Computing, 2005: 878-887.
[34]	GAMBOA H , FRED A L N , JAIN A K . Webbiometrics:user verification via web interaction[C]// Proceedings of 2007 Biometrics Symposium. Piscataway:IEEE Press, 2007: 1-6.
[35]	SHEN C , CAI Z M , GUAN X H . Continuous authentication for mouse dynamics:a pattern-growth approach[C]// Proceedings of IEEE/IFIP International Conference on Dependable Systems and Networks (DSN 2012). Piscataway:IEEE Press, 2012: 1-12.
[36]	ZHENG N , PALOSKI A , WANG H N . An efficient user verification system via mouse movements[C]// CCS '11:Proceedings of the 18th ACM Conference on Computer and Communications Security. 2011: 139-150.
[37]	LINZ . Research on agent-based human-information system trusted interaction in distributed cooperative work environment[J]. The Open Automation and Control Systems Journal, 2011,3(1): 1-7.

鼠标事件	原始数据中的名称	记号
鼠标移动	slide	m_t
按键按下	left down/right down	d_t
按键弹起	left up/right up	u_t

类型			特征名称			特征数
	v_x
	v_x
	v				最小值
	a		特征本身		最大值
与时间无关	v′′	×	瞬时频率	×	平均值	135
	v′′′		瞬时能量		标准差
	v′′′′				极差
	v_θ
	v_cur
	X		最小值
	Y		最大值
与时间无关	θ	×	平均值			25
	c		标准差
	cd		极差
	s_n
其他	t_n					3
	J

真实用户	FAR	FRR
A	0.05%	0.65%
B	0.18%	0.13%
C	0.09%	0.06%
D	0%	0%
E	0.15%	0.26%
F	0.25%	0.19%
G	0%	0.34%
H	0.06%	0.15%
I	0.09%	0.39%
J	0.24%	0.33%
K	0.36%	0.71%
L	0.08%	0.24%
M	0.06%	0.56%
N	0.03%	0.09%
O	0.06%	0.36%
P	0%	0.03%
Q	0.19%	0.22%
R	0.22%	0.26%
平均	0.12%	0.28%

外部入侵者	样本数量	FAR
入侵者1	470	0.72%
入侵者2	49	1.53%
入侵者3	124	1.41%
入侵者4	97	2.57%
入侵者5	162	1.77%
入侵者6	101	0.80%
平均	167.17	1.47%

分类器	FAR	FRR
Bagged tree（本文）	0.12%	0.28%
Boost trees	0.29%	0.64%
高斯核SVM	0.38%	0.89%
KNN	0.66%	1.83%

鼠标行为HHT变换的工业互联网用户身份认证

User authentication of industrial internet based on HHT transform of mouse behavior

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 37

相关文章 7

Metrics

推荐阅读 0

研究	方法	实验设计	结果
Ahmed等^[9]	MD-NN	22名被试者，短期非受控环境	FAR=2.464 9%
			FRR=2.461 4%
Gamboa等^[34]	Weibull分布	50名被试者，15 min短期受控游戏环境	EER=0.5%
Shen等^[35]	SVM（单核）	28名被试者，30 min短期非受控环境	FAR=0.37%
			FRR=1.12%
Feher等^[10]	MD	25名被试者，短期受控环境	EER=10%
Zheng等^[36]	MD-SVM	14名被试者，短期受控环境	FAR=2.96%
			FRR=0.86%
本文（内部检测）	HHT-BT	4名被试者，超过两年的长期非受控环境	FAR=0.12%
			FRR=0.28%
本文（外部检测）			FAR=1.47%

[1]	于亚, 伏玉笋. 工业互联网边缘终端初始接入可信度量方法研究[J]. 物联网学报, 2022, 6(4): 149-157.
[2]	亓晋, 王微, 陈孟玺, 许斌, 董振江, 孙雁飞. 工业互联网的概念、体系架构及关键技术[J]. 物联网学报, 2022, 6(2): 38-49.
[3]	蔡金莹, 向峰, 左颖, 钟雷, 周平. 面向工业互联网的制造服务协作等级协议轻量级框架研究[J]. 物联网学报, 2021, 5(3): 49-55.
[4]	孙思齐. 石油化工行业工业互联网发展现状分析及展望[J]. 物联网学报, 2021, 5(3): 126-132.
[5]	谭琛,陈美娟,Amuah Ebenezer Ackah. 基于区块链的分布式物联网设备身份认证机制研究[J]. 物联网学报, 2020, 4(2): 70-77.
[6]	常娥,冷卫杰. 工业互联网无线室内定位技术概述[J]. 物联网学报, 2020, 4(2): 129-135.
[7]	张宏科, 程煜钧, 杨冬. 工业网络技术现状与展望[J]. 物联网学报, 2017, 1(1): 13-20.