Full Text:   <835>

Summary:  <272>

CLC number: TP309.1

On-line Access: 2017-04-12

Received: 2016-09-09

Revision Accepted: 2016-11-30

Crosschecked: 2017-03-23

Cited: 0

Clicked: 1920

Citations:  Bibtex RefMan EndNote GB/T7714


Wen-yuan Xu


-   Go to

Article info.
Open peer comments

Frontiers of Information Technology & Electronic Engineering  2017 Vol.18 No.4 P.519-534


NIPAD: a non-invasive power-based anomaly detection scheme for programmable logic controllers

Author(s):  Yu-jun Xiao, Wen-yuan Xu, Zhen-hua Jia, Zhuo-ran Ma, Dong-lian Qi

Affiliation(s):  School of Electrical Engineering, Zhejiang University, Hangzhou 310027, China; more

Corresponding email(s):   xiaoyujun2@gmail.com, xuwenyuan@gmail.com, zhenhua@winlab.rutgers.edu, mzr@zju.edu.cn, qidl@zju.edu.cn

Key Words:  Industrial control system, Programmable logic controller, Side-channel, Anomaly detection, Long short-term memory neural networks

Yu-jun Xiao, Wen-yuan Xu, Zhen-hua Jia, Zhuo-ran Ma, Dong-lian Qi. NIPAD: a non-invasive power-based anomaly detection scheme for programmable logic controllers[J]. Frontiers of Information Technology & Electronic Engineering, 2017, 18(4): 519-534.

@article{title="NIPAD: a non-invasive power-based anomaly detection scheme for programmable logic controllers",
author="Yu-jun Xiao, Wen-yuan Xu, Zhen-hua Jia, Zhuo-ran Ma, Dong-lian Qi",
journal="Frontiers of Information Technology & Electronic Engineering",
publisher="Zhejiang University Press & Springer",

%0 Journal Article
%T NIPAD: a non-invasive power-based anomaly detection scheme for programmable logic controllers
%A Yu-jun Xiao
%A Wen-yuan Xu
%A Zhen-hua Jia
%A Zhuo-ran Ma
%A Dong-lian Qi
%J Frontiers of Information Technology & Electronic Engineering
%V 18
%N 4
%P 519-534
%@ 2095-9184
%D 2017
%I Zhejiang University Press & Springer
%DOI 10.1631/FITEE.1601540

T1 - NIPAD: a non-invasive power-based anomaly detection scheme for programmable logic controllers
A1 - Yu-jun Xiao
A1 - Wen-yuan Xu
A1 - Zhen-hua Jia
A1 - Zhuo-ran Ma
A1 - Dong-lian Qi
J0 - Frontiers of Information Technology & Electronic Engineering
VL - 18
IS - 4
SP - 519
EP - 534
%@ 2095-9184
Y1 - 2017
PB - Zhejiang University Press & Springer
ER -
DOI - 10.1631/FITEE.1601540

industrial control systems (ICSs) are widely used in critical infrastructures, making them popular targets for attacks to cause catastrophic physical damage. As one of the most critical components in ICSs, the programmable logic controller (PLC) controls the actuators directly. A PLC executing a malicious program can cause significant property loss or even casualties. The number of attacks targeted at PLCs has increased noticeably over the last few years, exposing the vulnerability of the PLC and the importance of PLC protection. Unfortunately, PLCs cannot be protected by traditional intrusion detection systems or antivirus software. Thus, an effective method for PLC protection is yet to be designed. Motivated by these concerns, we propose a non-invasive power-based anomaly detection scheme for PLCs. The basic idea is to detect malicious software execution in a PLC through analyzing its power consumption, which is measured by inserting a shunt resistor in series with the CPU in a PLC while it is executing instructions. To analyze the power measurements, we extract a discriminative feature set from the power trace, and then train a long short-term memory (LSTM) neural network with the features of normal samples to predict the next time step of a normal sample. Finally, an abnormal sample is identified through comparing the predicted sample and the actual sample. The advantages of our method are that it requires no software modification on the original system and is able to detect unknown attacks effectively. The method is evaluated on a lab testbed, and for a trojan attack whose difference from the normal program is around 0.63%, the detection accuracy reaches 99.83%.


概要:工业控制系统广泛应用于关键基础设施的建设中,关系到国计民生,因此,攻击者越来越多地将其作为攻击目标,并造成严重的破坏。可编程逻辑控制器(Programmable logic controller, PLC)作为工业控制系统中的核心组件,能够直接控制现场设备,一旦PLC中运行了恶意程序,则可能直接造成重大财产损失甚至是人员伤亡。近些年来,针对PLC的攻击事件显著增加,这表明PLC存在很大的脆弱性,同时也提醒人们保护PLC安全的重要性。不幸的是,传统的入侵检测系统和杀毒软件并不能很好地保护PLC的安全,因此,针对PLC的有效的安全防护方案有待被研究。基于上述背景,本文提出了一种非侵入式的基于功耗的PLC异常检测方案。该方案通过分析PLC运行时的功耗变化来检测PLC中是否运行异常程序,分为功耗信息获取与功耗分析两部分。采集功耗信息是通过在PLC的供电线上串入一个电阻实现的,当PLC运行时,测量电阻两端的电压即可获取CPU的功耗信息。为了更好的分析功耗信息,本文首先从原始功耗数据中提取有效的特征值组合,然后利用正常样本来训练一个基于长短记忆(long short-term memory, LSTM)单元的神经网络模型,利用该模型对后续正常样本进行预测,通过比较测量到的功耗信息与预测的功耗信息,可以确定当前PLC中运行的程序是否为正常程序。该方案的优点是无需对原工控系统的封装部分进行软硬件的修改,且无需负样本即可实现对未知攻击的检测。我们在实验室测试平台上对该方法进行了评估,实验表明,对于原程序,只需改动0.63%即可达到99.83%的准确率。


Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article


[1]Alcaraz, C., Zeadally, S., 2013. Critical control system protection in the 21st century. Computer, 46(10):74-83.

[2]Alcaraz, C., Zeadally, S., 2015. Critical infrastructure protection: requirements and challenges for the 21st century. Int. J. Crit. Infrastr. Protect., 8:53-66.

[3]Bencsáth, B., Pék, G., Buttyán, L., et al., 2012. The cousins of Stuxnet: Duqu, Flame, and Gauss. Fut. Int., 4(4):971-1003.

[4]Bolton, W., 2015. Programmable Logic Controllers (6th Ed.). Newnes, USA.

[5]Bullock, J., Conservatoire, U.C.E.B., 2007. LibXtract: a lightweight library for audio feature extraction. Proc. Int. Computer Music Conf., p.1-4.

[6]Candes, E.J., Tao, T., 2006. Near-optimal signal recovery from random projections: universal encoding strategies? IEEE Trans. Inform. Theory, 52(12):5406-5425.

[7]Cárdenas, A.A., Amin, S., Sastry, S., 2008. Research challenges for the security of control systems. Proc. 3rd Conf. on Hot Topics in Security, Article 6.

[8]Chen, T.M., Abu-Nimeh, S., 2011. Lessons from Stuxnet. Computer, 44(4):91-93.

[9]Clark, S.S., Ransford, B., Rahmati, A., et al., 2013. WattsUpDoc: power side channels to nonintrusively discover untargeted malware on embedded medical devices. Proc. USENIX Workshop on Health Information Technologies, p.1-11.

[10]Coletta, A., Armando, A., 2015. Security monitoring for industrial control systems. Proc. Conf. on Cybersecurity of Industrial Control Systems, p.48-62.

[11]Dalal, N., Triggs, B., 2005. Histograms of oriented gradients for human detection. Proc. IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, p.886-893.

[12]Formby, D., Srinivasan, P., Leonard, A., et al., 2016. Who’s in control of your control system? Device fingerprinting for cyber-physical systems. Proc. Network and Distributed System Security Symp., p.1-13.

[13]García-Teodoro, P., Díaz-Verdejo, J., Maciá-Fernández, G., emph{et al.}, 2009. Anomaly-based network intrusion detection: techniques, systems and challenges. Comput. Secur., 28(1-2):18-28.

[14]Gers, F.A., Schmidhuber, J.A., Cummins, F., 2000. Learning to forget: continual prediction with LSTM. Neur. Comput., 12(10):2451-2471.

[15]Gonzalez, C.A., Hinton, A., 2014. Detecting malicious software execution in programmable logic controllers using power fingerprinting. Proc. Int. Conf. on Critical Infrastructure Protection, p.15-27.

[16]Johnson, R.E., 2010. Survey of SCADA security challenges and potential attack vectors. Proc. Int. Conf. for Internet Technology and Secured Transactions, p.1-5.

[17]Kesler, B., 2011. The vulnerability of nuclear facilities to cyber attack. Strat. Insights, 10(1):15-25.

[18]Krotofil, M., Gollmann, D., 2013. Industrial control systems security: what is happening? Proc. 11th IEEE Int. Conf. on Industrial Informatics, p.670-675.

[19]Langner, R., 2011. Stuxnet: dissecting a cyberwarfare weapon. IEEE Secur. Priv. 9(3):49-51.

[20]Lee, H., Battle, A., Raina, R., et al., 2006. Efficient sparse coding algorithms. Proc. 19th Int. Conf. on Neural Information Processing Systems, p.801-808.

[21]Lowe, D.G., 2004. Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vis., 60(2):91-110.

[22]Macaulay, T., Singer, B.L., 2011. Cybersecurity for Industrial Control Systems: SCADA, DCS, PLC, HMI, and SIS. CRC Press, USA.

[23]Malhotra, P., Vig, L., Shroff, G., et al., 2015. Long short term memory networks for anomaly detection in time series. Proc. European Symp. on Artificial Neural Networks, Computational Intelligence and Maching Learning, p.89-94.

[24]Manevitz, L.M., Yousef, M., 2002. One-class SVMs for document classification. J. Mach. Learn. Res., 2:139-154.

[25]Mantere, M., Uusitalo, I., Sailio, M., et al., 2012. Challenges of machine learning based monitoring for industrial control system networks. Proc. 26th Int. Conf. on Advanced Information Networking and Applications Workshops, p.968-972.

[26]Morris, T., Vaughn, R., Dandass, Y., 2012. A retrofit network intrusion detection system for MODBUS RTU and ASCII industrial control systems. Proc. 45th Hawaii Int. Conf. on System Science, p.2338-2345.

[27]Nandakumar, K., Jain, A.K., 2004. Local correlation-based fingerprint matching. Proc. ICVGIP, p.503-508.

[28]Ni, B., Moulin, P., Yang, X., et al., 2015. Motion part regularization: improving action recognition via trajectory group selection. Proc. IEEE Conf. on Computer Vision and Pattern Recognition, p.3698-3706.

[29]Pearson, K., 1901. Mathematical contributions to the theory of evolution. X. Supplement to a memoir on skew variation. Phil. Trans. R. Soc. A, 197:443-459.

[30]Peng, Y., Xiang, C., Gao, H., et al., 2015. Industrial control system fingerprinting and anomaly detection. Proc. Int. Conf. on Critical Infrastructure Protection, p.73-85.

[31]Piggin, R., 2015. Are industrial control systems ready for the cloud? Int. J. Crit. Infrastr. Protect., 9(C):38-40.

[32]Ponomarev, S., Atkison, T., 2016. Industrial control system network intrusion detection by telemetry analysis. IEEE Trans. Depend. Sec. Comput., 13(2):252-260.

[33]Pretorius, B., van Niekerk, B., 2016. Cyber-security for ICS/SCADA: a South African perspective. Int. J. Cyber Warf. Terror., 6(3):1-16.

[34]Shang, W., Zeng, P., Wan, M., et al., 2016. Intrusion detection algorithm based on OCSVM in industrial control system. Secur. Commun. Netw., 9(10):1040-1049.

[35]Slay, J., Miller, M., 2007. Lessons learned from the Maroochy water breach. Proc. Int. Conf. on Critical Infrastructure Protection, p.73-82.

[36]Stone, S.J., Temple, M.A., Baldwin, R.O., 2015. Detecting anomalous programmable logic controller behavior using RF-based Hilbert transform features and a correlation-based verification process. Int. J. Crit. Infrastr. Protect., 9(C):41-51.

[37]Stouffer, K.A., Falco, J.A., Scarfone, K.A., 2011. Guide to Industrial Control Systems (ICS) Security: Supervisory Control and Data Acquisition (SCADA) Systems, Distributed Control Systems (DCS), and Other Control System Configurations such as Programmable Logic Controllers (PLC). Technical Report SP 800-82, National Institute of Standards and Technology, USA.

[38]Wang, H., Kläser, A., Schmid, C., et al., 2013. Dense trajectories and motion boundary descriptors for action recognition. Int. J. Comput. Vis., 103(1):60-79.

[39]Xu, J., Yang, G., Man, H., et al., 2013. L1 graph based on sparse coding for feature selection. Proc. Int. Symp. on Neural Networks, p.594-601.

[40]Zhong, W., Lu, H., Yang, M., 2012. Robust object tracking via sparsity-based collaborative model. Proc. IEEE Conf. on Computer Vision and Pattern Recognition, p.1838-1845.

Open peer comments: Debate/Discuss/Question/Opinion


Please provide your name, email address and a comment

Journal of Zhejiang University-SCIENCE, 38 Zheda Road, Hangzhou 310027, China
Tel: +86-571-87952783; E-mail: cjzhang@zju.edu.cn
Copyright © 2000 - Journal of Zhejiang University-SCIENCE