Advertisement

The Impact of Different Feature Scaling Methods on Intrusion Detection for in-Vehicle Controller Area Network (CAN)

  • Siti-Farhana LokmanEmail author
  • Abu Talib Othman
  • Muhamad Husaini Abu Bakar
  • Shahrulniza Musa
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 1132)

Abstract

Numerous security researchers have a growing interest in the vulnerabilities of the in-vehicle Controller Area Network (CAN) bus system to cyber-attacks. The adversaries can leverage these vulnerabilities in manipulating vehicle functions and harming the drivers’ safety. Some security mechanisms proposed for CAN bus in detecting anomalies have favoured over the one-class classification, where it constructs a decision boundary from normal instances. Nevertheless, the accuracy performance of the classifier is highly influenced by the data representation. Judging from this fact, this paper analyses the advantage of utilizing different feature scaling technique as in to obtain higher classification accuracy of the classifier algorithms. To serve this purpose, the CAN bus datasets in this paper are scaled using standardization, min-max, and quantile, and are evaluated using one-class classifier model used in automotive CAN bus. The results exhibit that integrating different feature scaling techniques could greatly enhance the classification accuracy of the classifiers.

Keywords

Anomaly-based detection Neural network Controller Area Network Feature scaling One-class classification 

References

  1. 1.
    Sakiz, F., Sen, S.: A survey of attacks and detection mechanisms on intelligent trans-portation systems: VANETs and IoV. Ad Hoc Netw. 61, 33–50 (2017)CrossRefGoogle Scholar
  2. 2.
    Miller, C., Valasek, C.: Remote exploitation of an unaltered passenger vehicle. Black Hat USA (2015)Google Scholar
  3. 3.
    Koscher, K., et al.: Experimental security analysis of a modern automobile. In: 2010 IEEE Symposium on Security and Privacy (SP), pp. 447–462 (2010)Google Scholar
  4. 4.
    Hoppe, T., Kiltz, S., Dittmann, J.: Security threats to automotive CAN networks – practical examples and selected short-term countermeasures. In: Harrison, M.D., Sujan, M.-A. (eds.) SAFECOMP 2008. LNCS, vol. 5219, pp. 235–248. Springer, Heidelberg (2008).  https://doi.org/10.1007/978-3-540-87698-4_21CrossRefGoogle Scholar
  5. 5.
    Hoppe, T., Kiltz, S., Dittmann, J.: Applying intrusion detection to automotive it-early insights and remaining challenges. J. Inform. Assur. Secur. (JIAS) 4(6), 226–235 (2009)Google Scholar
  6. 6.
    Martinelli, F., Mercaldo, F., Nardone, V., Santone, A.: Car hacking identification through fuzzy logic algorithms. In: IEEE International Conference on Fuzzy Systems, Naples (2017)Google Scholar
  7. 7.
    Tomlinson, A., Bryans, J., Shaikh, S.A.: Using a one-class compound classifier to detect in-vehicle network attacks. In GECCO 2018 Companion: Genetic and Evolutionary Computation Conference Companion. ACM, Kyoto (2018).  https://doi.org/10.1145/3205651.3208223
  8. 8.
    Weber, M., Klug, S., Sax, E., Zimmer, B.: Embedded hybrid anomaly detection for automotive CAN communication (2018)Google Scholar
  9. 9.
    Xing, Y., Lv, C., Wang, H., Cao, D. Recognizing driver braking intention with vehicle data using unsupervised learning methods (2017)Google Scholar
  10. 10.
    Loukas, G., Vuong, T., Heartfield, R., Sakellari, G., Yoon, Y., Gan, D.: Cloud-based cyber-physical intrusion detection for vehicles using deep learning. IEEE Access 6, 3491–3508 (2017)CrossRefGoogle Scholar
  11. 11.
    Nawi, N.M., et al.: The effect of pre-processing techniques and optimal parameters selection on back propagation neural networks. Int. J. Adv. Sci. Eng. Inform. Technol. 7(3), 770–777 (2017)CrossRefGoogle Scholar
  12. 12.
    Kumar, D.A., Venugopalan, S.: The effect of normalization on intrusion detection classifiers (Naïve Bayes and J48). Int. J. Future Revolut. Comput. Sci. Commun. Eng. 3, 60–64 (2017)Google Scholar
  13. 13.
    Kang, M.J., Kang, J.W.: Intrusion detection system using deep neural network for in-vehicle network security. PLoS One 11(6), e0155781 (2016)CrossRefGoogle Scholar
  14. 14.
    Wasicek, A., Weimerskirch, A.: Recognizing manipulated electronic control units (No. 2015-01-0202). SAE Technical Paper (2015)Google Scholar
  15. 15.
    Taylor, A., Leblanc, S., Japkowicz, N.: Anomaly detection in automobile control network data with long short-term memory networks. In: 2016 IEEE International Conference on Data Science and Advanced Analytics (DSAA), pp. 130–139 (2016)Google Scholar
  16. 16.
    Pan, M., Zhang, J.: Quantile normalization for combining gene-expression da-tasets. Biotechnol. Biotechnol. Equip. 32(3), 751–758 (2018)MathSciNetCrossRefGoogle Scholar
  17. 17.
    Upender, B.P., Dean, A.G.: Variability of CAN network performance. In: Proceedings of the 3rd International CAN Conference ICC (1996)Google Scholar
  18. 18.
    Taylor, A., Japkowicz, N., Leblanc, S.: Frequency-based anomaly detection for the automotive CAN bus. In: 2015 World Congress on Industrial Control Systems Security (WCICSS), pp. 45–49. IEEE (2015)Google Scholar
  19. 19.
    Lokman, S.F., Othman, A.T., Bakar, M.H.A., Razuwan, R.: Stacked sparse autoencoders-based outlier discovery for in-vehicle controller area network (CAN). Int. J. Eng. Technol. 7(4.33), 375–380 (2018).  https://doi.org/10.14419/ijet.v7i4.33.26078CrossRefGoogle Scholar
  20. 20.
    Hicks, S.C., Okrah, K., Paulson, J.N., Quackenbush, J., Irizarry, R.A., Bravo, H.C.: Smooth quantile normalization. Biostatistics 19(2), 185–198 (2017)MathSciNetCrossRefGoogle Scholar
  21. 21.
    Hansen, K.D., Irizarry, R.A., Wu, Z.: Removing technical variability in RNA-seq data using conditional quantile normalization. Biostatistics 13(2), 204–216 (2012)CrossRefGoogle Scholar
  22. 22.
    Monot, A., Navet, N., Bavoux, B.: Impact of clock drifts on CAN frame response time distributions. In: ETFA2011, pp. 1–4. IEEE (2011)Google Scholar
  23. 23.
    Potter, K., Hagen, H., Kerren, A., Dannenmann, P.: Methods for presenting statistical information: the box plot. Vis. Large Unstr. Data Sets 4, 97–106 (2006)Google Scholar
  24. 24.
    Moya, M.M., Hush, D.R.: Network constraints and multi-objective optimization for one-class classification. Neural Netw. 9(3), 463–474 (1996)CrossRefGoogle Scholar
  25. 25.
    Ghafoori, Z., Erfani, S.M., Rajasegarar, S., Bezdek, J.C., Karunasekera, S., Leckie, C.: Efficient unsupervised parameter estimation for one-class support vector machines. IEEE Trans. Neural Netw. Learn. Syst. 29(10), 5057–5070 (2018)CrossRefGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2020

Authors and Affiliations

  1. 1.System Engineering and Energy LaboratoryUniversiti Kuala Lumpur, Malaysian-Spanish InstituteKulimMalaysia
  2. 2.Universiti Kuala Lumpur, Malaysian Institute of Information TechnologyKuala LumpurMalaysia

Personalised recommendations