Framework for the Establishment of Resource-Aware Data Mining Techniques on Critical Infrastructures
Nowadays, the development of modern societies is based on the availability of essential services by means of industrial control systems or SCADA systems which form part of what has come to be known as critical infrastructures. SCADA systems are usually implemented in a distributed manner, in which some remote terminal units (RTU) are in charge of compiling all of the information from the sensors in the field. The implementation of any efficient protection mechanism in these RTUs demands a “context” and “resource aware” behavior, through the development of intelligent methods that in an efficient way could allow the device to react in a proactive way. However, RTUs are characterized by computational and storage limitations which make it difficult to provide the “intelligence” necessary to develop new decentralized protection systems, which could be useful for the early incident detection based on data mining techniques. This work deals with the problem of executing a classification algorithm in a device with limited computational possibilities. The design presented is characterized by its modularity, adaptability to the available resources, together with its capacity to be reused in other systems with similar characteristics. Results of the experiments carried out are also presented.
KeywordsData Mining Bayesian Network Intrusion Detection System Data Mining Technique Critical Infrastructure
Unable to display preview. Download preview PDF.
- 3.Bernstein, A., Provost, F.: An intelligent assistant for the knowledge discovery process. Information Systems Working Papers Series (2001)Google Scholar
- 4.Krutz, R.L.: Securing SCADA systems, pp. 4–16. Wiley Publishing, Inc. (2006)Google Scholar
- 5.Linda, O., Vollmer, T., Manic, M.: Neural network based intrusion detection system for critical infrastructures. In: Proceedings of the 2009 International Joint Conference on Neural Networks, IJCNN 2009, pp. 102–109. IEEE Press, Piscataway (2009)Google Scholar
- 7.Saitta, L., May, M.: Blueprint in Ubiquitous Knowledge Discovery (2007)Google Scholar
- 9.Menasalvas, E., Eibe, S., Gomes, J., Zanda, A.: Context-aware recommendations in ubiquitous devices (2008)Google Scholar
- 10.Ning, P., Jajodia, S.: Intrusion Detection Basics. In: Bidgoli, H, ed. (2006)Google Scholar
- 11.Scarfone, K., Mell, P.: Intrusion detection and prevention systems. In: Handbook of Information and Communication Security, pp. 177–192 (2010)Google Scholar
- 13.Witten, I.H., Frank, E.: Data Mining: Practical machine learning tools and techniques, 2nd edn. Morgan Kaufmann (2005)Google Scholar
- 14.Wu, C.H., Yang, C.H., Lo, S.C., Vichare, N., Rhem, E., Pecht, M.: Automatic data mining for telemetry database of computer systems. Microelectronics Reliability (October 2010)Google Scholar
- 15.Yu, Z., Tsai, J.J.P., Weigert, T.: An adaptive automatically tuning intrusion detection system. ACM Transactions on Autonomous and Adaptive Systems 3 (August 2008)Google Scholar
- 17.Zhang, J., Zulkernine, M.: Anomaly based network intrusion detection with unsupervised outlier detection. In: 2006 IEEE International Conference on Communications, pp. 2388–2393 (2006)Google Scholar