Improving ICS security through Honeynets and Machine Learning techniques

Abstract

The internet of things(IoT), the Industrial Internet of Things (IIoT), and Cyber-Physical Systems (CPS) can be seen everywhere, Home applications, Buildings, Cars, Space Industry, Military, Health Care, and in many other fields. On the other hand, they become an easier target for attackers, due to many reasons including the limitation of hardware, so from that point, companies start working to build a secure systems by keep themselves updated about their system threats and vulnerabilities, and also by studying how the attackers can gets into their system, how they act, what is the attack flow, and also the identity of the attackers by trapping and tricking them into believing that they have got access to the actual system or assets . And that’s what it's called a Honeypot. [1] As the technology keeps changing and becomes more powerful, so do the attackers, and for that reason companies should use new techniques to enhance Honeypots efficacy by making it undetectable by cybercriminals, more usable and make use of the information that the honeypots gather in a more efficient way. Moreover, Machine Learning (ML) techniques are able to provide intelligence to IoT, IIoT, and ICS systems and networks, and enhance its ability to deal with various security problems, hence, in this research, we are developing a new solution that improves the architecture of SCADA (An ICS System) by adding CamouflageNet Honeynet into it and ML techniques, in order to defend and acquisition system security performance.