Anomaly detection in SCADA systems : a network based approach

Abstract

Supervisory Control and Data Acquisition (SCADA) networks are commonly deployed to aid the operation of large industrial facilities, such as water treatment facilities. Historically, these networks were composed by special-purpose embedded devices communicating through proprietary protocols. However, modern deployments commonly make use of commercial off-the-shelf devices and standard communication protocols, such as TCP/IP. Furthermore, these networks are becoming increasingly interconnected, allowing communication with corporate networks and even the Internet. As a result, SCADA networks become vulnerable to cyber attacks, being exposed to the same threats that plague traditional IT systems. In our view, measurements play an essential role in validating results in network research; therefore, our first objective is to understand how SCADA networks are utilized in practice. To this end, we provide the first comprehensive analysis of real-world SCADA traffic. We analyze five network packet traces collected at four different critical infrastructures: two water treatment facilities, one gas utility, and one electricity and gas utility. We show, for instance, that exiting network traffic models developed for traditional IT networks cannot be directly applied to SCADA network traffic. We also confirm two SCADA traffic characteristics: the stable connection matrix and the traffic periodicity, and propose two intrusion detection approaches that exploit them. In order to exploit the stable connection matrix, we investigate the use of whitelists at the flow level. We show that flow whitelists have a manageable size, considering the number of hosts in the network, and that it is possible to overcome the main sources of instability in the whitelists. In order to exploit the traffic periodicity, we focus our attention to connections used to retrieve data from devices in the field network. We propose PeriodAnalyzer, an approach that uses deep packet inspection to automatically identify the different messages and the frequency at which they are issued. Once such normal behavior is learned, PeriodAnalyzer can be used to detect data injection and Denial of Service attacks.