NeTS: Small: Collaborative Research: Measurement and Modeling of Industrial Control Networks

University Of Texas At Dallas, Richardson TX

Investigators

Abstract

Operators of industrial control systems have limited software tools for obtaining situational awareness of their networks and processes. This project seeks to characterize network traffic for a rich variety of industrial control networks. Because the research community lacks access to (mostly proprietary) data sets from operating control systems, our ability to identify anomalous behavior and protect weaknesses is a concern for critical U.S. infrastructure. The research team has obtained a collection of network data sets that will allow them to study the performance (and security) properties of various types of industrial control systems, including wastewater treatment, power, and solar energy. By examining the similarities and differences of these data sets, this analysis will yield models of critical infrastructure control system traffic that will aid in anomaly detection and further our understanding of these networks in general. The project will characterize network flows, and perform deep-packet inspection in a protocol- and process-aware fashion. The project will develop open-source tools implementing anomaly detection algorithms to be used by researchers and the process control industry. Since the anomaly detection tools are passive, they can ultimately be deployed in the operational environments of industrial partners to help them improve the visibility and understanding of their control networks. This work will be performed in collaboration with researchers in Israel under the US-Israel Binational Science Foundation joint funding program.