ZOE: Content-Based Anomaly Detection for Industrial Control Systems
Authors:
- Christian WressneggerCWChristian Wressnegger
- Ansgar KellnerAKAnsgar Kellner
- Konrad RieckKRKonrad Rieck
- 1 June 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 127-138
Abstract
Due its complexity and a multitude of proprietary components, industrial control systems are an immanently difficult field of application for intrusion detection. Proprietary binary protocols and the lack of public specifications have forced the research community to move away from content-based detection to more abstract concepts. In this paper, we show that in contrast to prior belief the content of unknown binary protocols can very well be modeled. ZOE derives prototype models that are specific to individual types of messages in order to capture the characteristics of arbitrary binary protocols and enable detecting different forms of attacks as anomalies. In an evaluation based on 6 days of network traffic recorded at a large power plant (1,900 MW) with over 92,000 unique devices, we demonstrate that ZOE improves upon related approaches by up to an order of magnitude in detection performance, but also significantly decreases false positives.Keywords
This publication has 49 references indexed in Scilit:
- Last Line of Defense: A Novel IDS Approach Against Advanced Threats in Industrial Control Systems2017
- Limiting the Impact of Stealthy Attacks on Industrial Control Systems2016
- Comprehensive Analysis and Detection of Flash-Based Malware2016
- Exploiting Bro for Intrusion Detection in a SCADA System2016
- Coding sensor outputs for injection attacks detection2014
- A close look on n -grams in intrusion detection2013
- Learning stateful models for network honeypots2012
- ScriptGen: an automated script generation tool for honeyd2006
- Testing network-based intrusion detection signatures using mutant exploits2004
- Service specific anomaly detection for network intrusion detection2002