Visible to the public Learning from Mutants: Using Code Mutation to Learn and Monitor Invariants of a Cyber-Physical System

TitleLearning from Mutants: Using Code Mutation to Learn and Monitor Invariants of a Cyber-Physical System
Publication TypeConference Paper
Year of Publication2018
AuthorsChen, Yuqi, Poskitt, Christopher M., Sun, Jun
Conference Name2018 IEEE Symposium on Security and Privacy (SP)
Date Publishedmay
Keywordsactuators, anomaly detection, attacks, attestation, code mutation, code-modification attacks, CPS, cps privacy, cross-validation, cyber physical systems, cyber-physical system, data logs, Data models, feature extraction, formal verification, Human Behavior, human factors, invariants, learning (artificial intelligence), learnt model, machine learning, model checking, Monitoring, Mutation testing, privacy, program diagnostics, program testing, pubcrawl, real-world water purification plant, Sensors, Software, Statistical Model Checking, supervised machine learning, Support vector machines, SVM-based model, system modelling, water treatment systems
AbstractCyber-physical systems (CPS) consist of sensors, actuators, and controllers all communicating over a network; if any subset becomes compromised, an attacker could cause significant damage. With access to data logs and a model of the CPS, the physical effects of an attack could potentially be detected before any damage is done. Manually building a model that is accurate enough in practice, however, is extremely difficult. In this paper, we propose a novel approach for constructing models of CPS automatically, by applying supervised machine learning to data traces obtained after systematically seeding their software components with faults ("mutants"). We demonstrate the efficacy of this approach on the simulator of a real-world water purification plant, presenting a framework that automatically generates mutants, collects data traces, and learns an SVM-based model. Using cross-validation and statistical model checking, we show that the learnt model characterises an invariant physical property of the system. Furthermore, we demonstrate the usefulness of the invariant by subjecting the system to 55 network and code-modification attacks, and showing that it can detect 85% of them from the data logs generated at runtime.
DOI10.1109/SP.2018.00016
Citation Keychen_learning_2018