Accession Number : AD1051267


Title :   CORESAFE: A Formal Approach against Code Replacement Attacks on Cyber Physical Systems


Descriptive Note : Technical Report,20 Sep 2016,19 Sep 2017


Corporate Author : INDIAN INSTITUTE OF TECHNOLOGY KANPUR Kanpur India


Personal Author(s) : Shukla,Sandeep ; Das,Debleena ; Banerjee,Anshuman


Full Text : https://apps.dtic.mil/dtic/tr/fulltext/u2/1051267.pdf


Report Date : 19 Apr 2018


Pagination or Media Count : 15


Abstract : Industrial Control Systems (ICS) used in manufacturing, power generators and other critical infrastructure monitoring and control are ripe targets for cyber-attacks these days. Examples of such attacks are abundant such as attacks on Iranian nuclear enrichment plant with Stuxnet in 2009, on German steel plant in 2014, Ukrainian power system in 2015 and 2016. Usually in ICS, multiple control loops work concurrently and share various resources including the communication bus through which they interact with sensors and actuators. Real-time scheduling of concurrent control applications while competing for shared resources demands a delicate balance between performance and real-time constraints. A possible insider attack could be the replacement of a previously vetted control application or other components in the system, during a system update. In this project, we worked on an automated framework that addresses the effect of such replacement attacks from the perspective of loss of control performance. Given a set of control components, a control objective to be satisfied by the control ensemble, the question of schedulability and synthesis of a scheduler that can ensure the desired control performance has been recently studied in literature. In this project, we extended this idea further to build an automata theoretic framework for assessment of replacement attacks on schedulability. We have built an end-to-end framework that takes in a set of control components, their variants (after replacement), a control objective to be guaranteed, and performs an automated schedulability assessment. We report some preliminary experiments of our framework on simple benchmarks.


Descriptors :   computer network security , malware , cyberattacks , control systems , automata theory


Subject Categories : Computer Systems Management and Standards


Distribution Statement : APPROVED FOR PUBLIC RELEASE