Enforcing memory safety in cyber-physical systems

Abstract

Cyber-Physical Systems (CPS) integrate computations and communications with physical processes and are being widely adopted in various application areas. However, the increasing prevalence of cyber attacks targeting them poses a growing security concern. In particular, attacks exploiting memory-safety vulnerabilities constitute a major attack vector against CPS, because embedded systems often rely on unsafe but fast programming languages to meet their hard time constraints. A wide range of countermeasures has been developed to provide protection against these attacks. However, the most reliable countermeasures incur in high runtime overheads. In this work, we explore the applicability of strong countermeasures against memory-safety attacks in the context of realistic Industrial Control Systems (ICS). To this end, we design an experimental setup, based on a secure water treatment plant (SWaT) to empirically measure the memory safety overhead (MSO) caused by memory-safe compilation of the Programmable Logic Controller (PLC). We then quantify the tolerability of this overhead in terms of the expected real-time constraints of SWaT. Our results show high effectiveness of the security measure in detecting memory-safety violations and a MSO (197.86 μ s per scan-cycle) that is also tolerable for the SWaT simulation. We also discuss how different parameters impact the execution time of PLCs and the resulting absolute MSO.