Balancing Security And Schedulability: WCET Evaluation And Security Optimization In CPS
Abstract
—As real-time cyber-physical systems (CPS) become increasingly prevalent and interconnected, ensuring their security becomes essential. Yet, integrating runtime security mechanisms into real-time systems is challenging due to non-uniform performance overheads that may differ significantly under average- and worst-case behaviors. This paper systematically examines how commonly used runtime memory-safety defenses affect worst-case task performance in real-world CPS applications. The experiments reveal that security mechanisms vary markedly in their computational impact, providing practical guidance for their deployment. They also highlight the opportunity to selectively combine defenses to enhance overall protection. The paper further presents an optimization framework, QUASARRT, and a case study on ArduPilot demonstrating how system designers can apply these findings to maximize their defense coverage while maintaining real-time schedulability. Results indicate that QUASARRT is effective at finding the most secure feasible task configurations, even under complex schedulability analysis.
Recommended Citation
Z. Li et al., "Balancing Security And Schedulability: WCET Evaluation And Security Optimization In CPS," Proceedings of the IEEE Real Time and Embedded Technology and Applications Symposium Rtas, pp. 344 - 359, Institute of Electrical and Electronics Engineers, Jan 2026.
The definitive version is available at https://doi.org/10.1109/RTAS68450.2026.00038
Department(s)
Computer Science
International Standard Serial Number (ISSN)
1545-3421
Document Type
Article - Conference proceedings
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2026 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.
Publication Date
01 Jan 2026
Comments
National Aeronautics and Space Administration, Grant 80NSSC21K1741