TaskShuffler++: Real-Time Schedule Randomization for Reducing Worst-Case Vulnerability to Timing Inference Attacks
This addresses security vulnerabilities for real-time systems, but it is incremental as it builds on existing randomization techniques.
The paper tackles the problem of timing inference attacks in real-time systems by introducing a schedule randomization algorithm that reduces an adversary's ability to predict task execution times while maintaining schedulability, resulting in a significant reduction in the best chance for correct prediction.
This paper presents a schedule randomization algorithm that reduces the vulnerability of real-time systems to timing inference attacks which attempt to learn the timing of task execution. It utilizes run-time information readily available at each scheduling decision point to increase the level of uncertainty in task schedules, while preserving the original schedulability. The randomization algorithm significantly reduces an adversary's best chance to correctly predict what tasks would run at arbitrary times. This paper also proposes an information-theoretic measure that can quantify the worst-case vulnerability, from the defender's perspective, of an arbitrary real-time schedule.