Integrating security constraints into fixed priority real-time schedulers

Traditionally, most real-time systems (RTS) were considered to be invulnerable to security breaches and external attacks. This was mainly due to the use of proprietary hardware and protocols in such systems along with physical isolation. Hence, security and RTS were considered to be separate domains. This assumption is being challenged by recent events that highlight the vulnerabilities in such systems. In this paper, we focus on how to integrate security as a first-class principle in the design of RTS. We demonstrate how certain security requirements can be cast as real-time scheduling constraints. We use information leakage as a motivating problem to illustrate our techniques and focus on the class of fixed-priority real-time schedulers. We evaluate our approach both analytically as well as using simulations and discuss the tradeoffs in using such an approach. Our work shows that many real-time task sets can be scheduled using our methods without significant performance impact.