Security and Reliability of Safety-Critical RTOS

Abstract

Real-Time Operating System (RTOS) presents a computing environment with the ability to react to events within a strictlydefined period. Modern domain-specific (e.g., aerospace, industrial control, defense, and medical) embedded systems include safety-critical applications whose failures would result in catastrophic events. Such systems avoid the general-purpose operating system and leverage RTOS to work in increasingly dynamic and harsh environments. Moreover, many of these systems operate over networks, making them susceptible to various attacks. Hence, the reliability and security of embedded systems running on top of RTOS are equally important. Although research on these two crucial components has been conducted and presented for traditional operating systems, there has been a significant research thrust on RTOS due to the recent emergence of the internet of things (IoT). This article presents an in-depth and systematic review of the security and reliability of safety-critical systems. We highlight the best practices and recent research efforts and their feasibility in the context of safety-critical RTOS. We compare the solutions qualitatively and discuss their pros and cons. We conclude the article with open-ended research questions.

Measurement of Kernel Preemption Time of Real Time Operating Systems

“…Hard RTS is subject to crash or may cause physical hazards upon not meeting the deadline. A mission critical system, for example, always has a hard deadline [2]. In order to schedule the tasks of the system, all parameters including temporal and resource parameters for task preemption should be known well in advance [3].…”