Non-intrusive In-situ Requirements Monitoring for Embedded Systems

Abstract

Accounting for all operating conditions of a system at the design stage is typically infeasible for complex systems. Monitoring and verifying system requirements at runtime enable a system to continuously and introspectively ensure the system is operating correctly in the presence of dynamic execution scenarios. In this dissertation, we present a requirements-driven methodology enabling efficient system-level runtime monitoring of embedded systems. The presented methodology constructs a hierarchical runtime monitoring graph from system requirements specified using multiple UML sequence diagrams, which are already commonly used in software development, and state-based hardware models, which are common in hardware design. The requirements models for both software and hardware components can then be integrated to create a system-level requirements model that will be used at runtime to additionally verify the interactions between hardware and software components. Non-intrusive, on-chip hardware dynamically monitors the system-level execution and communication, verifies the execution and the communication adhere to the requirements model, and in the event of a failure provides detailed information that can be analyzed to determine the root cause. Using case studies of a collision-avoidance system and smart connect pacemaker prototypes, we analyze the relationship between event coverage, detection rate, detection latency, root cause analysis, and hardware requirements.