Architectural Invariants and Correctness of IoT-based Systems
This work addresses critical reliability issues for IoT applications in areas like smart manufacturing and health monitoring, though it appears incremental as it builds on existing formal methods.
The authors tackled the problem of ensuring correctness and reliability in IoT-based systems by proposing a generic formal modeling framework based on architectural invariants, which they implemented and experimented with using Event-B.
Internet of Things applications impact more and more industrial areas such as smart manufacturing, smart health monitoring and home automation; physical objects or devices equipped with sensors and actuators are interconnected and then controlled with software applications. Ensuring the correct construction, the well-functioning and the reliability of these applications constitute important issues for some of these applications which can be critical in case of dysfunction. We propose on the basis of the formal model of their common architectural properties, a generic framework for the formal modelling of IoT-based applications, the rigorous analysis of their consistency properties, their rigorous construction and evolution. Specific properties can be gradually added and checked. The proposed framework is then implemented and experimented using Event-B. We exploit the observation that the main requirements of the IoT-based physical architectures and control software are common to all IoT-based applications; this leaded us to the definition of the generic formal model together with invariant architectural and consistency properties. The proposed approach is generic, extensible, and can be profitably adapted to more general hybrid or cyber-physical systems. Our current implementation is independent of the formal model, it can be achieved in various other formal analysis environments.