Modeling & Verification of an Adaptive IoT System using Uppaal

Design time verification of Internet Of Things (IoT) systems helps in detection of errors at an early stage. This study focuses on the modeling and verifying IoT systems that are self-adaptive. Self Adaptive Systems (SAS) adapt autonomously to changing environment and internal dynamics to achieve required functionalities. The adopted case study is the Smart Home Automation System. A self adaptive smart home system is modelled using Uppaal, an integrated tool used for modeling and verification of real-time systems. In Uppaal, systems are modelled using a network of timed automata and verified using statements in Linear Temporal Logic. Two subsystems of smart home, fire detection and automated lighting system are modelled and its properties are verified to ensure correctness of the system. A typical SAS has a feedback loop consisting of MAPE-K (M-Monitor, A-Analyze, P-Plan, E-Execute, K-Knowledge) components, for keeping track of changes in the system, automatically update the knowledge base and execute corrective actions to cope up with the changes in the system. Self adaptive nature of fire detection subsystem in the case study enables the system to automatically switch to inputs from temperature sensors for fire detection in case of failure/ malfunctioning smoke sensor, thus preventing system failure. Automated lighting system changes the light intensity depending on changing activity of a person in a particular area. Modelling and verification of SAS is essential for avoiding run-time failures of safety-critical systems.