Addressing Non-Functional Requirements of Adaptive IoT Systems: A Model-Driven Approach

Non-functional requirements (NFR) of IoT systems increase the complexity of system development. The success of such systems also largely depends on dealing with NFRs correctly. However, interdependencies among NFRs often introduce conflicts. These conflicts impede implementing the system with all specified NFRs. Furthermore, the heterogeneous nature of IoT systems makes it critical to incorporate NFRs in the early stages of software development. This PhD thesis proposes a model-driven requirements engineering procedure to address different NFRs of adaptive IoT systems. This approach will incorporate non-functional requirements at different levels of abstraction with model-driven techniques to minimize conflicts among elicited NFRs. We are extending use case models, soft goal models, and behavioural models to elicit, analyze, and specify interoperability, scalability, availability, and context-awareness of IoT systems. As interoperability and context-awareness are two NFRs that affect the adaptiveness of IoT systems most, we addressed these two NFRs first. Availability and scalability NFRs will be incorporated as this thesis progresses.