Smart Campus: Model Driven Security for Systems of Systems

Socio-technical systems underpinning smart and connected cities are good examples of systems of systems – where each constituent system act and behave independents but these systems are connected with each other and provide each other different services. Security is one of the key aspects of systems of systems, e.g., smart city systems or smart campus systems. It is extremely complex and challenging undertaking to model and analyse a secure systems of systems. Model Driven Engineering (MDE) is expected to bring much needed support to bring effectiveness and efficiency in terms of automating some of the tedious and error-prone tasks. To support a MDE approach to secure smart cities systems, we need appropriate meta-models and suitable tools that can convert the meta-models in graphical editors that can support the modelling and analysis of secure smart cities applications. In collaboration with researchers from France, we have been working on a project aimed at developing graphical editors and model cases of secure systems in a smart cities, or smart campus. Apart from senior researchers from Adelaide and France, Zi Yang Pang has been working on summer scholarship on this project. The project has the following goals and scope and we are hoping to apply the meta-model language for secure smart cities systems very soon and afterwards we plan to extend the tools to provide end-to-end support. 

Systems-of-Systems (SoS) Some examples of Systems-of-Systems are internet of things, smart grids, smart cities, smart waster management systems, smart waste management system, and smart transportation. As our collaborator Jamel wrote that “If security were taken into consideration only at the end of the SoS development life-cycle, all posterior modifications and analysis would be very expensive. That is why security has to be taken into account as soon as possible in the design of an SoS. For this, the proposition of a new modelling language (based on UML, MoDELO, SysML, UMLSec,…) to model, simulate and predict SoS Security is in progress. This language comprise three components: an abstract syntax/MetaModel (set of concepts and relations), a graphical concrete syntax (set of predefined icons) and semantics/meaning (inherited from Multi-Agent Systems). A remaining challenge is the definition of software tools related to these components. One efficient approach to do that is Model Driven Engineering (MDE) which is a software engineering technique enabling the definition of modeling languages and (semi)automatical generation of tools, such as graphical or textual editors, code generators and model transformations.