IoT and Smart Cities: Modelling and Experimentation

Internet of Things (IoT) is a recent paradigm that envisions a near future, in which the objects of everyday life will communicate with one another and with the users, becoming an integral part of the Internet. The application of the IoT paradigm to an urban context is of particular interest, as it responds to the need to adopt ICT solutions in the city management, thus realizing the Smart City concept. Creating IoT and Smart City platforms poses many issues and challenges. Building suitable solutions that guarantee an interoperability of platform nodes and easy access, requires appropriate tools and approaches that allow to timely understand the effectiveness of solutions. This thesis investigates the above mentioned issues through two methodological approaches: mathematical modelling and experimenta- tion. On one hand, a mathematical model for multi-hop networks based on semi- Markov chains is presented, allowing to properly capture the behaviour of each node in the network while accounting for the dependencies among all links. On the other hand, a methodology for spatial downscaling of testbeds is proposed, implemented, and then exploited for experimental performance evaluation of proprietary but also standardised protocol solutions, considering smart lighting and smart building scenarios. The proposed downscaling procedure allows to create an indoor well-accessible testbed, such that experimentation conditions and performance on this testbed closely match the typical operating conditions and performance where the final solutions are expected to be deployed.