Experimentation as a service over Internet of Things enabled smart citiesphilosophiae

Résumé

During the last decade, the great evolution of Wireless Sensor Network (WSN) and the continuous development and miniaturization of computers, have made possible the emergence of a new global paradigm: the Internet of Things (IoT). This new paradigm has removed from the equation the human-to-human or human-to-machine interactions, triggering the creation of a new myriad of applications without the need of human intervention. From home automation to service provision in cities and communities, such novel technologies represent an outstanding evolution for many traditional services. At the same time, the rapid increase of world population living in urban areas required the implementation of new services so as to keep current living standards. Therefore, this situation has led to an increased coexistence between novel IoT deployments and cities, boosting the development of new services based on such relationship, the so-called Smart City revolution. Furthermore, the scientific community has been benefited by the introduction of new technologies in scenarios that were not accessible before, such as cities and urban areas. From laboratories to the urban scenario, the time to develop new solutions for cities worldwide has been shortened, as the possibility of deploying new pilots in them has become a reality. There are several services where this evolution is more visible, like urban mobility, or the continuous environmental monitoring of certain parameters. Hereof, services regarding crowd management in cities are particularly relevant nowadays. Unexpected events, such as terrorist attacks or the recent emergency due to highly contagious diseases, has made crowd management an essential part of the Smart City. However, despite the benefits associated to IoT-based deployments in cities, they have brought to light new issues that were not faced before in controlled environments. Just to mention a few, we can highlight the difficulties associated to the introduction of constantly evolving technologies in cities; the use of different information data models, that hinders the adoption of successful services from other cities; or the increased difficulty while dealing with an infrastructure that is usually difficult to access, limiting the experimentation possibilities. In this context, the concept of Experimentation as a Service (EaaS) emerges as a solution to enable the experimentation over IoT-enabled Smart Cities, providing an interoperable framework to carry out the experiments in multiple IoT deployments. Throughout this dissertation, the author presents the contributions carried out on three main pillars, namely deployment, provision and experimentation, that underpin the design and implementation of an EaaS solution for the development of IoT-based Smart City services and applications. These contributions are: 1) a thorough description of the Smart City deployment which has been carried out in the city of Santander, including the discussion of the lessons learned elicited from the experience working with IoT devices in an urban environment; (2) the development and provision of EaaS models to enable third-parties such as citizens, service providers and the scientific community, with the means to access the Smart City IoT resources, through semantic-enabled federation with other international IoT testbeds; (3) the implementation of an experiment on top of the experimentation framework, the Crowd Counting and Location System (CCLS), a privacy-aware solution for crowd management in the Smart City. In a nutshell, the aforementioned contributions conform a holistic view of an IoT-enabled Smart City for the support of an EaaS model, from the perspective of three key-role players in its development: the deployment manager; the Smart City experimentation and service provider; and the researcher. Each of them related to the work presented in this dissertation.