|
Descripción
|
|
|---|---|
| The vision of a world populated by smart objects has been a constant since the dawn of contemporary telecommunication in the late nineteenth century. However, it was Mark Weiser who, in the early 1990s, established the fundamentals of Pervasive Computing, which were essential for the foundation of a new concept of object interconnection called the Internet of Things (IoT). This recent paradigm has extended the capillarity of the original Internet integrating devices with sensorial capabilities in close relation with the real scenarios, to implement applications for Smart Cities, Industry 4.0, Connected Car or Personal Health. An ultimate challenge for the spreading of IoT is the removal of the current high technological fragmentation that affects aspects such as hardware platforms, programming models and services, semantics and security. The doctoral thesis describes a holistic solution that simplifies the complexity of the Internet of Things through contributions on two connected paradigms: Web of Things and Model-Driven Development Methodologies. Firstly, the thesis proposes a reference architecture called Web of Things Open Platform (WoTOP), which aims at orchestrating interoperability between different technologies and standards of the Internet of Things, to create an enriched ecosystem of sensorial and logical resources. This architecture also includes an application domain layer that defines an API to enable the access to those resources through web technologies. WoTOP is based on an open architecture so it can be extended and modified depending on the needs of specific projects. The thesis also includes a light version of WoTOP, ready to be embedded in mobile devices to serve, for example, as a framework to develop applications involving mobile-connectable sensors (e.g. wearables for personal health solutions). Secondly, in order to explore the universalization of IoT technologies, the proposed framework provides a Model-Driven Resource-Oriented (MDRO) methodology for the development of smart objects networks. MDRO is designed to be used by users without specific knowledge in modeling and programming languages. The main advantage of using MDRO lies in the support and automation during the development of services for smart spaces. These characteristics are achieved through the semantic verification mechanisms of the models as well as their automated transformation into program code. The MDRO methodology is complemented by a development environment which is based on an ad-hoc modeling language designed as an UML extension, the Smart Space Modeling Language (SSML). For validation, the performance of WoTOP implementation has been evaluated through a simulated environment based on a Queueing Theory M/G/1 model, which demonstrates the scalability of the platform at the same time that serves as a tool to size real deployments. In particular, a set of services for smart buildings considering ambient, localization, parking and resources management features have been designed using the tool. Additionally, the MDRO methodology together with a functional version of its associated development environment has been used to deliver several real services in the Experience Lab of Future Spaces, in particular of energy consumption, ambient control and security. These services were deployed on a smart space infrastructure based on the WoTOP architecture. | |
|
Internacional
|
No |
|
ISBN
|
|
|
Tipo de Tesis
|
Doctoral |
|
Calificación
|
Sobresaliente cum laude |
|
Fecha
|
06/09/2017 |