A Semantic Implementation for Data Management in an Open Building Automation Platform.
Autor: Jie Song
Director: Asunción Santamaría
We are living in a world full of embedded systems, which have formed an essential part of
our daily lives. With the development of the Internet of Things, more and more Things such as sensors, actuators and smart devices are connected to collect and exchange data. However, the use of different communication protocols leads to high cost and long development time for the integration of different technologies in the same system. As a possible solution, the concept "Web of Things" (WoT) emerges with the idea of using Web as the application-layer for the IoT, drawing the attention of several researchers.
One of the issues that the IoT is facing is "data silos", i.e the isolation of data. This issue appears because the meaning of the data is not clear explained. The Web of Things dedicates to improve the semantic interoperability of the IoT, which requires the simultaneous transmission of the meaning with the data, i.e. semantics.
This thesis focus on the implementation of a semantic data management system for an open IoT platform. The platform dedicates to solve the integration problem among different communication technologies: capillary networks and traditional home automation protocols.
Besides, the platform defines a common vocabulary for providing semantic description to Things, including the functionalities, representation of the generated data and related concepts such as location or user authorization. In addition, the platform implements a Web API, which allows users or software programs to interact with Things through a standard
On the other hand, the involvement with real devices into the Web requires higher level of security and privacy protection, and the "multi-client, multi-user" paradigm demands more precise approach for deciding the authority of each user. To face this challenge, a new concept "Private Knowledge Graph" (PKG) has been proposed which represents a knowledge graph generated for the use of a specific agent. Besides, a fine-grained access control mechanism is designed on the basis Web Services and PKGs.
Finally, a theoretical case study is presented to explain the working process of the semantic data management system. Furthermore, a practical case is provided to validate the performance of the designed system, verifying that it facilitates the integration of Things and applications.