Architectural Model for Collaboration in the Internet of Things; A Fog Computing Based Approach

摘要

Through sensors, actuators and other Internet-connected devices, applications and services are becoming able to perceive and react on the real world. Seamlessly integrating people, and devices is no longer a futuristic idea. Converging the physical world with the human-made realm into one network is rather a present and promising approach called The Internet of Things (IoT).;A closer look at the phenomenon of IoT reveals many problems. The current trends are focusing on Cloud-centric approaches to deal with the heterogeneity and the scale of this network. The blessing of the Cloud computing becomes, however, a burden on latency-sensitive applications, which require processing and storage mechanisms in their proximity to meet low-latency, location and better context-awareness requirements. In addition to mobility support and high geographical distribution requirements. Fog computing is a new concept that focuses on extending the Cloud paradigm to the edge of the Internet of Things, via providing communication, computing, and access management support.;This research project foresees and is driven by the promising opportunities of the concept behind Fog computing. In this thesis, we leverage this new concept by delivering a Collaboration Architecture for the Fog computing. This architecture constitutes a referential model to better design and to implement Fog platforms. It powers the freedom of abstraction to make development and deployment at the Fog nodes easier and more efficient. Moreover, it provides a nest where IoT-connected objects can interact and collaborate. To this end, we introduce expressive mechanisms to define and abstract objects, data analytics, and services. To leverage Fog nodes with dynamic services and service-based collaboration, we propose the concept of Operation: a formal way to dynamically generate new services through mechanisms such as aggregation, composition, and transformation. Finally, we deliver a comprehensive study and a collaboration-oriented access control model for the proposed architecture.