A fog-computing architecture for preventive healthcare and assisted living in smart ambients

In the last years, the emergence of pervasive connected devices and the development of the cloud computing paradigms brought a revolution in health-care and industrial applications. Cloud and Internet of Things (IoT) exploit large scale service providers to vastly reduce costs and gather Big Data. However, cloud-based services still face various issues related to: high bandwidth requirements, unpredictable delays, and security and safety concerns. These issues are critical to health-care and Active and Assisted Living (AAL) where a correct and timely reaction can result in saving a life or drastically reducing a disability (e.g. after a stroke). In this scenario, we present a flexible multi-level architecture using the fog approach, a computing paradigm in which heterogeneous devices at the edge of the network collect data, compute a task with minimal latency, and produce physical actions meaningful for the user, leveraging upon context and location awareness. In this paper, we envision also an edge node built upon Field-Programmable Gate Array (FPGA) technology. The hardware of a FPGA node can be reconfigured to produce maximum performance in tasks, to guarantee a minimal delay, or the capacity to scale on the number of devices connected, with a minimal power consumption. We present two case studies for assistive smart ambients and health applications designed on our Fog architecture.