IoT Ambient Assisted Living: Scalable Analytics Architecture and Flexible Process
Guillaume Gingras, Mehdi Adda, Abdenour Bouzouane, Hussein Ibrahim, Clémence Dallaire · Procedia Computer Science · 2020
With the recent advances in IoT, Ambient Assisted Living (AAL) became an active field that attempts to assist individuals in their Activities of Daily Living (ADL). One researched venue deriving from these advances is how the technology and analytics could benefit the prevention and treatment of chronic diseases in the escalating number of elderly people experiencing health issues. Many architectures are proposed in the literature, but they lack modularity and flexibility for different types of sensors and do not have a way of selecting the appropriate algorithms to perform a given task. In this paper, we propose a four layered and highly modular architecture for health analytics of elderly people. Moreover, we propose a novel automated process for selecting the appropriate algorithms for a task at hand. In the final analysis, we evaluate the approach by implementing part of the architecture on fog nodes and the cloud. Finally, we deploy affordable consumer grade sensors in an apartment in order to move toward the use of the system proposed.