Abstract
Context: A fog computing architecture that is geographically distributed and to which a variety of heterogeneous devices are ubiquitously connected at the end of a network in order to provide collaboratively variable and flexible communication, computation, and storage services. Fog computing has many advantages and it is suited for the applications whereby real-time, high response time, and low latency are of the utmost importance, especially healthcare applications. Objectives: The aim of this study was to present a systematic literature review of the technologies for fog computing in the healthcare IoT systems field and analyze the previous. Providing motivation, limitations faced by researchers, and suggestions proposed to analysts for improving this essential research field. Methods: The investigations were systematically performed on fog computing in the healthcare field by all studies; furthermore, the four databases Web of Science (WoS), ScienceDirect, IEEE Xplore Digital Library, and Scopus from 2007 to 2017 were used to analyze their architecture, applications, and performance evaluation.
1. Introduction
A number of IoT services, such as computation resources, storage capabilities, heterogeneity, high processing, and others that brought a technological revolution, are provided by cloud computing. The cloud provides the virtualization of computing resources at various levels [1]. Almost all the human life domains have adopted cloud computing [2]. However, cloud computing has drawbacks in terms of high delays which have an adverse effect on the IoT tasks that require a real-time response. Furthermore, it does not match industrial control systems which require a lowdelay response time [1]. In 2012, Cisco announced an infrastructure paradigm called fog computing, which is a new computing concept, so as to tackle the limitations of cloud computing [3].
7. Learned lessons
Numerous lessons related to fog computing have been learned. Fog computing, without a doubt, decreased latency in contrast to cloud computing in healthcare IoT systems. Researchers show that simulation and experimental proportions provide many advantages such as distributed processing, privacy, security, scalability, fault tolerance, and low latency. These advantages are beneficial for vital signs patient monitoring systems, which demand substantial reliability, mobility, context awareness, and processing in realtime.
