مقاله انگلیسی رایگان در مورد تکنولوژی های توانمندسازی برای محاسبات مه در سیستم های خدمات درمانی – الزویر 2019

1- Introduction

2- Research methodology

3- Fog computing in healthcare IoT systems

4- Related works

5- Limitations

6- Discussion and open issues

7- Learned lessons

8- Conclusion

References

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. Results: A total of 99 articles were selected on fog computing in healthcare applications with deferent methods and techniques depending on our inclusion and exclusion criteria. The taxonomy results were divided into three major classes; frameworks and models, systems (implemented or architecture), review and survey. Discussion: Fog computing is considered suitable for the applications that require real-time, low latency, and high response time, especially in healthcare applications. All these studies demonstrate that resource sharing provides low latency, better scalability, distributed processing, better security, fault tolerance, and privacy in order to present better fog infrastructure. Learned lessons: numerous lessons related to fog computing. Fog computing without a doubt decreased latency in contrast to cloud computing. Researchers show that simulation and experimental proportions ensure substantial reductions of latency is provided. Which it is very important for healthcare IoT systems due to real-time requirements. Conclusion: Research domains on fog computing in healthcare applications differ, yet they are equally important for the most parts. We conclude that this review will help accentuating research capabilities and consequently expanding and making extra research domains.

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 low-delay 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]. They asserted that fog computing is applicable at three networking levels: (1) the collection of data from the devices in the edge (sensors, vehicles, roadways, and ships); (2) multiple devices connecting to a network and sending all the data; (3) the collected data from the devices should be processed in less than a second along with decision making [4]. The term fog computing shifts capabilities of the cloud near to the end user, and provides storage, computation, and communication to edge devices, which facilitate and enhance mobility, privacy, security, low latency, and network bandwidth so that fog computing can perfectly match latency-sensitive or realtime applications [5]. On the one hand, fog computing infrastructure consists of plenty of fog nodes, edge device networks, and even virtualized data centers or IoT devices that are connected to these nodes [6]. These are connected to the cloud for the purpose of implementing large storage and rich computing [7]. The distribution of functions between the cloud and the fog nodes is considered a crucial factor [8]. Millisecond to sub-second latency offered by fog, even faster than real-time interaction, supports multitenancy and performs better in low-latency applications [9]. The concept of fog computing has been designed to satisfy the applications that require low latency with a real-time response such as healthcare IoT systems.