Article Preview
Top1. Introduction
The internet of things (IoT) technology has remarkably facilitated several day-to-day routines. IoT has attracted a lot of attention among academia and industries in recent years. The primary aim of IoT is to facilitate the interaction between humans and the things surrounding them (Fazeldehkordi, Owe, & Noll, 2019). Things refer to objects that are used in our daily life ranging from simple things such as smart bulbs to more sophisticated ones such as heart disease detectors (Hussain, Hussain, Hassan, & Hossain, 2019). Several felids of the modern society but not limited to healthcare, automobile, industrial appliances, sports, and environment have witnessed many useful IoT applications (Fazeldehkordi et al., 2019)(Al-garadi, Mohamed, Al-ali, Du, & Guizani, 2018). The European IoT Research Cluster(IERC) defined IoT as a global network of physical and virtual objects, which have an identity, physical attributes, and virtual personalities, that communicate through an advanced protocol based on standardized self-configuration capabilities and use smart interfaces which are connected to computer networks (Djenna & Saidouni, 2019). In the IoT environment, things mainly collect data and receive data, while the processing is done in fog nodes that are closer to the sensors or remotely in the Cloud (Alhazmi & Aloufi, 2019). There are several IoT architectures have been proposed to illustrate the different layers of IoT stack (Adat & Gupta, 2018; Fremantle, 2015; Guo, Ren, Zhang, Zhang, & Hu, 2017; Hsieh & Lai, 2011; Khan, Khan, Zaheer, & Khan, 2012; Li, Xuan, & Wen, 2011; Salman, Elhajj, Kayssi, & Chehab, 2016; Suarez et al., 2016; Tan, 2010; Wu, Lu, Ling, Sun, & Du, 2010; Zhou et al., 2013). However, the underlying IoT architecture is composed of three layers, and it is very abstract and lack of several details (Alshohoumi, Sarrab, AlHamadani, & Al-Abri, 2019): perception layer which includes different internet-based things (sensors) that are used for collecting data such as sensors which collect health-related data (e.g., vital signs), network layer which comprises different communication protocols and network devices for transmitting and processing the collected data, and application layer which is responsible for providing the end-user interface that can be utilized by IoT users (Djenna & Saidouni, 2019)(Alhazmi & Aloufi, 2019). Contrarily, the new IoT architectures, as reported in (Sood & Mahajan, 2019) (Alshohoumi et al., 2019), consist of several layers and introduces more details related to data analysis, processing, storage, and data security (Alhazmi & Aloufi, 2019).
H-IoT plays a significant role in improving health care through leveraging many critical IoT healthcare applications for monitoring patient’s health. For example, IoT can be used to identify and control the hypertension attack at an early stage. In H-IoT, IoT-based medical sensors can be used to predict the risk of hypertension attack. It can provide real-time notifications through the system to the users as well as the doctors to diagnose their health(Sood & Mahajan, 2019). H-IoT can help health professionals to monitor their patients’ health status and which can improve health care management and patients’ treatment through learning the collected data to extract useful insights into diseases (Macdermott, Kendrick, Idowu, Ashall, & Shi, 2019). The key functionality in H-IoT is real-time monitoring based on the collected data from patients through medical things(Macdermott et al., 2019). Although H-IoT promises socio-economic growth and significant impact in health-related wellness, it encounters several challenges related to the real-time monitoring (e.g., delay, latency, and response time), reliability, security of the collected data, and patients’ privacy (Macdermott et al., 2019)(Sood & Mahajan, 2019)(Djenna & Saidouni, 2019)(Alharam & El-Madany, 2017). This research proposes a fog-based framework for data classification into different fog virtual servers according to their criticality level so that it will help in preserving the privacy of the collected data and improving the response time while detecting the abnormal measurements.
The work in this paper is framed in different sections as follows: Section 2 presents the related work which includes IoT in healthcare, the role of fog-computing in H-IoT, and analysis of various types of data collected by IoT medical applications. Section 3 introduces the proposed approach. The discussion is presented in section 4. Section 5 concludes the paper with a future research direction.