Internet of Things (Iot) For Next-Generation Smart Systems: A Review of Current Challenges, Future Trends and Prospects For Emerging 5G-Iot Scenarios

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/338874911
Internet of Things (IoT) For Next-Generation Smart Systems: A Review of

Current Challenges, Future Trends and Prospects for Emerging 5G-IoT
Scenarios
Article in IEEE Access · January 2020

DOI: 10.1109/ACCESS.2020.2970118
CITATIONS READS
793 6,761
5 authors, including:
Bilal A Khawaja Farah Sabir

Islamic University of Madinah | National University of Sciences and Technology NED University of Engineering and Technology, Karachi
74 PUBLICATIONS 1,782 CITATIONS 2 PUBLICATIONS 806 CITATIONS
SEE PROFILE SEE PROFILE
Sameer Qazi Muhammed Mustaqim

PAF Karachi Institute of Economics & Technology National University of Sciences and Technology
52 PUBLICATIONS 1,206 CITATIONS 29 PUBLICATIONS 1,215 CITATIONS
SEE PROFILE SEE PROFILE
All content following this page was uploaded by Bilal A Khawaja on 04 February 2020.
The user has requested enhancement of the downloaded file.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2020.2970118, IEEE Access
Date of publication xxxx 00, 0000, date of current version xxxx 00, 0000.
Digital Object Identifier 10.1109/ACCESS.2017.Doi Number
Internet of Things (IoT) For Next-Generation

Smart Systems: A Review of Current
Challenges, Future Trends and Prospects for
Emerging 5G-IoT Scenarios
Kinza Shafique1, Bilal A. Khawaja2, 3, Senior Member, IEEE, Farah Sabir4, Sameer Qazi4,
Member, IEEE, and Muhammad Mustaqim3
1
Department of Electrical Engineering, DHA Suffa University, Karachi, Pakistan, (e-mail: [email protected]).
2
Faculty of Engineering, Department of Electrical Engineering, Islamic University of Madinah, P.O. BOX 170 Madinah, 41411, Saudi Arabia (e-mail:
[email protected]).
3
Department of Electronic and Power Engineering (EPE), PN-Engineering College (PNEC), National University of Sciences and Technology
(NUST), Karachi, Pakistan, (e-mail: [email protected], [email protected]).
4
College of Engineering, PAF Karachi Institute of Economics and Technology, Karachi 75190, Pakistan, (e-mail: [email protected],
[email protected]).
Corresponding authors: Kinza Shafique, Bilal A. Khawaja (e-mail: [email protected], [email protected]).
This work was supported in part by the Deanship of Scientific Research, Islamic University of Madinah, Saudi Arabia, under the Research Grant No. 29/40.
ABSTRACT The Internet of Things (IoT)-centric concepts like augmented reality, high-resolution video
streaming, self-driven cars, smart environment, e-health care, etc have a ubiquitous presence now. These
applications require higher data-rates, large bandwidth, increased capacity, low latency and high throughput.
In light of these emerging concepts, IoT has revolutionized the world by providing seamless connectivity
between heterogeneous networks (HetNets). The eventual aim of IoT is to introduce the plug and play
technology providing the end-user, ease of operation, remotely access control and configurability. This paper
presents the IoT technology from a bird’s eye view covering its statistical/architectural trends, use cases,
challenges and future prospects. The paper also presents a detailed and extensive overview of the emerging
5G-IoT scenario. Fifth Generation (5G) cellular networks provide key enabling technologies for ubiquitous
deployment of the IoT technology. These include carrier aggregation, multiple-input multiple-output
(MIMO), massive-MIMO (M-MIMO), coordinated multipoint processing (CoMP), device-to-device (D2D)
communications, centralized radio access network (CRAN), software-defined wireless sensor networking
(SD-WSN), network function virtualization (NFV) and cognitive radios (CRs). This paper presents an
exhaustive review for these key enabling technologies and also discusses the new emerging use cases of 5G-
IoT driven by the advances in artificial intelligence, machine and deep learning, ongoing 5G initiatives,
quality of service (QoS) requirements in 5G and its standardization issues. Finally, the paper discusses
challenges in the implementation of 5G-IoT due to high data-rates requiring both cloud-based platforms and
IoT devices based edge computing.
INDEX TERMS Internet of things (IoT), 5G, Carrier Aggregation, CoMP, CRAN, CRs, HetNets, MIMO,
M-MIMO, NFV, SD-WSN, QoS.
I. INTRODUCTION back to the 1980s with the idea of ubiquitous computing

Momentous developments in wireless sensor networks, whose objective was to embed technology in every day’s life
telecommunications and informatics have paved the [3]. Currently, the IoT is envisaged both at the individual and
realization of pervasive intelligence [1]- [2] which envisions professional level. For an individual, IoT plays a pivotal role
the future Internet of things (IoT). The genesis of IoT goes in enhancing living standards in the form of e-health, smart
2169-3536 © 2017 IEEE. Translations and content mining are permitted for academic research only.
VOLUME XX, 2017 Personal use is also permitted, but republication/redistribution requires IEEE permission. 1
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
home and smart learning. For a professional, IoT finds its increase of 67% [10]. IoT is one of the emerging trends of the
application in automation, smart supply chain and current decade. Also, as per Gartner’s IT Hype cycle [11], it is
transportation, remote monitoring and logistics. forecasted in 2011 that IoT will take 5-10 years for market
According to the reports from Ericsson [4], about 28 billion adoption. Hence, the US $1.7 trillion is also expected as
smart devices will be connected across the world by 2021. expenditure on IoT by 2020 as per International Data
Furthermore, machine-to-machine (M2M) communication is Corporation (IDC) [9].
employed in more than 15 billion devices [4]. Further research
envisaged that around 6-7 devices will be carried per person A. IoT VISION – ANYTIME, ANYWHERE, ANYTHING
by 2020[5]. The term IoT was first coined by Kevin Ashton in 1999 with
A vision of the all-communicating world is stated as the reference to the supply chain management [12]. The concept
system that provides per area data volume to be increased by of IoT revolves around the word “smartness” – “an ability to
1000 times, the number of connected devices and the user independently obtain and apply knowledge”[13]. Therefore,
data-rate is to be increased by 10 to 100 times. Furthermore, IoT refers to the "things or devices and sensors" that are smart,
the extended battery life up to 10 times for massive machine uniquely addressable based on their communication protocols,
communication devices and end-to-end latency to be reduced and are adaptable and autonomous with inherent security [14].
by 5 times [6]. Hence, recent trends show researchers huge Atzoriet al.[3] have characterized IoT in three visions. Internet
interests towards the amalgam of various technologies such as Oriented–the vision focusses on connectivity between the
integration of sensors and embedded systems with cyber- objects; Things Oriented- the vision focusses on generic
physical systems (CPS), device-to-device communications objects; and Knowledge Oriented – the vision focusses on how
(D2D) and 5G wireless systems with IoT as a centre. to represent, store and organize information. These visions
Currently, new business models set for IoT implementation paved towards International Telecommunication Union (ITU)
requires massive connectivity, high privacy and security, vision of IoT, which defines it as “from anytime, anyplace
complete coverage, ultra-high reliability and ultra-low connectivity for anyone; we will now have the connectivity for
latency. The trending 5G enabled IoT encompasses increased anything” [15]. In a nutshell, the ultimate objective is “to plug
data-rates, better coverage and high throughput hence and play smart objects”.
providing solutions to business models and enabling IoT to
robots, actuators and drones [7]. B. ARCHITECTURAL TRENDS AND OBJECTIVES
So, in this paper, the author’s present a comparative analysis There are three components that form the basis of IoT
and detailed review of the IoT technology covering its architecture:
statistical/architectural trends, use cases, challenges and future (1) Hardware: It comprises of sensor nodes, its embedded
prospects. The paper also presents a detailed and extensive communication and interfacing circuitry.
overview of emerging 5G-IoT scenarios, ongoing 5G (2) Middleware: It comprises of data storage, analysis and
initiatives all over the globe, QoS requirements in 5G and its handling resources.
standardization issues and the impact of the integration of 5G (3) Presentation layer: It comprises of efficient visualization
with IoT and artificial intelligence (AI). tools that are compatible with various platforms for different
The rest of the article is organized as follows: Section II applications and present the data to end-user in an
presents a detailed background and overview on IoT – its understandable form.
vision, architectural trends and objectives, opportunities and The parameters affecting the architecture of IoT are
prospects, projects and research trends. Section III, gives a manifold. Hence, current research efforts have been made to
detailed overview of 5G enabled IoT where its features, devise the most optimized architecture that handles network
architectural trends, quality of service and standardization issues such as scalability, security, addressability, and efficient
issues are discussed in detail. Section IV presents the emerging energy utilization.
concept of 5G Intelligent IoT with reference to the AI-based As for the future, the number of devices connected to the
systems. Section V gives an overview of challenging prospects network will rise. Hence, the architecture of IoT must cater to
of 5G-IoT, and finally, Section VI concludes the paper. it. Scalability, energy consumption, and addressing issues are
all considered as challenges for successful deployment of IoT.
II. BACKGROUND, MOTIVATION AND OVERVIEW Research is carried out in solving the scalability issues by
It is estimated that by 2025, the internet nodes may reside developing various multi-hop routing protocols covering a
in every single object hence causing the number of devices larger area and are self-adapting. These fall into three
connected to the internet to rise [8]. According to Cisco, there domains: 1) Data-centric, 2) Location-based and 3)
will be 500 billion devices connected to the internet by the year hierarchical [16].
2030. Similarly, Telefonica predicted in 2013 that 90% of the The energy consumption issues are addressed by using
cars will be connected to the internet by 2020 [9]. However, energy harvesting techniques [17], [18], devising energy-
the survey in 2015 suggests that more than 250 million efficient MAC protocols [19] and cross-layer protocols [20].
vehicles will be connected globally by 2020, which will be an On a large scale, the authors in [3] have suggested the
VOLUME XX, 2017 3
deployment of a combination of internet protocol (IPv6) and address the energy limitation issue by devising the hierarchal
low-power wireless personal area networks (6LoWPAN) [21]. architecture which comprises of three layers: Sensing and
The 6LoWPAN technology targets at integrating the low control layer (SCL), information processing layer (IPL) and
power sensor nodes working at IEEE 802.15.4 protocol into application layer (AL). This not only defines the energy
IPv6 networks comprising of 10128 addresses. The concept of efficiency but also increases the hardware resource utilization
Mobile IP was presented in [22] rather than the use of home of both the SCL and the IPL layers. The main feature of the
location register (HLR) and visitor location register (VLR) as architecture is based on exchanging the energy-related
it does not require any centralized server. On a smaller scale, information between the SCL and the IPL, hence controlling
European Coordinated Action aimed at redefining the RFID the sleep time interval of the sensing nodes. The SCL
standards, so that RFID applications can be shifted to IoT [23]. comprises of sensor nodes (SNs) which are controlled by
Handling a huge amount of big IoT data from all the nodes gateway nodes (eGNs) which in turn is controlled by a base
of a network is a tedious task. Also, the energy efficiency of station (eNode). The SNs work in two modes: periodic mode
the data centres must be considered. Hence, to cater for these and trigger mode suitable for periodic and critical events,
issues, artificial intelligence techniques, novel fusion respectively.
algorithms, state-of-the-art temporal machine learning On the other hand, the eGNs control the sleep time interval
methods, and neural networks must be deployed for automated of the SNs depending upon the previous usage history of the
decision making and energy efficiency [16]. SNs and the quality of information available at the nodes.
Security and privacy are one of the major open issues in IoT Another important factor in this regard is the conflict factor,
architecture as end-user data should be protected from which is a measure of duplicate information by different
eavesdropping and interference. Data should be authenticated, sensors, battery level and coefficient of variation. The
and its integrity must be maintained at the user end. Various coefficient of variation is a measure of variation between
cryptographic algorithms are proposed for data authentication, current and previously sensed values. If there is no variation
but it possesses serious energy and bandwidth consumption in the information received in the past and present, then the
issues. Hence, key cryptographic schemes are proposed and sleep interval of sensors is prolonged, and otherwise, it is
presented [24]-[25]. The security and privacy of an IoT reduced. The eNodes setups the communication between the
network are also challenged whenever a node enters in or SNs and the cloud resources, which to be allocated as per
whenever the applications running on nodes are needed to be requirement. The eNodes also assign the SNs to the generator
installed or updated. In this context, remote wireless nodes depending upon their battery level and distance. The
reprogramming protocol is proposed [3]. This protocol allows information processing layer stores, processes, and analyzes
the node to look for any malicious attack during installation the extracted data using a cloud-computing platform through
and to verify each code. These are mostly based on a standard energy-efficient resource allocator. The resource allocator is
protocol called Deluge [26]. used for processing data in an efficient manner by allocating
Another challenge facing IoT successful deployment is the the hardware resources according to the needs of Sensing and
lack of a universal platform, protocol, and a programming Control Layer and the information analyzer. The information
language. Today, all connected devices follow a different set analyzer calculates the level of information extracted from the
of protocols and platforms. The need of the time is to have data and decides the sleep interval of the SNs. The application-
collaboration among the connecting devices. For this, large dependent information converter then translates the
enterprises such as LG, Samsung and Philips etc. should join information in the human-readable form and store it in the
hands and make a consortium for the development of universal storage media. The AL provides services to the end-users by
coding language and platform. The proposed solution can using visualization tools. Hence, energy efficiency is achieved
solve the compatibility issues of IoT up to a significant extent by controlling the sleep time interval of the SNs using the
[13]. eGNs and the BS in sensing and control layer. It is possible by
The IoT must be suitable for all the applications having allocating the hardware resources accordingly using the
elastic traffic and in-elastic traffic [16]. The former (elastic energy-efficient resource allocator in the IPL. The hierarchal
traffic) focusses on applications that are delay tolerant, which architecture is also compared with other architectures such as
includes monitoring applications; network scheduling etc. self-organized things (SoT) [28], energy-efficient index tree
While the latter (in-elastic) requires a prompt response from (EGF-tree) [29], energy-efficiency hierarchical clustering
the network, such as video streaming. Hence, IoT must satisfy index tree (ECH-tree) [30] and object group localization
both the traffics having different key performance indicators (OGL) [31]. After a thorough comparison, itis then concluded
with the high QoS. that the proposed architecture shows a functionality edge over
Keeping in mind the above open issues, a set of starting others.
guidelines are provided in the IoT architecture reference For efficient data handling, a cloud-based architecture is
model (ARM) [3]. presented by Gubbi et al. [16] and Zhou et al. [32],
Similarly, various IoT architectures have been proposed in respectively.
the literature addressing various issues. Kaur et al. [27],
VOLUME XX, 2017 3
Another architecture combining the third-generation (3G) In general, companies are going to take an immediate
networks and power line communications (PLC), also referred investment or a wait-and-see approach to investment based on
to as 3G-PLC, is presented in [33] to address the scalability the maturity level of the specific IoT technologies [35]. During
issues. Similarly, for large scale networks, self-configuring the initial deployment phase, the margin should be kept for
peer-to-peer based architecture is also presented [34]. adapting to changes. For this, open software and hardware-
based IoT solutions should be proposed [12]. In this regards,
C. IoT OPPORTUNITIES AND PROSPECTS one approach of cost-cutting is to use smartphones serving as
IoT offers many business opportunities, which allow IoT nodes [16].
companies to build new business strategies and models to
implement the concept. Not only business opportunities, but D. IoT PROJECTS
also the realization of efficient and resourceful research Due to the extensive research and interest in the IoT domain
opportunities to scholars and investigators of multi- since the concept was first introduced in 1980s, various
disciplinary fields. Hence, it combines business studies, countries such as Canada, China, Brazil and UAE have
engineering skills, science and humanities all under one implemented the concept of the smart-cities, smart cranes,
umbrella. Also, IoT transforms the world into a smart world, smart flood warning system and smart-grid, respectively [12].
where everything is easily accessible in less time and effort, as These projects are successfully implemented and are running
depicted in Fig. 1 (a-b), respectively. at the Rio Operations Centre, SK Solution, Yellow River
Conservancy Commission and BC Hydro, respectively [12].
China had also invested $100 million in studying the IoT
industrial standards and technologies in Shanghai as a part of
its 12th 5-year plan[16],[35]. Substantial efforts are underway
to merge cross-domain research activities spanning machine-
to-machine (M2M) communication, wireless sensor network
(WSN) and RFID into unified IoT framework [16]. The
European Research Cluster on IoT [3], [16], [36], an
organization of the European Union, is currently funding 33
IoT projects [36]. The focus of the organization is to develop
IoT architectures, technological and knowledge
interoperability with security, reliability and scalability [3],
[16], [36-37]. The reason behind the interest of the first world
countries in IoT trends and developments is its prevailed
benefits, as shown in Fig. 1(b). The outlined benefits hence
proved advantageous for the country’s economic
(a) sustainability, growth, urbanization, infrastructure,
employment rate, citizens’ health and services.
Utilization
Improvement
E. IoT - A CORPORATE SECTOR VISION
The IoT deployments also enabled the enterprises to meet
their business needs [9]. Microsoft proposed a cloud-based
New Minimum architecture in which the nodes send their data to the cloud
Enhanced User gateway for processing. Similarly, the IoT based video
Services
IoT Interaction solution is implemented by Banco de Cordoba [9]. With this
implementation, the trainers are not required to visit fields;
hence, increased their productivity level. The designed
solution also enables security and marketing efficiency. Also,
Daimler employed IoT architecture and used IBM services to
launch their smart cars concept, which is popularly referred to
Cost Proactive as Car2go [38].
Reduction Maintenance
(b)
Figure1: Block diagram showing (a) IoT based smart applications

and (b) IoT benefits as the cutting edge technology
VOLUME XX, 2017 3
10.0
9.0 Mobile Subscriptions (Billion)
8.2
Mobile Subscriptions (Billion)
8.0
7.0
6.0
4.8
5.0
4.0
3.0 2.5
2.0
1.0
0.0
2G (Voice Services) 3G (Voice / Data services) 4G (Broadband Internet
Services)
Figure 2: Mobile devices subscriptions with respect to different wireless technologies
In a nutshell, the technical specifications currently become

F. IoT - RECENT TRENDS part of published literature on IoT enable smart environment
Weiser et al.[26], [39], introduced the concept of ubiquitous is summarised in Table 1.
computing, which later evolved the vision of the smart
environment. In the current decade, the ‘smart environment’ G. WIRELESS TECHNOLOGY AND IoT
concept has become a booming technology. IoT is a heterogeneous network that will be connecting
The concept is diverse as it covers transportation/logistics, around 7 billion devices by 2025 using different wireless
healthcare, utilities, personal home/offices and much more. technologies and standards such as 2G, 3G, 4G, Bluetooth and
During this decade, concepts like augmented maps, Wi-Fi etc. as predicted in 2015 by Machina Research [50]. The
autonomous car, mobile ticketing, and passenger counting in mobile devices subscriptions as per 2G/3G/4G technologies
transportation/logistics domain have been successfully are shown in Fig. 2 [51]-[52]. The specifications of previously
implemented. The continuous improvement in these used wireless technologies are listed in Table 2 [53]. Currently
technologies is also currently in practice. The concept of IoT used IoT networks use these technologies as listed in Table 3
enabled Robot taxi, which is underway as a futuristic [51].
application [40]-[41]. Similarly, remote patient monitoring, It is important to realize that the above discussed
smart biosensors, smart ambulances, wearable devices, tele- technologies are not fully optimized for IoT business models
medicines in IoT-enabled healthcare domain benefitted the and devices which requires low power and less data-rate [51],
society manifold [42]-[43]. Public utility infrastructure has [54]. Therefore, a drift from various wireless technologies
been improved to a large extent, with the concept of smart such as 1G, 2G, 2.5G, 3G, 3.5G and 4G towards 5G can be
metering and smart-grid systems [44]. observed. The 5G technology is highlighted as it addresses the
Not only the country’s infrastructure has been improved, major challenges of a cellular network more effectively as
but also the end-users are also benefitted at a personal level compared to its predecessors [55]-[56]. These challenges are
with the concept of smart homes [45] and smart-cities [46]- appended below:
[47], which prove cost-effective and convenient. Smart health-
care manages the health of the consumer efficiently. The  Large bandwidth
concept of the smart gym [48] allows the end-user to monitor  Higher data-rate
his exercise schedule regularly. Since today’s human beings  Massive connectivity
are also socially active; there is a need of time to automatically  Low end to end latency
update their social activities on social media. This concept has  Cost-effective
already been introduced by Tweetbot [49]. A digital diary  Consistent Quality of Service
application that records personal data and updates it on the  Device computational capabilities
Google calendar [3]. Hence, IoT also proved beneficial at the  Device intelligence services
end-user level with the challenges of security and privacy.
VOLUME XX, 2017 3
TABLE I: IOT ENABLED SMART ENVIRONMENT
Wireless local
Communication area network
Applications Network Types Modules References
Enablers (WLAN)
standards
Architectures, protocols, and enabling
Wi-Fi, 3G, 4G, technologies for urban IoT.
Smart-Cities MAN, WRANs 802.11 [46], [57]
Satellite Integrated information centre for the
smart-city
Cloud-based home solution for detection
Smart Homes Wi-Fi WLAN 802.11 of faulty location using software-defined [58]
networks (SDNs)
Real-time monitoring system for
powering transmission-lines to avoid
Smart-Grid 3G, 4G, Satellite WLAN, WANs 802.11 [59]-[60]
disasters.
Smart-grid control
Smart Access control for services inside a
Wi-Fi WLAN 802.11 [13]
Buildings typical smart building
Smart WAN, WRANs, Smart-ticketing, smart passenger
Wi-Fi, Satellite 802.11 [61]
Transport MANs counting
Wi-Fi, 3G, 4G, WLAN, WPANs,
Smart Health 802.15.4 Remote health care [43], [62]
Satellite WANs
WLAN, WPANs, Energy-efficient remote monitoring and

Smart Industry Wi-Fi, Satellite 802.11 [63]
WANs optimized decision-making.
VOLUME XX, 2017 3
TABLE II: COMPARISON OF PREVIOUS WIRELESS TECHNOLOGIES FROM 1G - 4G
Wireless
Data-Rate Services Bandwidth Reference
Technology
1G 2.4Kbps Voice 30KHz [64]
2G 10Kbps Voice, Data 200KHz [64]-[65]
2.5G 50Kbps/200Kbps Voice, Data 200KHz [64]-[65]
3G 384Kbps Voice, Data, Video Calling 5 MHz [64]-[65]
3.5G 5-30Mbps Voice, Data, Video Calling 5MHz [64]-[65]

1.4MHz-
3.75G 100-200Mbps HD video, peer to peer file sharing, composite Web services, online gaming [64]-[66]
20MHz
DL 3Gbps 1.4MHz-
4G HD video, multimedia, data services at much higher data-rate [66]-[67]
UL 1.5Gbps 20MHz
TABLE III: WIRELESS TECHNOLOGY CURRENTLY EMPLOYED FOR IOT B. SPECTRUM REQUIREMENTS of 5G-IoT
Innovation in wireless domains, next-generation
technologies and development of 5G enable IoT requires
Type of Technology Wireless Technology
cutting-edge services and solutions together with the need of
Personal area network (PAN) Bluetooth Low Energy, Zigbee broadband spectrum to meet the demands of rapidly growing
Wireless local area network traffic. Therefore, a combination of low-band, mid-band and
Wi-Fi
(WLAN) high-band spectrum is desirable to manage the use cases of 5G
Cellular communication 3GPP, 4G (LTE) enable IoT successfully as suggested by 5G Americas [72].
Using a combination of different bands help to address certain
Wide area network (WAN) LP-WA
use cases better than others. The comparison of each band with
respect to different use cases presented in Table 5 [72].
III. VISION AND DEVELOPMENT OF 5G-ENABLED IOT Apart from the generalized spectrum requirements, a new
FROM 5G CELLULAR TECHNOLOGIES
air interface called New Radio (NR) is defined by 3GPP for
5G [73]. Its specification falls into two categories. Firstly,
A. 5G ENABLED IoT – GLOBAL INITIATIVES
FR1, which refers to spectrum below 6GHz [73]. Here, FCC
Various initiatives are taken all around the world for
has provided two licensed spectrums for 5G deployment. 1)
adopting and standardizing 5G enabled IoT, as listed in Table
Citizens Broadband Radio Service (CBRS) operating at 3.55
4 [68]. Different European projects beyond 4G can be found
– 3.7GHz. 2) C-band operating at 3.70 – 4.2GHz, respectively.
in [69]. Similarly, International Mobile Telecommunications
Others possible spectrum ranges suggested by FCC are 6-7
(IMT) also initiated the research and technology practices
GHz band (5.925-6.425 GHz and 6.425-7.125 GHz) which are
back in 2013 and the standardization in 2016 [68]. In 2015, it
unused and can support wider bandwidths than its predecessor
was decided that the Third Generation Partnership Project
(LTE) [72]. Secondly, FR2, which is a higher frequency
(3GPP) will formulate a group called technical specification
mmWave band [73] has significant unused spectrum having
group (TSG), which will be responsible for establishing 5G
large bandwidths which is suitable for 5G deployments.
RAN [70].
Therefore, 24GHz, 28GHz, 37GHz, 39GHz, and 47GHz are
During the same era, International Telecommunication
also identified for 5G deployment by FCC, and their auction
Union-Radio Communication (ITU-R) has taken the
is expected nearly [72]. For instance, in Nov 2018, 28GHz
responsibility of defining and specifying 5G technology by
band auction has begun. Other bands to be auctioned in 2019.
2020 [71].
In addition, bands such as 32 GHz, 42 GHz and 50 GHz are
TABLE IV: 5G INITIATIVES IN DIFFERENT COUNTRIES also under consideration [72]. The details of the spectrum
allocation of each NRs can be viewed in [73].
Country Initiative After looking at the above-mentioned spectrum
United States 4G America characteristics/requirements of 5G technology, there is still a
China IMT-2020 (5G) gap between the actual implementation of a 5G technology
Japan 2020 and beyond and the promises made by the next-generation wireless
Korea 5G forum network.
Europe 5G Private Public Partnership (5GPPP)
UK 5G Innovation Centre (5GIC)
VOLUME XX, 2017 3
TABLE V: COMPARISON OF THE SPECTRUM WITH RESPECT TO DIFFERENT 5G USE CASES
Spectrum Type Characteristics Use Cases

 Higher Coverage and Mobility 1. Massive Machine-Type Communications
Low-frequency band (below 2GHz)
 Wider channels availability 2. Indoor applications
Higher frequency millimetre waves  Short-range with low latency. 1. Enhanced mobile broadband Communications
(mmWave) bands  High capacity due to wider channelization 2. Urban and sub-urban applications
 Short-range with low latency and high capacity 1. 5G implementation in uncrowded/open areas
Mid-frequency bands
transmission for few macro-based stations 2. Urban deployment.
Hence certain technologies must be involved in the

implementation of 5G deployment [74]. For instance, A 5G transceiver must be capable of transmitting and
transmission/reception at mmWaves have high path-loss, and receiving the data at the same time and on the same frequency
their rate of absorption by the atmospheric parameters such as – hence a transceiver must be full-duplex and operate in both
rain and greenery are very high. Hence, it is likely that a Time Division Duplexing (TDD) and Frequency Division
smaller cellular architecture is needed in a form micro-, pico- Duplexing (FDD) modes. This is achieved by using silicon
or femto-cell to improve the coverage and decrease the path- transistors acting as fast switches allowing it to
loss at mmWaves. This paved the way towards the new transmit/receive on the same frequency [74].
concept – small, low-power cellular base station (BS). These Nevertheless, the above mentioned technologies are still in
cells are low power, compact and portable BSs which are their infancy and are under research to cater for their
placed meters apart. Thousands of these small cells form a drawbacks and implement a successful 5G enable the system.
dense network acting as a relay and boosting a signal between Furthermore, other related technologies are also discussed in
the end-users and BSs. As these small cells are handling the later section.
mmWaves, so the antennas attached to these BSs are smaller
and lighter as compared to their traditional counterparts. C. FEATURES EMPLOYED IN 5G PHY LAYER TO
Hence, these BSs can be easily fitted on poles or buildings top. SUPPORT 5G-IoT
Also, another benefit of using a small cell is spectrum re- MIMO, CoMP, and the HetNets, etc. are some of the
usage. To further enhance the advantages of small cells and features that have been standardized for LTE/LTE Advanced
reduce the complexity of extensive cells and antennas running (LTE-A) technology. These technologies show encouraging
in an urban environment – a M-MIMO technology is also results in providing massive connectivity and high data-rate.
incorporated. Therefore, 5G technology employs these concepts. These
A 4G BS currently employs dozens of antenna ports to concepts are discussed at first to get their insight in the next
handle its traffic efficiently. However, in case of 5G traffic, sub-sections.
these ports must be increased significantly from dozens to i.CARRIER AGGREGATION
hundreds to enhance the capacity by a factor of at least 22. Carrier aggregation was introduced in 4G LTE– A based on
This is solved by incorporating M-MIMO technology, which Release 10 of the 3GPP. It aggregates up to 5 component
targets at incorporating many antennas on a single BS. This carriers (CCs) of LTE–A having 20 MHz of bandwidth thus
creates another problem: Interference. To cater for it, a beam- enhancing the overall bandwidth to 100 MHz. In carrier
forming technology is incorporated [53], [74]. As the 5G aggregation, it is possible that the mobile device may receive
technology aimed at integrating various heterogeneous access more than one CC. Multiple CCs with different bandwidths
technologies leading to the implementation of the vision of can be aggregated in the uplink and downlink, where the
IoT. The 5G technology employs beam division multiple number of uplink aggregated CC may not be higher than the
access (BDMA), an advanced technology in which BS allows number of aggregated CC in the downlink. The CC may
an orthogonal beam to each mobile device and BDMA belong to the same band known as intra-band carrier
technique will divide the beam as per the position of the aggregation or to the different band known as inter-band
mobile devices. This enables the multiple accesses to the carrier aggregation. In each type of carrier aggregation, it is
devices hence increases the capacity manifold [53] with also possible to perform contiguous and non-contiguous
reduced interference. component carrier aggregation. In uplink and downlink, one
Besides using mmWaves for enhanced data-rates, M- of the CC must be selected as a primary component carrier
MIMO and beamforming for spectrum efficiency, 5G (PCC) while others as a secondary component carrier (SCC).
technology also requires high throughput with low latency. The number of CCs in different 3GPP releases has increased
This necessity is fulfilled by integrating the full-duplex significantly. In rel-10, there was only 2 CC in downlink while
technology, which targets at antennas’ transceiving no CC in uplink; in rel-11 carrier aggregation is improved
methodology [74]. drastically with the addition of 2 CC in the uplink, in rel-12
FDD/TDD is added. In rel-13, the number of 20 MHz
VOLUME XX, 2017 3
aggregated CCs are increased from 5 to 32. Even though carried out at 73GHz in an urban micro-cell environment for
carrier aggregation is a powerful technique but due to inter- BS diversity in CoMP style manner. The CoMP is a
and intra-band carrier aggregation, intermodulation product transceiver technique, by which the interference issues can be
may interfere with the signal. lessened. This is done by coordinating the transmission and
reception between the spatially distributed BSs using the
ii.MASSIVE-MIMO (M-MIMO)
channel state information. The technique includes coordinated
The MIMO technology is considered a necessary part of scheduling and joint transmission [68].
LTE-A and is based on the concept of spatial multiplexing.
Data streams from multiple antennas are multiplexed and are iv. HETEROGENEOUS NETWORKS (HETNETS)
transmitted over a variety of spatially separated channels. A network comprising of different types of cell layers
On the other hand, M-MIMO is an integral part of the 5G (femto-cells, pico-cell, micro-cell and macro-cell) and
infrastructure. At mmWave, many antenna elements are different RATs. These networks consist of low power nodes
needed so that highly directional narrow beam can be needed for data offloading [76]. The HetNets supports the
produced to counteract the path-loss. This technique becomes “green” aspect of 5G by maximizing the spectral efficiency
feasible to implement the high-order multi-user MIMO (MU- through reusing the spectrum tightly and with low uplink and
MIMO) to enhance the small cell capacity. In 5G radio access downlink power transmission [68] making it spectrum and
network (RAN), is based on M-MIMO in “macro-assisted energy-efficient. Sharing the spectrum by the massive number
small cells”. The control-plane traffic is provided in the macro of user equipment – an ultra-dense network (UDN) requires
cell at lower band frequencies using omni-directional antenna an intelligent interference mitigation technique. In order to
while user-data traffic is delivered using highly directional M- cater for the interference, the HetNets uses enhanced inter-cell
MIMO beam at mmWave band frequencies. M-MIMO allows interference coordination (e-ICIC) and further enhanced ICIC
the possibility of distributed MIMO, where multiple narrow (feICIC) [68]. These features allow it to handle massive traffic
beams simultaneously get transmitted to the same mobile and large node density; hence; making it suitable for satisfying
station from the BS at a different location to improve the requirements of service-driven 5G enabled IoT. The
throughput and reduce the correlation among the antenna HetNet enabled 5G-IoT based solutions are provided in [51].
elements. A subsequent of MIMO technology, which targets
v. D2D COMMUNICATIONS
at the spectrum efficiency by using the arrays of a few hundred
antennas having one time and frequency slot to serve tens of The HetNet realizes the coordination between macro-cell
user equipment. Hence, leveraging the MIMO technology on BS and the low power BS. However, for short-range
communication, it does not prove to be efficient. Hence, the
a larger scale. As compared to conventional MIMO
D2D communication is evolved that allows low power
technology, instead of using pilot waveforms for channel
consumption, better QoS and load balancing between the
estimation, M-MIMO employs TDD mode and rely on the
devices for short-range communication (< 200m). In light of
reciprocity mechanism between the uplink and the downlink this, the BS will either have the full control over allocating
channels [53]. M-MIMO has proved to be beneficial in the resources among source, destination and relaying nodes
providing improved radiation efficiency up to 100 times, or have partial control. Hence, keeping in view this scenario,
increased capacity by order of 10, increased protection against in [53], the authors [53] classified D2D communication into
interference and intended jamming, a significant decrease in four main categories.
latency and have low power and low-cost setup [75].  Device relaying with BS controlled link information.
iii.COORDINATED MULTIPOINT PROCESSING (CoMP)  Direct D2D communication with BS controlled link
information.
Coordinated Multipoint (CoMP) was first introduced and
 Device relaying with device-controlled link
standardized by 3GPP in Rel-10 and employed in LTE-A.
information.
CoMP transmission in downlink and reception in uplink is a
 Direct D2D communication with device-controlled link
very effective way to enhance the cell-edge user throughput.
information.
CoMP utilizes distributed MIMO for transmission and
However, D2D communication offers, significant security
reception from different antennas, which may not belong to and interference management issues [77] which must cater
the same cell to reduce, received spatial interference and before its deployment. Hence, research efforts [53] are
enhanced the received signal quality. CoMP is a very effective underway to provide a solution to these issues.
technique if deployed using MU-MIMO to increase cell edge
coverage and reduce the outages caused by blocking and vi.CENTRALIZED RADIO ACCESS NETWORK (CRAN)
channel conditions. There were several studies conducted The CRAN is another concept towards greener and
using CoMP with MIMO by NTT Docomo and Ericsson in cleaner communication by reallocating the functions of BSs.
Stockholm, Sweden to see the effectiveness in terms of It only allocates the radio functions to the BS that is remote
coordination between distributed MIMO and how the radio unit or remote radio head. Other unit like baseband unit
combination of both technologies will improve user data-rates (BBU) is distributed to the cloud-based central processor.
at mmWave frequency bands. At mmWave, the study was This enables centralized intelligence, cooperative
VOLUME XX, 2017 3
communication among cells, improved cell utilization and D. FEATURES EMPLOYED IN 5G NETWORKING LAYER
reduces complexity and cost at the BS end [68]. TO SUPPORT 5G-IoT
To support the massive connectivity of devices, different These network architectures previously hardware-based,
5G standards are developed for Cellular IoT. In the first will now have to manage the exponentially growing traffic,
phase, M2M communication, narrowband IoT (NB-IoT), a deployment/induction of new nodes, efficient network
low power wide area technology aiming to ease the massive forwarding infrastructures and the network expansion
deployment of IoT [78] are specified in the 3GPP Release flexibility. To achieve these objectives, different network-
14. Nowadays, the 3GPP standardization is working on the enabling technologies are developed which are now
enhancements, so that it can cater to the growing demand. emerging. Some of them are discussed below.
These enhancements are summarized in Table 6 [78].
TABLE VI: ENHANCEMENTS PROPOSED FOR 5G ON CELLULAR IOT
Open Standards Enhancements Proposed

Machine type  Reduced complexity
communications (MTC)  Increased transport block size to improve the data-rate
or M2M  Ensure Voice over LTE
Communications  Ensure multi-cast transmission
 Ensure enhanced coverage area
 Enhance MTC related to reception and transmission in case of time difference measurements
Narrowband Internet of  Multicast support for downlink transmission
things (NB-IoT)  Mobility and enhanced service continuity
 Lowering the maximum transmit power which enables reduced form-factor of wearable devices
 Increased transport block size with 1352 bits for downlink and 1800 bits for uplink to support high data-
rate, low latency and reduced power consumption
 Co-existence of NB-IoT User equipment with CDMA to lower the adjacent channel leakage ratio (ACLR)
up to 49dB
 New band supports for NB-IoT for effective utilization
 D2D communication is employed as an extension to NB-IoT to provide routing cellular links
Also, the New Services and Markets Technology Enablers

i. SOFTWARE-DEFINED WIRELESS SENSOR NETWORKING
(SMARTERs) by 3GPP are currently reviewing over 70 use
(SD-WSN)
cases for 5G service requirements. These use cases as per
there service requirements are categorized in Table 7 [78]- Conventional methods to deploy cellular technology are
[79]. generally hardware-based. The hardware-based deployment
Other than the above mentioned standards, there are other limits the flexibility of network expansion. Hence, SD-WSN,
network enablers to support 5G enabled IoT. It is evident that a promising paradigm, is developed to overcome this
the conventional features of a network are becoming limitation [78], [80]-[81]. The SD-WSN is a blend of SDN
obsolete with the increasing demands and requirements inside the WSNs. The SDN found its applications in data
needed to support massive connectivity of devices and centers (wired communication networks) and for Internet
providing them with the high QoS in case of massive and connectivity [82]. Currently, it is envisaged as 5G
critical IoT. technology enabler [83]-[84]. The primary purpose of using
this model for deployment of 5G networks is to decentralize
TABLE VII: SMARTER USE CASES the control logic plane from the network device [78] and
Services Use Cases providing the centralized mechanism to program the entire
Smart buildings, smart-cities, smart network.
Massive MTC (mMTC)
wearables, etc. Due to the increasing demands of the massive connected
Enhanced Mobile User movement from urban to rural areas devices, a centralized software-based paradigm like SDN is
Broadband (eMBB) or urban to sub-urban areas necessary to fulfil the requirements of maintaining consistent
Industrial control applications (drone or QoS. The SDN will simplify the network management issues
Critical Communications
robots), Tactile internet such as link/node down issues, allocation of resources, and
Network related functions such as deployment of new links/nodes by manifold. Different
Network Operations
slicing, routing, internetworking, proposed SDN solutions are Soft Air [51], Cloud RAN [85]
enhancements, security and and CONTENT [51]. Hence, it is seen as a pivotal enabler of
optimizations future next-generation 5G technology.
Enhancement of Vehicle
Self-driven cars, autonomous flights, etc.
to everything (eV2x)
VOLUME XX, 2017 3
ii. NETWORK FUNCTION VIRTUALIZATION (NFV) is not only possible at the beginning but also during the
The NFV virtualizes the network functions. These transmission.
functions can then be implemented into software packages A generic architecture of a cognitive radio consists of RF
which can later be used for providing network service front-end and a baseband processor [87]. To adapt to a time-
requirements [86]. The NFV and SDN are mutually changing RF environment, a control bus is attached to each
exclusive. The NFV concept is originated from the component. A cognitive radio differs from its traditional
perception of virtual machines, which can be installed on peers by its RF front end, which comprises of a wideband
different operating systems on the same server. Some antenna, power amplifier and an adaptive filter. These
foreseen use cases related to NFV are core virtualization and characteristics of an RF front end enables its wideband
centralized baseband processing within RANs [85]. The sensing capability. Also, a CR should be capable of detecting
NFV is a promising candidate for the successful deployment weak signals in a wide spectrum [89]-[90].
of 5G enabled IoT. Since, the network virtualization concept As the CR aims at adapting to the needs of a radio
supports the network scalability mechanism, network slicing environment, it follows a cognitive radio cycle [87]. The
over the distributed cloud, real-time processing capability cycle comprises of three basic steps: 1) Spectrum Sensing 2)
(optimizing speed and capacity in sliced networks) and Spectrum Analysis 3) Spectrum Decision.
maintains its heterogeneity – one of the significant features At first, the radio environment is sensed for white holes,
and requirements of 5G over IoT [51], [78]. Not only this, primary user appearance, variations in traffic and user
the NFV is also energy and cost-efficient, thereby reducing movements, etc. This is performed by various techniques
the capital and operational expenditure. Therefore, a lot of such as transmitter detection, cooperative detection and
research is going on this technology as recently (2019) Intel interference-based detection for cognitive radio networks
and Ericsson have developed the 5G platform that (XG network) [87]. Then the estimations/analysis on
incorporates the software and hardware management, NFV characteristics of spectrum holes is performed. This analyzes
and cloud services for 5G [75]. the parameters which define the quality of spectrum such as
interference, path-loss, wireless link errors, link-layer delay,
and holding time [87]. Finally, depending on the above-
iii. COGNITIVE RADIOS (CRs) mentioned steps, CR decides the suitable data-rate,
The current IoT applications from massive to critical IoT bandwidth and transmission mode of any transmission. This
represents massive connectivity and overloading of the transmission is in full consent with QoS, user requirements
network resources resulting in the scarcity of spectrum. and features of the spectrum. In [87], different spectrum
Hence, there is a dire need to use the spectrum efficiently decision rules are discussed depending upon the QoS of each
and intelligently to satisfy the growing demand. The CR use case.
handles this issue perceptively, by using/sharing the All these features make the CR an efficient candidate for
spectrum resources in an opportunistic manner [78]. the successful deployment in 5G.
A CR can be defined by [77] “a radio that can change its Other than these technologies, the domain of cloud
transmitter parameters based on the interaction with the computing is also putting up efforts in recognizing a 5G
environment in which it operates”. Hence, this leads to the enabled IoT [53]. The efforts put forward under this domain
concept of cognitive capability and reconfigurability. are summarized in Table 8.
The former refers to capturing the spatial and temporal
variations with reduced interference in a radio environment. TABLE VIII: CLOUD COMPUTING TECHNOLOGY FOR 5G NETWORKS
Subsequently, an optimized spectrum can be captured for
transmission. This is usually an unused spectrum called Cloud Computing
spectrum hole or white space. If it is being used by any technology for 5G Key Aspects
licensed user, then interference is reduced by three ways. RANs
Firstly, hopping towards the other spectrum hole. Secondly, Mobile cloud
Targets at fulfilling the resources
by keeping its power level low and thirdly by changing the computing
(computational power, memory, energy,
used modulation scheme. [87] Furthermore, the CR also storage) of mobile devices [91]
manages the spectrum by selecting the best available channel Targets at centralizing the RANs
RAN as a service
and allows spectrum sharing. Another striking property of functionalities as service on the cloud [53]
the CR is the spectrum mobility that is to vacate the allocated Joint RAN Backhaul Flexible centralized dynamic adjustment of
channel when a primary user arrives [78]. Operation network routes
The latter refers to the dynamic programmability with
respect to the radio environment. Hence, a CR can transmit
and receive using different frequencies, modulation
schemes, transmission powers, communication
technologies, protocol parameters and transmission access
technologies depending on its hardware [88]. This scalability
VOLUME XX, 2017 3
E. ARCHITECTURAL VIEW OF 5G-IoT and radio frequency integrated circuit (RFIC) [94]-[95]. A
As discussed in the previous section, the IoT technology comparison in the configuration of the antenna array at
has immense demands. Hence, a 5G architecture should be 28GHz for 4G and 5G technology is shown in [53].
capable of proving a scalable network, network The mmWave spectrum was underutilized since the past
virtualization facility, cloud services, network densification decade due to its inappropriateness for mobile
capability, mobility control services, radio access control, communications, particularly because of propagation issues
efficient resource allocation mechanism, and big IoT data such as path-loss, blocking, atmospheric and rain absorption
analysis tools. In a nutshell, a 5G-IoT based architecture etc. These issues are resolved now up to a great extent by
should provide an independent HetNet, that is self- using large antenna arrays and then steering up the beam
configurable as per the application requirement. A cellular energy and collecting it coherently.
5G architecture mainly consists of a front-haul, mid-haul,
and back-haul networks. F. QOS IN 5G-IoT
The front-haul network connects the remote radio-head The QoS in a 5G cellular network can be comprehended
(RRH) to the BBU, as shown in Fig. 3. Back-haul refers to by its spectral efficiency and latency [53]. The spectral
the connection between the BBU to the core wired network efficiency in a 5G network can be achieved by using non-
often made from coaxial cable and/or optical fibre. Mid-haul orthogonal signals and radio access methods, whereas
refers to the connection between RRH and the next link latency demands vary in case of a user and control plane
[68]. traffic.
Figure 3:Multi-tier Architecture showing the front-haul, back-haul and core network highlighting the peer-to-peer communication, macro- and femto-cells
The two logical layers existing in the 5G cellular network A typical 4G and 5G network cell spectral efficiency in
architecture are radio network and network cloud. The 5G the uplink and downlink is shown in Table 9, and the latency
cellular network architecture is explained in detail in [68], in the control and user plane is shown in Table 10 [96],
[92]-[93]. respectively.
In [53], the authors provide a mmWave solution for
implementing a 5G cellular network. The authors explained TABLE IX: CELL SPECTRAL EFFICIENCY FOR 5G TECHNOLOGY
that by using steerable antennas at the BS and the mobile Cell Spectral 5G
4G Technology
station with state-of-the-art CMOS technology and the Efficiency Technology
mmWave's spectrum, high data-rate can be achieved. Not Uplink (bps/Hz) 05 1.8
only this, but this will also support mobile communication Downlink (bps/Hz) 10 2.6
and the back-haul network. Furthermore, the lower
wavelength of the mmWaves makes it an excellent candidate TABLE X: LATENCY IN 5G NETWORK
for the polarization and various spatial processing techniques
5G
such as MIMO, M-MIMO and antenna beamforming, hence Latency 4G Technology
Technology
key architecture enabler of the 5G-IoT.
User plane (msec) 1 10
As far as a prototype building is concerned, realizing a
28GHz antenna array directly on the printed circuit board Control plane (msec) 50 100
(PCB) also reduces the insertion-loss between the antenna
VOLUME XX, 2017 3
It can be seen from Table 9 and Table 10, that about 50% patterns and associations) and providing various ways of
decrease in latency can be seen in the control plane in case communicating [104], as depicted in Fig. 4.
of 5G network with significant improvement in the cell Moreover, IoT big data analytics have proven to bring
spectral efficiency. value to the society. For example, it is reported that, by
detecting damaged pipes and fixing them, the Department of
G. STANDARDIZATION IN 5G-IoT Park Management in Miami, USA, has saved about one
The standardization process in the 5G-IoT involves million USD on their water bills [105]. Previous wireless
mainly two types of standards. One is the technology technologies did not support effective means of
standards that deal with network technology, protocols, and communication among machines. The increased data-rate,
wireless communication and data aggregation standards. reduced end-to-end latency, improved coverage, and the
Second is a regulatory standard that comprises of security support for large amounts of devices hold the potential to
and privacy of data. satisfy the most demanding IoT applications in terms of
communication requirements [106]. The integration of 5G
IV. ARTIFICIAL INTELLIGENCE DRIVEN USE CASES with IoT and AI has a huge impact on business by making
FOR 5G-IOT NETWORKS predictions using available data and taking control decisions
for enhancing the technology to improve quality-of-life.
To make the network self-configurable, self-organized,
and self-adaptive, the connected devices must be intelligent V. CHALLENGES AND ISSUES
to make decisions in an organized manner[97]. With the Current trends show that the future 5G-IoT network
continuous development of the IoT and communications should be capable of supporting the massive connectivity of
technology, especially under the epoch of 5G, mobile tasks devices by providing high and consistent QoS. A 5G network
with big scales of data have a strong demand in deep learning has to cater for massive as well as critical IoT. A massive IoT
such as virtual speech recognition and video classification deals with the applications that need such pervasive user
[98]. equipment that is economical, consumes less energy,
The 5G Intelligent IoT (5G I-IoT) is envisioned to address scalable and covers a large area. Whereas, critical IoT deals
some issues, such as enormous data processing and with those applications that require availability, reliability
communication channel congestion. With the combination of and safety and high QoS such as remote health care system.
5G technologies and AI algorithms, the visions of the 5G I- The traffic generated by such IoTs is different from that
IoT are processing big data intelligently, improving the generated by cellular systems in many aspects. First, unlike
effective utilization of channels and optimizing the case of broadband access, most IoT traffic is in the
communication channels [99]. Moreover, the incorporation uplink. Moreover, IoT networks’ messages are typically
of AI in the core components of the firmware will provide a small in size and sparse in time. Furthermore, IoT devices
secure environment for the execution of applications to make are limited in energy and computation resources. These IoT
informed decisions without any interruption [97]. devices’ characteristics make their access to 5G systems
The high-speed mobile networking of 5G and the different from classical cellular devices [107].
improvement of IoT promote the development of big data Identifying the right system parameter configuration for
applications. Many applications in 5G, such as face the specific IoT use case is a big challenge [108]. Among
recognition, natural language processing, and so on, are able many challenges discussed for IoT access technology,
to operate in the terminal [97]. coverage enhancement (CE) is one of the primary and
In healthcare, for instance, these advantages are useful in challenging physical layer objectives to improve the
improving the system for millions of people. Chen et al. maximum coupling required to support tactile internet and
[100] designed a personalized emotion-aware healthcare multimedia applications [109]. Security/privacy, energy
system using 5G. It focuses on the emotional care, efficiency and massive connectivity are other major
particularly for children, and mentally ill and elderly people. concerns. Hence, research efforts are being actively pursued
In [101], Genetic algorithm and simulated annealing were by researchers in each domain.
used to determine the optimal position for 5G drone base Apart from these initiatives, researchers are also putting
stations given the constraints of coverage, energy and cost. efforts in providing solutions to the challenges related to the
Another possible other application of AI is concerned with architecture, physical layer, channel access; energy
the implementation of 5G networks, which is shown in the efficiency and spectrum efficiency of 5G enabled IoT
CogNet project [102] in which an architecture of an technology. Some of these researches and their solutions to
autonomic self-managing network extending NFV the proposed challenges are listed in Table 11. The extensive
management with the machine learning-based decision- research efforts are due to the high demands and
making mechanism is discussed. requirements of 5G enabled IoT networks driven by
The treasure of data generated from interconnected IoT previously never thought-of use cases [109].
devices can be used to predict crimes and accidents based on
the existing data [103]. Thus, it is providing new ideas that
can become projects for tech companies, generating big data
(large volume of data which can be used to reveal trends,
VOLUME XX, 2017 3
Figure 4: Architectural Scenario Where 5G Meets Artificial Intelligence
Future intelligent systems require such IoT nodes to work

These requirements include high data-rate for high under varying operating conditions, and with the network
definition video streaming, augmented reality (AR) and and cloud maximizing system intelligence while minimizing
virtual reality (VR) applications; The applications typically energy/information [112]. The method to maximize the
require very high data-rates, up to 25Mbps [51]. The IoT- overall performance of the network is to incorporate
based 5G network is completely fragmented because it lacks intelligence and self-adaptation into a self-organizing
the component of interoperability. This encourages network (SON). The main goal of the SON is to improve
exploration of new intelligent, open-source standardization service quality and reduce the costs associated with network
for seamless communication among attached IoT devices by operations by reducing human involvement while enhancing
making the operating system compatible for every object network performance[113].
[110]. This standardization could be implemented at the The primary motives for using AI related technologies in
cloud platform to connect and enable interoperability [97]. the 5G-IoT infrastructures is to enable the network to
Nowadays, deep learning is applied to IoT and mobile intelligently adjust its configurations as the requirements or
applications for implementing real-time or near real-time parameters of the environment change [102]. The new 5G
data processing tasks. Considering the limited performance network should be able to provide efficient solutions for
of data transfer networks, transmitting massive data to Cloud radio resource management (RRM), mobility management
for deep learning will consume much energy and produce (MM), management and orchestration (MANO), and service
much transmission delay, which reduces the efficiency of provisioning management (SPM), making dedicated purpose
deep learning tasks [98]. The challenge of the cloud-centric networks no longer necessary instead dynamic
system was considered and an autonomous and incremental reconfigurations of networks as the concept of NS [114].
computing framework and architecture for deep learning-
based IoT applications was proposed to tackle it [111].
VOLUME XX, 2017 3
VI. CONCLUSION
This paper presented an exhaustive review of the 5G
wireless technologies that have become key enablers for the
ubiquitous deployment of the IoT technology. The survey
presented a review of the evolution of cellular wireless
technologies making a case how 5G wireless technology
improved upon its predecessor technologies making
ubiquitous deployment of IoT possible. The various
architectural components of the 5G networks are also
discussed, with special emphasis to the key improvements to
the physical and network layer of 5G networks over its
predecessors.
The paper also discusses in great depth the challenges of
QoS requirements in modern day 5G-IoT whose traffic
characteristics differs significantly from other legacy 5G
network applications, being predominantly in the uplink
direction instead of the downlink. High data transmission
rates with low latency from the 5G-IoT nodes are vital for
the cloud-based application layer programs running state of
the art artificial intelligence, machine and deep learning
algorithms for efficient real-time data processing and
prediction. Such modern day applications running on top of
5G-IoT, e.g. smart transport, smart healthcare etc are also
discussed and benchmarks for acceptable performance, key
performance indicators (KPIs) are also presented. Another
challenge discussed in this paper is that of the
standardization issues due to the several heterogeneous
nodes participating in the 5G-IoT network (HetNets).
The comprehensive review presented in this paper will
help in more coordinated efforts from both industry and
academia for bringing about improvements in the 5G-IoT
technologies.
VOLUME XX, 2017 3
TABLE XI: TRENDING RESEARCH ON CHALLENGES OF 5G-IOT
Article Challenges Solution Reference

Suitable radio numerology is designed with a
Enabling Massive IoT in
Flexibility in the physical layer radio random-access channel to support massive
5G and beyond Systems:
framework? of 5G technology to satisfy connection density and is capable of dealing with [115]
PHY radio frame design
the diverse requirements of IoT channel impairment and imperfections in the
considerations
transceiver
A Survey of Client- Huge signalling overhead in network
A review on client-controlledHetNets for 5G
controlled HetNets for control schemes for network edge
networks is provided in-depth with distributed and [116]
5G devices and leveraging Radio access
hybrid control approaches
technologies (RATs)
The next-generation of Different companies such as Tata communications, DellIoT Services, Sierra Wireless are
[117]
IoT presented which are developing IoT technology all around the globe
Long-Range IoT
Technologies: The Dawn Efficient low power WAN (LPWAN) LoRa is presented as the latest and most promising
[118]
of LoRa™ enabling technologies for IoT technology
Extracting and exploiting Scarcity in spectrum resources to

inherent sparsity for provide support for IoT devices to Wide-spectrum management techniques, D2D
efficient IoT Support in enable 5G technology. Also, the radio communications, and the concept of edge cloud [107]
5G: Challenges and access channel has many limitations to solutions are presented
potential solutions handle IoT enabled 5G devices
Solutions for the wireless powering of IoT devices

using near and far-field techniques is provided.
Keeping in mind the 5G-IoT scenarios,
Furthermore, a new networking model called
Energizing 5G the efficient recharging of ubiquitous [2]
Wireless power communication network is
IoT devices is a tedious task
introduced that integrates wireless power
transmission and communication.
5G cellular network architecture is presented with
its enabling technologies such as M-MIMO, D2D
To cope up with the needs of 5G enable
A Survey of 5G Network: communication. Other associated emerging
IoT such as better data-rate, reduced
Architecture and technologies are also presented, such as ultra-dense [53]
latency, consistent quality of service
Emerging Technologies networks, cognitive radios, millimetre-wave (mm-
and huge spectrum resources
wave) solutions for 5G networks and cloud
technologies, etc.
A joint radio access and backhaul framework is
The requirements for maintaining the
presented which addresses the QoS issues
high quality of service in a 5G
5G Backhaul Challenges efficiently. The framework called Backhaul as a
paradigm present a bottleneck:
and Emerging Service (BHaaS) which is a part of SDN with
backhaul. The requirements of backhaul [68]
Research Directions: A Radio access network (RAN) intelligence, Self-
must be addressed as it is responsible
Survey optimizing network (SON) and caching
for connecting the highly dense and
capabilities has a complete vision of end-to-end
heavy traffic cells to the core
network and also enables optimization.
The limited computing resource and

A Heuristic Offloading battery consumption of mobile devices Computation offloading is a key technique for the
(MDs), mobile tasks are often offloaded deep learning edge services in 5G networks. In
Method for Deep
to the remote infrastructure, like cloud order to shorten the transmission delay of deep [98]
Learning Edge Services platforms, which leads to the learning tasks, a heuristic offloading method is
in 5G Networks unavoidable offloading transmission devised and shown
delay.
VOLUME XX, 2017 3
ACKNOWLEDGEMENT Overview, Assumptions, Problem Statement, and Goals,” RFC

The authors would like to acknowledge the funding from Editor, RFC4919, Aug. 2007.
[22] C. Perkins, IP Mobility Support for IPv4, IETF RFC 3344, Aug.
the Deanship of Scientific Research, Islamic University of 2002.
Madinah, Saudi Arabia under the Research Grant No. 29/40. [23] A. Nilssen, Security and privacy standardization in internet of
things, in: eMatch’09 – Future Internet Workshop, Oslo, Norway,
REFERENCES September 2009.
[24] M. Feldhofer, S. Dominikus, and J. Wolkerstorfer, “Strong
Authentication for RFID Systems Using the AES Algorithm,” in
[1] L. Roselli et al., “Smart Surfaces: Large Area Electronics Systems Cryptographic Hardware and Embedded Systems - CHES 2004,
for Internet of Things Enabled by Energy Harvesting,” Proc. IEEE, vol. 3156, M. Joye and J.-J. Quisquater, Eds. Berlin, Heidelberg:
vol. 102, no. 11, pp. 1723–1746, Nov. 2014. Springer Berlin Heidelberg, 2004, pp. 357–370.
[2] A. Costanzo and D. Masotti, “Energizing 5G: Near- and Far-Field [25] B. Calmels, S. Canard, M. Girault, and H. Sibert, “Low-Cost
Wireless Energy and Data Trantransfer as an Enabling Technology Cryptography for Privacy in RFID Systems,” in Smart Card
for the 5G IoT,” IEEE Microw. Mag., vol. 18, no. 3, pp. 125–136, Research and Advanced Applications, vol. 3928, J. Domingo-
May 2017. Ferrer, J. Posegga, and D. Schreckling, Eds. Berlin, Heidelberg:
[3] L. Atzori, A. Iera, and G. Morabito, “The Internet of Things: A Springer Berlin Heidelberg, 2006, pp. 237–251.
survey,” Comput. Netw., vol. 54, no. 15, pp. 2787–2805, Oct. 2010. [26] M. Weiser, “The computer for the 21 st century,” ACM SIGMOBILE
[4] Cellular Networks for Massive IoT: Enabling Low Power Wide Area Mob. Comput. Commun. Rev., vol. 3, no. 3, pp. 3–11, Jul. 1999.
Applications, Ericsson, Stockholm, Sweden, 2016, pp. 1-13. [27] N. Kaur and S. K. Sood, “An Energy-Efficient Architecture for the
[5] W. Ejaz et al., “Internet of Things (IoT) in 5G Wireless Internet of Things (IoT),” IEEE Syst. J., vol. 11, no. 2, pp. 796–805,
Communications,” IEEE Access, vol. 4, pp. 10310–10314, 2016. Jun. 2017.
[6] A. Osseiran et al., “Scenarios for 5G mobile and wireless [28] Ö. U. Akgül and B. Canberk, “Self-Organized Things (SoT): An
communications: the vision of the METIS project,” IEEE Commun. energy efficient next generation network management,” Comput.
Mag., vol. 52, no. 5, pp. 26–35, May 2014. Commun., vol. 74, pp. 52–62, Jan. 2016.
[7] D. Niyato, M. Maso, D. I. Kim, A. Xhafa, M. Zorzi, and A. Dutta, [29] Z. Zhou, J. Tang, L.-J. Zhang, K. Ning, and Q. Wang, “EGF-tree:
“Practical Perspectives on IoT in 5G Networks: From Theory to an energy-efficient index tree for facilitating multi-region query
Industrial Challenges and Business Opportunities,” IEEE Commun. aggregation in the internet of things,” Pers. Ubiquitous Comput.,
Mag., vol. 55, no. 2, pp. 68–69, Feb. 2017. vol. 18, no. 4, pp. 951–966, Apr. 2014.
[8] National Intelligence Council, Disruptive Civil Technologies – Six [30] J. Tang, Z. Zhou, J. Niu, and Q. Wang, “An energy efficient
Technologies with Potential Impacts on US Interests Out to 2025 – hierarchical clustering index tree for facilitating time-correlated
Conference Report CR 2008-07, Apr. 2008. . region queries in the Internet of Things,” J. Netw. Comput. Appl.,
[9] I. Yaqoob et al., “Internet of Things Architecture: Recent Advances, vol. 40, pp. 1–11, Apr. 2014.
Taxonomy, Requirements, and Open Challenges,” IEEE Wirel. [31] S. D’Oro, L. Galluccio, G. Morabito, and S. Palazzo, “Exploiting
Commun., vol. 24, no. 3, pp. 10–16, Jun. 2017. Object Group Localization in the Internet of Things: Performance
[10] X. Krasniqi and E. Hajrizi, “Use of IoT Technology to Drive the Analysis,” IEEE Trans. Veh. Technol., vol. 64, no. 8, pp. 3645–
Automotive Industry from Connected to Full Autonomous 3656, Aug. 2015.
Vehicles,” IFAC-Pap., vol. 49, no. 29, pp. 269–274, 2016. [32] J. Zhou et al., “CloudThings: A common architecture for integrating
[11] Gartner’s hype cycle special report for 2011, Gartner Inc., 2012. the Internet of Things with Cloud Computing,” in Proceedings of
http://www.gartner.com/technology/research/hype-cycles/. . the 2013 IEEE 17th International Conference on Computer
[12] K. Ashton, That “Internet of Things” thing, RFiD Journal, 2009. Supported Cooperative Work in Design (CSCWD), Whistler, BC,
[13] E. Ahmed, I. Yaqoob, A. Gani, M. Imran, and M. Guizani, “Internet- Canada, 2013, pp. 651–657.
of-things-based smart environments: state of the art, taxonomy, and [33] H.-C. Hsieh and C.-H. Lai, “Internet of Things Architecture Based
open research challenges,” IEEE Wirel. Commun., vol. 23, no. 5, pp. on Integrated PLC and 3G Communication Networks,” in 2011
10–16, Oct. 2016. IEEE 17th International Conference on Parallel and Distributed
[14] INFSO D.4 Networked Enterprise & RFID INFSO G.2 Micro & Systems, Tainan, Taiwan, 2011, pp. 853–856.
Nanosystems, in: Co-operation with the Working Group RFID of the [34] S. Cirani et al., “A Scalable and Self-Configuring Architecture for
ETP EPOSS, Internet of Things in 2020, Roadmap for the Future, Service Discovery in the Internet of Things,” IEEE Internet Things
Version 1.1, 27 May 2008. J., vol. 1, no. 5, pp. 508–521, Oct. 2014.
[15] ITU Internet Reports, The Internet of Things, Nov. 2005. [35] I. Lee and K. Lee, “The Internet of Things (IoT): Applications,
[16] J. Gubbi, R. Buyya, S. Marusic, and M. Palaniswami, “Internet of investments, and challenges for enterprises,” Bus. Horiz., vol. 58,
Things (IoT): A vision, architectural elements, and future no. 4, pp. 431–440, Jul. 2015.
directions,” Future Gener. Comput. Syst., vol. 29, no. 7, pp. 1645– [36] “Internet of things, pan european research and innovation vision,”
1660, Sep. 2013. 2011. [Online]. Available: http://www.internet-of-things-
[17] K. Shafique et al., “Energy Harvesting Using a Low-Cost Rectenna research.eu/ pdf/IERC_IoT-
for Internet of Things (IoT) Applications,” IEEE Access, vol. 6, pp. Pan%20European%20Research%20and%20Innovation%
30932–30941, 2018. 20Vision_2011_web.pdf.
[18] Q. Awais, Y. Jin, H. T. Chattha, M. Jamil, H. Qiang, and B. A. [37] V. Gazis et al., “Short Paper: IoT: Challenges, projects,
Khawaja, “A Compact Rectenna System With High Conversion architectures,” in 2015 18th International Conference on
Efficiency for Wireless Energy Harvesting,” IEEE Access, vol. 6, Intelligence in Next Generation Networks, Paris, France, 2015, pp.
pp. 35857–35866, 2018. 145–147.
[19] B.-M. Pang, H.-S. Shi, and Y.-X. Li, “An Energy-Efficient MAC [38] “Daimler’s car2go: Rent a Smart anywhere, anytime”, 2012-12-21.
Protocol for Wireless Sensor Network,” in Future Wireless [39] M. Weiser, R. Gold, and J. S. Brown, “The origins of ubiquitous
Networks and Information Systems, vol. 143, Y. Zhang, Ed. Berlin, computing research at PARC in the late 1980s,” IBM Syst. J., vol.
Heidelberg: Springer Berlin Heidelberg, 2012, pp. 163–170. 38, no. 4, pp. 693–696, 1999.
[20] S. Rani, S. H. Ahmed, R. Talwar, J. Malhotra, and H. Song, “IoMT: [40] M. Zhang, T. Yu, and G. F. Zhai, “Smart Transport System Based
A Reliable Cross Layer Protocol for Internet of Multimedia on ‘The Internet of Things,’” Appl. Mech. Mater., vol. 48–49, pp.
Things,” IEEE Internet Things J., vol. 4, no. 3, pp. 832–839, Jun. 1073–1076, Feb. 2011.
2017. [41] LIN, Hong-En & Zito, Rocco. (2005). A review of travel-time
[21] N. Kushalnagar, G. Montenegro, and C. Schumacher, “IPv6 over prediction in transport and logistics. Proceedings of the Eastern
Low-Power Wireless Personal Area Networks (6LoWPANs): Asia Society for Transportation Studies. 5
VOLUME XX, 2017 3
[42] Wearable Technology Market Worth 31.27 Billion USD by 2020", systems,” in 2015 12th Annual IEEE Consumer Communications
Dec 2015. and Networking Conference (CCNC), Las Vegas, NV, USA, 2015,
[43] J. B. Wang et al., “Wearable Sensor/Device (Fitbit One) and SMS pp. 826–834.
Text-Messaging Prompts to Increase Physical Activity in [63] F. Shrouf, J. Ordieres, and G. Miragliotta, “Smart factories in
Overweight and Obese Adults: A Randomized Controlled Trial,” Industry 4.0: A review of the concept and of energy management
Telemed. E-Health, vol. 21, no. 10, pp. 782–792, Oct. 2015. approached in production based on the Internet of Things
[44] J. Lloret, J. Tomas, A. Canovas, and L. Parra, “An Integrated IoT paradigm,” in 2014 IEEE International Conference on Industrial
Architecture for Smart Metering,” IEEE Commun. Mag., vol. 54, Engineering and Engineering Management, Selangor DarulEhsan,
no. 12, pp. 50–57, Dec. 2016. Malaysia, 2014, pp. 697–701.
[45] E. Park, Y. Cho, J. Han, and S. J. Kwon, “Comprehensive [64] K. R. Santhi, V. K. Srivastava, G. SenthilKumaran, and A. Butare,
Approaches to User Acceptance of Internet of Things in a Smart “Goals of true broad band’s wireless next wave (4G-5G),” in 2003
Home Environment,” IEEE Internet Things J., vol. 4, no. 6, pp. IEEE 58th Vehicular Technology Conference. VTC 2003-Fall
2342–2350, Dec. 2017. (IEEE Cat. No.03CH37484), Orlando, FL, USA, 2003, pp. 2317-
[46] A. Zanella, N. Bui, A. Castellani, L. Vangelista, and M. Zorzi, 2321 Vol.4.
“Internet of Things for Smart Cities,” IEEE Internet Things J., vol. [65] T. Halonen, J. Romero, and J. Melero, Eds., GSM, GPRS and EDGE
1, no. 1, pp. 22–32, Feb. 2014. Performance: Evolution towards 3G/UMTS, 1st ed. Wiley, 2003.
[47] S. P. Mohanty, U. Choppali, and E. Kougianos, “Everything you [66] “J. G. Andrews, A. Ghosh, and R. Muhamed, Fundamentals of
wanted to know about smart cities: The Internet of things is the WiMAX, Englewood Cliffs, NJ, USA: Prentice-Hall, 2007.” .
backbone,” IEEE Consum. Electron. Mag., vol. 5, no. 3, pp. 60–70, [67] S. Sesia, I. Toufik, and M. Baker, Eds., LTE - The UMTS Long Term
Jul. 2016. Evolution: From Theory to Practice. Chichester, UK: John Wiley &
[48] Z. Zhao, S. Ali Etemad, and A. Arya, “Gamification of Exercise and Sons, Ltd, 2011.
Fitness using Wearable Activity Trackers,” in Proceedings of the [68] M. Jaber, M. A. Imran, R. Tafazolli, and A. Tukmanov, “5G
10th International Symposium on Computer Science in Sports Backhaul Challenges and Emerging Research Directions: A
(ISCSS), vol. 392, P. Chung, A. Soltoggio, C. W. Dawson, Q. Meng, Survey,” IEEE Access, vol. 4, pp. 1743–1766, 2016.
and M. Pain, Eds. Cham: Springer International Publishing, 2016, [69] R. Chávez-Santiago et al., “5G: The Convergence of Wireless
pp. 233–240. Communications,” Wirel. Pers. Commun., vol. 83, no. 3, pp. 1617–
[49] X. Li, R. Lu, X. Liang, X. Shen, J. Chen, and X. Lin, “Smart 1642, Aug. 2015.
community: an internet of things application,” IEEE Commun. [70] 3GPP TSG-RAN. 5G: From Myth to Reality, accessed on Feb. 13,
Mag., vol. 49, no. 11, pp. 68–75, Nov. 2011. 2019. [Online]. Available: http://www.3gpp.org/speci_cations-
[50] M. J. Marcus, “5G and ‘IMT for 2020 and beyond’ [Spectrum Policy groups/ranplenary. .
and Regulatory Issues],” IEEE Wirel. Commun., vol. 22, no. 4, pp. [71] Working Party 5D (WP 5D) IMT Systems, ITU Towards IMT for
2–3, Aug. 2015. 2020 and Beyond, accessed on Feb. 13, 2019. [Online]. Available:
[51] S. Li, L. D. Xu, and S. Zhao, “5G Internet of Things: A survey,” J. http://www.itu.int/en/ITU-R. .
Ind. Inf. Integr., vol. 10, pp. 1–9, Jun. 2018. [72] 5G Spectrum Vision, 5G Americas White Paper- 5G Spectrum
[52] “https://www.ericsson.com/en/press-releases/2017/11/ericsson- Vision, Feb 2019. .
predicts-1-billion-5g-subscriptions-in-2023. . [73]https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDe
[53] A. Gupta and R. K. Jha, “A Survey of 5G Network: Architecture tails.aspx?specificationId=3283. .
and Emerging Technologies,” IEEE Access, vol. 3, pp. 1206–1232, [74] Amy Nordrum, et.all, ‘Everything you need to know about 5G’,
2015. IEEE Spectrum, Jan 27, 2019.
[54] Nokia, ‘LTE evolution for IoT connectivity,’ Nokia, Espoo, Finland, [75] Corinne Reichert, ‘Intel and Ericsson develop 5G platform’, Feb 12,
White Paper, 2017, pp. 1-18. . 2019.
[55] M. Fallgren et al., "Scenarios, Requirements and KPIs for 5G [76] A. Aijaz, H. Aghvami, and M. Amani, “A survey on mobile data
Mobile and Wireless System", document ICT-317669- offloading: technical and business perspectives,” IEEE Wirel.
METIS/D1.1, Apr. 2013. Commun., vol. 20, no. 2, pp. 104–112, Apr. 2013.
[56] Industry Proposal for a Public Private Partnership (PPP) in [77] “FCC, ET Docket No 03-222 Notice of proposed rule makingand
Horizon 2020 (Draft Version 2.1), Horizon 2020 Advanced 5G order, December 2003.”
Network Infrastructure for the Future Internet. [78] G. A. Akpakwu, B. J. Silva, G. P. Hancke, and A. M. Abu-Mahfouz,
[57] I. Ganchev, Z. Ji and M. O’Droma, “A Generic IoT Architecture for “A Survey on 5G Networks for the Internet of Things:
Smart Cities,” in 25th IET Irish Signals & Systems Conference 2014 Communication Technologies and Challenges,” IEEE Access, vol.
and 2014 China-Ireland International Conference on Information 6, pp. 3619–3647, 2018.
and Communities Technologies (ISSC 2014/CIICT 2014), Limerick, [79] Feasibility Study on New Service and Market Technology Enablers,
Ireland, 2014, pp. 196–199. document 3GPP TR 22.891 v14.2.0, 3GPP, 2016. [Online].
[58] Y. Kim and Y. Lee, “Automatic Generation of Social Relationships Available:
between Internet of Things in Smart Home Using SDN-Based Home htpps://portal.3gpp.org/desktopmodules/Speci_cations/Speci_catio
Cloud,” in 2015 IEEE 29th International Conference on Advanced n details.aspx?speci_cationId=2897. .
Information Networking and Applications Workshops, Gwangiu, [80] H. I. Kobo, A. M. Abu-Mahfouz, and G. P. Hancke, “A Survey on
South Korea, 2015, pp. 662–667. Software-Defined Wireless Sensor Networks: Challenges and
[59] Q. Ou, Y. Zhen, X. Li, Y. Zhang, and L. Zeng, “Application of Design Requirements,” IEEE Access, vol. 5, pp. 1872–1899, 2017.
Internet of Things in Smart Grid Power Transmission,” in 2012 [81] M. Ndiaye, G. Hancke, and A. Abu-Mahfouz, “Software Defined
Third FTRA International Conference on Mobile, Ubiquitous, and Networking for Improved Wireless Sensor Network Management:
Intelligent Computing, Vancouver, Canada, 2012, pp. 96–100. A Survey,” Sensors, vol. 17, no. 5, p. 1031, May 2017.
[60] Y. Y. Ma, Y. F. Wang, W. M. Lin, and T. Zhang, “Research on [82] W. Xia, Y. Wen, C. H. Foh, D. Niyato, and H. Xie, “A Survey on
application and security protection of Internet of Things in smart Software-Defined Networking,” IEEE Commun. Surv. Tutor., vol.
grid,” in IET International Conference on Information Science and 17, no. 1, pp. 27–51, 2015.
Control Engineering 2012 (ICISCE 2012), Shenzhen, China, 2012, [83] A. Hakiri and P. Berthou, “Leveraging SDN for the 5G Networks:
pp. 1.54-1.54. Trends, Prospects, and Challenges,” in Software Defined Mobile
[61] E. Ahmed, I. Yaqoob, A. Gani, M. Imran, and M. Guizani, “Internet- Networks (SDMN), M. Liyanage, A. Gurtov, and M. Ylianttila, Eds.
of-things-based smart environments: state of the art, taxonomy, and Chichester, UK: John Wiley & Sons, Ltd, 2015, pp. 61–80.
open research challenges,” IEEE Wirel. Commun., vol. 23, no. 5, pp. [84] I. F. Akyildiz, P. Wang, and S.-C. Lin, “SoftAir: A software defined
10–16, Oct. 2016. networking architecture for 5G wireless systems,” Comput. Netw.,
[62] A.-M. Rahmani et al., “Smart e-Health Gateway: Bringing vol. 85, pp. 1–18, Jul. 2015.
intelligence to Internet-of-Things based ubiquitous healthcare
VOLUME XX, 2017 3
[85] Ericsson, ‘Cloud RAN. The benefits of virtualization, centralization [105] M. Chen, S. Mao, Y. Zhang, and V. C. M. Leung, “Related
and coordination,’ Ericsson, Stockholm, Sweden, White Paper, Uen Technologies,” in Big Data, Cham: Springer International
284 23-3271, 2015. . Publishing, 2014, pp. 11–18.
[86] I. F. Akyildiz, S.-C. Lin, and P. Wang, “Wireless software-defined [106] M. R. Palattellaet al., “Internet of Things in the 5G Era: Enablers,
networks (W-SDNs) and network function virtualization (NFV) for Architecture, and Business Models,” IEEE J. Sel. Areas Commun.,
5G cellular systems: An overview and qualitative evaluation,” vol. 34, no. 3, pp. 510–527, Mar. 2016.
Comput. Netw., vol. 93, pp. 66–79, Dec. 2015. [107] B. Khalfi, B. Hamdaoui, and M. Guizani, “Extracting and
[87] I. F. Akyildiz, W.-Y. Lee, M. C. Vuran, and S. Mohanty, “NeXt Exploiting Inherent Sparsity for Efficient IoT Support in 5G:
generation/dynamic spectrum access/cognitive radio wireless Challenges and Potential Solutions,” IEEE Wirel. Commun., vol. 24,
networks: A survey,” Comput. Netw., vol. 50, no. 13, pp. 2127– no. 5, pp. 68–73, Oct. 2017.
2159, Sep. 2006. [108] P. Annamalai, J. Bapat, and D. Das, “Emerging Access
[88] F. K. Jondral, “Software-Defined Radio—Basics and Evolution to Technologies and Open Challenges in 5G IoT: From Physical Layer
Cognitive Radio,” EURASIP J. Wirel. Commun. Netw., vol. 2005, Perspective,” in 2018 IEEE International Conference on Advanced
no. 3, p. 652784, Dec. 2005. Networks and Telecommunications Systems (ANTS), Indore, India,
[89] Qing Zhao, Lang Tong, and A. Swami, “Decentralized cognitive 2018, pp. 1–6.
mac for dynamic spectrum access,” in First IEEE International [109] J. Skold, Research Trends and IMT Beyond 2020, Feb. 2014.
Symposium on New Frontiers in Dynamic Spectrum Access [Online]. Available: https://www.itu.int/en/ITU-D/Regional-
Networks, 2005. DySPAN2005., Baltimore, MD, USA, 2005, pp. Presence/AsiaPacific/Documents/. .
224–232. [110] A. Al-Fuqaha, M. Guizani, M. Mohammadi, M. Aledhari, and M.
[90] Haitao Zheng and Lili Cao, “Device-centric spectrum Ayyash, “Internet of Things: A Survey on Enabling Technologies,
management,” in First IEEE International Symposium on New Protocols, and Applications,” IEEE Commun. Surv. Tutor., vol. 17,
Frontiers in Dynamic Spectrum Access Networks, 2005. no. 4, pp. 2347–2376, 2015.
DySPAN2005., Baltimore, MD, USA, 2005, pp. 56–65. [111] M. Song et al., “In-Situ AI: Towards Autonomous and Incremental
[91] A. R. Khan, M. Othman, S. A. Madani, and S. U. Khan, “A Survey Deep Learning for IoT Systems,” in 2018 IEEE International
of Mobile Cloud Computing Application Models,” IEEE Commun. Symposium on High Performance Computer Architecture (HPCA),
Surv. Tutor., vol. 16, no. 1, pp. 393–413, 2014. Vienna, 2018, pp. 92–103.
[92] P. Agyapong, M. Iwamura, D. Staehle, W. Kiess, and A. [112] B. Chatterjee, N. Cao, A. Raychowdhury, and S. Sen, “Context-
Benjebbour, “Design considerations for a 5G network architecture,” Aware Intelligence in Resource-Constrained IoT Nodes:
IEEE Commun. Mag., vol. 52, no. 11, pp. 65–75, Nov. 2014. Opportunities and Challenges,” IEEE Des. Test, vol. 36, no. 2, pp.
[93] 5G PPP Architecture working group, White Paper, July 2016. 7–40, Apr. 2019.
Available Online: https://5g-ppp.eu/wp- [113] J. Guo et al., “A Deep Reinforcement Learning based Mechanism
content/uploads/2014/02/5G-PPP-5G-Architecture-WP-July- for Cell Outage Compensation in Massive IoT Environments,” in
2016.pdf. 2019 15th International Wireless Communications & Mobile
[94] Wonbin Hong, Kwang-Hyun Baek, Youngju Lee, Yoongeon Kim, Computing Conference (IWCMC), Tangier, Morocco, 2019, pp.
and Seung-Tae Ko, “Study and prototyping of practically large- 284–289.
scale mmWave antenna systems for 5G cellular devices,” IEEE [114] X. Zhou, R. Li, T. Chen, and H. Zhang, “Network slicing as a
Commun. Mag., vol. 52, no. 9, pp. 63–69, Sep. 2014. service: enabling enterprises’ own software-defined cellular
[95] M. K. Ishfaq et al., “Compact Four-Element Phased Antenna Array networks,” IEEE Commun. Mag., vol. 54, no. 7, pp. 146–153, Jul.
for 5G Applications,” IEEE Access, vol. 7, pp. 161103–161111, 2016.
2019. [115] A. Ijazet al., “Enabling Massive IoT in 5G and Beyond Systems:
[96] Y. Park, 5G Vision and Requirements, 5G Forum, Korea, Feb. 2014. PHY Radio Frame Design Considerations,” IEEE Access, vol. 4, pp.
[97] N. Javaid, A. Sher, H. Nasir, and N. Guizani, “Intelligence in IoT- 3322–3339, 2016.
Based 5G Networks: Opportunities and Challenges,” IEEE [116] M. Wang, J. Chen, E. Aryafar, and M. Chiang, “A Survey of Client-
Commun. Mag., vol. 56, no. 10, pp. 94–100, Oct. 2018. Controlled HetNets for 5G,” IEEE Access, vol. 5, pp. 2842–2854,
[98] X. Xu, D. Li, Z. Dai, S. Li, and X. Chen, “A Heuristic Offloading 2017.
Method for Deep Learning Edge Services in 5G Networks,” IEEE [117] “The Tech Wire Asia, ‘The next generation of IoT’, [Available on
Access, vol. 7, pp. 67734–67744, 2019. line 14 Jan 2018], http://techwireasia.com/2017/08/next-generation-
[99] D. Wang, D. Chen, B. Song, N. Guizani, X. Yu, and X. Du, “From iot/.” .
IoT to 5G I-IoT: The Next Generation IoT-Based Intelligent [118] L. Vangelista, A. Zanella, and M. Zorzi, “Long-Range IoT
Algorithms and 5G Technologies,” IEEE Commun. Mag., vol. 56, Technologies: The Dawn of LoRaTM,” in Future Access Enablers
no. 10, pp. 114–120, Oct. 2018. for Ubiquitous and Intelligent Infrastructures, vol. 159, V.
[100] M. Chen, J. Yang, Y. Hao, S. Mao, and K. Hwang, “A 5G Cognitive Atanasovski and A. Leon-Garcia, Eds. Cham: Springer International
System for Healthcare,” Big Data Cogn. Comput., vol. 1, no. 1, p. Publishing, 2015, pp. 51–58.
2, Mar. 2017.
[101] F. Al-Turjman, J. P. Lemayian, S. Alturjman, and L. Mostarda,
“Enhanced Deployment Strategy for the 5G Drone-BS Using
Artificial Intelligence,” IEEE Access, vol. 7, pp. 75999–76008,
2019.
[102] P. Kiss, A. Reale, C. J. Ferrari, and Z. Istenes, “Deployment of IoT
applications on 5G edge,” in 2018 IEEE International Conference
on Future IoT Technologies (Future IoT), Eger, 2018, pp. 1–9.
[103] R. Arridha, S. Sukaridhoto, D. Pramadihanto, and N. Funabiki, Kinza Shafique received the BE degree in
“Classification extension based on IoT-big data analytic for smart Electronics Engineering and MS degree in
environment monitoring and analytic in real-time system,” Int. J. Electrical Engineering from National
Space-Based Situated Comput., vol. 7, no. 2, p. 82, 2017. University of Sciences and Technology
[104] H. Uddin et al., “IoT for 5G/B5G Applications in Smart Homes, Karachi, Pakistan, in 2012 and 2015,
Smart Cities, Wearables and Connected Cars,” in 2019 IEEE 24th respectively. Since Jan 2016, she is working
International Workshop on Computer Aided Modeling and Design as a Lecturer in Electrical Engineering
of Communication Links and Networks (CAMAD), Limassol, Department, DHA Suffa University, Karachi,
Cyprus, 2019, pp. 1–5. Pakistan. Her research interests are in RF
circuits and antenna designing / optimization
for next generation wireless systems and body area networks. She is also
VOLUME XX, 2017 3
actively involved in the research related to RF energy harvesting techniques Dr Sameer Qazi received his B.E. degree from
for next generation Internet of Things (IoT) devices. National University of Sciences and Technology,
Pakistan, in 2001 and the MS and PhD degrees
from the University of New South Wales,
Australia, in 2004 and 2009. He is currently
working as Associate Professor and Head of
Electrical Engineering Department in the College
of Engineering at Karachi Institute of Economics
and Technology, Pakistan. His research interests are Network Traffic Matrix
Bilal A. Khawaja (M'08-SM'18) received Estimation, Non-Negative Matrix Factorization Architectures and
the B.S degree in computer engineering from Applications; UAV based video surveillance applications and Internet of
Sir Syed University of Engineering and Things (IoT). Previously, he has worked as Assistant Professor in
Technology, Karachi, Pakistan, in 2002, the Department of Electronics and Power Engineering, National University of
M.Sc. degree in communication engineering Sciences and Technology (NUST-PNEC), Pakistan and as Assistant
and signal processing from the University of Professor and Head of Electrical Engineering Department at DHA Suffa
Plymouth, Plymouth, U.K., in 2005, and the University, Pakistan.
Ph.D. degree in Electrical Engineering from
the University of Bristol, Bristol, U.K. in
2010. From 2003 to 2004, he was a Software
Engineer with Simcon International (Pvt.)
Ltd, Pakistan. From 2010 to 2016, he was an
Assistant Professor with the Electronics and Power Engineering
Department, PN-Engineering College, National University of Science and Muhammed Mustaqim received the BS
Technology (NUST), Karachi, Pakistan. In 2015, he was a Visiting Degree from the University of Central Florida
Postdoctoral Researcher with the Lightwave Systems Research Laboratory, (UCF), Florida, USA, in 2007 and MS degree in
Queens University, Kingston, Canada, involved in the Natural Sciences and Electrical Engineering from PAF-KIET,
Engineering Research Council (NSERC)-Canada CREATE Next Karachi, Pakistan in 2012. He is currently
Generation Optical Network (NGON) project on the characterization and pursuing his PhD. His research interests are in
measurements of 25GHz RF signal generation optical comb sources. He is Software Defined Radios (SDRs), Wireless
currently an Associate Professor with the Department of Electrical Networks, RF and Microwave Systems. Since
Engineering, Faculty of Engineering, Islamic University of Madina, 2010, he is working as a Lecturer at the
Madina, Saudi Arabia. He has authored and co-authored several journals Electronics and Power Engineering Department, PN-Engineering College,
and IEEE proceeding publications. His current research interests include National University of Science and Technology (NUST), Karachi, Pakistan.
next-generation of millimetre-wave (mm-wave) radio-over-fiber and optical
communication systems, mm-wave and THz signal generation mode-locked
lasers and RF transceiver design and antennas design/characterization for
the Wi-Fi/IoT/UAVs/FANETs/5G systems/UWB wireless body area
networks, wireless sensor networks, and millimetre-wave frequency bands.
He is an active reviewer for many reputed IEEE journals and letters.
Farah Sabir received the BE degree in

Electrical Engineering from DHA SUFFA
University, Karachi, Pakistan in 2017 and MS
degree in Electrical Power Systems from NED
University of Engineering and Technology
Karachi, Pakistan in 2019. Since Feb 2017, she
is working as a Lecturer in Electrical
Engineering Department, PAF-KIET
University, Karachi, Pakistan. Her research
interests are in Electrical power systems and
smart metering systems. She is also involved in
the research related to modelling and analysis of distance relays in FACTs
compensated transmission lines and Internet of Things (IoT) devices.
VOLUME XX, 2017 3
View publication stats

Internet of Things (Iot) For Next-Generation Smart Systems: A Review of Current Challenges, Future Trends and Prospects For Emerging 5G-Iot Scenarios

Uploaded by

Copyright:

Available Formats

Internet of Things (Iot) For Next-Generation Smart Systems: A Review of Current Challenges, Future Trends and Prospects For Emerging 5G-Iot Scenarios

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Internet of Things (Iot) For Next-Generation Smart Systems: A Review of Current Challenges, Future Trends and Prospects For Emerging 5G-Iot Scenarios

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

Internet of Things (IoT) For Next-Generation Smart Systems: A Review of

Article in IEEE Access · January 2020

Bilal A Khawaja Farah Sabir

SEE PROFILE SEE PROFILE

Sameer Qazi Muhammed Mustaqim

SEE PROFILE SEE PROFILE

The user has requested enhancement of the downloaded file.

Internet of Things (IoT) For Next-Generation

I. INTRODUCTION back to the 1980s with the idea of ubiquitous computing

VOLUME XX, 2017 3

VOLUME XX, 2017 3

Figure1: Block diagram showing (a) IoT based smart applications

VOLUME XX, 2017 3

Figure 2: Mobile devices subscriptions with respect to different wireless technologies

In a nutshell, the technical specifications currently become

VOLUME XX, 2017 3

TABLE I: IOT ENABLED SMART ENVIRONMENT

WLAN, WPANs, Energy-efficient remote monitoring and

VOLUME XX, 2017 3

TABLE II: COMPARISON OF PREVIOUS WIRELESS TECHNOLOGIES FROM 1G - 4G

3.5G 5-30Mbps Voice, Data, Video Calling 5MHz [64]-[65]

VOLUME XX, 2017 3

TABLE V: COMPARISON OF THE SPECTRUM WITH RESPECT TO DIFFERENT 5G USE CASES

Spectrum Type Characteristics Use Cases

Hence certain technologies must be involved in the

VOLUME XX, 2017 3

VOLUME XX, 2017 3

TABLE VI: ENHANCEMENTS PROPOSED FOR 5G ON CELLULAR IOT

Open Standards Enhancements Proposed

Also, the New Services and Markets Technology Enablers

VOLUME XX, 2017 3

VOLUME XX, 2017 3

VOLUME XX, 2017 3

VOLUME XX, 2017 3

Figure 4: Architectural Scenario Where 5G Meets Artificial Intelligence

Future intelligent systems require such IoT nodes to work

VOLUME XX, 2017 3

VOLUME XX, 2017 3

TABLE XI: TRENDING RESEARCH ON CHALLENGES OF 5G-IOT

Article Challenges Solution Reference

Extracting and exploiting Scarcity in spectrum resources to

Solutions for the wireless powering of IoT devices

The limited computing resource and

VOLUME XX, 2017 3

ACKNOWLEDGEMENT Overview, Assumptions, Problem Statement, and Goals,” RFC

VOLUME XX, 2017 3

VOLUME XX, 2017 3

VOLUME XX, 2017 3

Farah Sabir received the BE degree in

VOLUME XX, 2017 3

You might also like