Received 18 November 2021, accepted 21 January 2022, date of publication 1 February 2022, date of current version 30 March 2023.

Digital Object Identifier 10.1109/ACCESS.2022.3148473

Tactile Based Intelligence Touch Technology in

IoT Configured WCN in B5G/6G-A Survey
MANTISHA GUPTA 1 , (Student Member, IEEE),
RAKESH KUMAR JHA 2 , (Senior Member, IEEE), AND
SANJEEV JAIN 3 , (Senior Member, IEEE)
1 Schoolof Electronics and Communication Engineering (SoECE), Shri Mata Vaishno Devi University, Katra, Jammu, Jammu and Kashmir 182320, India
2 Department of Electronics and Communication Engineering, Indian Institute of Information Technology, Design and Manufacturing, Jabalpur (IIITDM
Jabalpur), Jabalpur 482005, India
3 Computer Science Department, Central University of Jammu, Jammu, Jammu and Kashmir 181143, India

Corresponding author: Rakesh Kumar Jha ([email protected])

This work was supported in part by the 5G and Internet of Things (IoT) Laboratory, School of Electronics and Communication
Engineering (SoECE), Technology Business Incubation Center (TBIC), Technical Education Quality Improvement Programme of
Government of India-III (TEQIP-III) at Shri Mata Vaishno Devi University (SMVDU), Katra, Jammu; and in part by the 6G and IoT
Innovation Laboratory, Indian Institute of Information Technology, Design and Manufacturing (IIITDM), Jabalpur.

ABSTRACT Touch-enabled sensation and actuation are expected to be the most promising, straightforward,
and important uses of the B5G/6G communication networks. In light of the next generation (6G) systems’
prerequisite for low latency, the infrastructure should be reconfigurable, intelligent, and interoperable in the
real-time existing wireless network. It has a drastic impact on society due to its high precision, accuracy,
reliability, and efficiency, combined with the ability to connect a user from remote areas. Hence, the
touch-enabled interaction is primarily concerned with the real-time transmission of tactile-based haptic
information over the internet, in addition to the usual audio, visual, and data traffic, thus enabling a paradigm
shift towards a real-time control and steering communication system. The existing system latency and
overhead often have delays and limitations on the application’s usability. In light of the aforementioned
concerns, the study proposes an intelligent touch-enabled system for B5G/6G and an IoT-based wireless
communication network, incorporating AR/VR technologies. The tactile internet and network-slicing serve
as the backbone of touch technology and incorporates intelligence from techniques such as artificial
intelligence and machine/deep learning. The survey also introduces a layered and interfacing architecture
with its E2E solution for the intelligent touch-based wireless communication system. It is anticipated for the
upcoming 6G system to provide numerous opportunities for various sectors to utilize AR/VR technology in
robotics and healthcare facilities to help in addressing several problems faced by society. Conclusively the
article presents a few use cases concerning the deployment of touch infrastructure in automation, robotics,
and intelligent healthcare systems, assisting in the diagnosis and treatment of the prevailing Covid-19 cases.
The paper concludes with some considerable future research aspects of the proposed system with a few
ongoing projects concerning the development and incorporation of the 6G wireless communication system.

INDEX TERMS 6G, AI, AR, intelligence, IoT, ML, network slicing, tactile internet, VR.

I. INTRODUCTION population. These, in the present day, are proficient enough

Mobile and wireless communications have a decisive to connect to the devices and people for an unprecedented
role in the current economy with the technologies like exchange of multimedia and data content. These services
2G,3G,4G,5G, GPRS, EDGE which successfully satisfies hence enjoy their fastest growth in history due to their
the user end. These have a significant role in the business, enabling technologies, encouraging its widespread deploy-
education, logistics, and other primary industrial applica- ment and further intensifying the communication and the
tions, while effectively connecting the majority of the world’s industrial sector [1].
As per the Cisco Visual Networking Index (VNI), there
The associate editor coordinating the review of this manuscript and is effective forecasting of the impact of the visual net-
approving it for publication was Mohamed M. A. Moustafa . working applications on global networks, incrementing

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME 11, 2023 30639
 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

FIGURE 1. General architecture of existing wireless and IoT-based scenario.

from about 11.5 exabytes in 2017 to an expected surge of Along the lines, the system also introduces extreme base
77 exabytes towards 2023. The compound annual growth stations with high device densities and an unparalleled num-
rate (CAGR) is expected to be about 74 percent of the present ber of antennas, as well as high carrier frequencies with large
mobile traffic, 66 percent of the cellular (Wi-Fi) traffic, bandwidths [4], [5]. From the onset of 2020 onwards and
accompanied by smartphones, dominating more than 90 per- till this date, the existing wireless communication networks
cent of the mobile data traffic in the coming few years [2]. have been standardized and deployed globally, with eMBB,
The Internet of Things (IoT) has therefore been a novel MMTC, and (URLLC) being the key 5G/B5G communica-
paradigm that is swiftly accelerating in the modern wireless tion scenarios [6]–[8]. The widely researched, IoT-enabled
communication scenario, connecting billions of devices with wireless network architecture, compatible with the B5G/6G
seamless access to the internet and applications, altogether system has been effectively illustrated in Fig. 1.
forming an internet of everything (IoE). Conclusively the The figure represents more or less every possible IoT
main strength of the IoT lies in the significant impact it configured service and application, requiring high data
has on a variety of facets of everyday life and user behav- rate and low latency, interacting with the surrounding
ior [3]. Since then, there has been a worldwide increase in environment. All these together constitute a smart sys-
the development of cellular networks over the past decade. tem. This smart system entails all the possible smart
Table 1 hereby provides a list of all the commonly used devices, gadgets, and sensors, all in all actuating a smart
acronyms throughout the paper for better understanding. and automatized infrastructure. This infrastructure, there-
Many technical challenges have instantiated the designing fore, highlights the utilization of the D2D communica-
of a robust wireless network, capable of delivering the neces- tion network [9]–[11], the massive MIMO[12], small cell
sary performance to support the emerging applications. The access points (SCA) [13], the IoT [14] with the network
previous generations have seen a paradigm shift in cellular cloud [15], [16], all in all forming a part of the said 5G/B5G
technology and unlike these, the B5G/6G is said to be an cellular framework [17], [18].
integrative structure of the present 5G air interface spectrum. All these along with the number of other sensor-based
This in combination with the LTE and Wi-Fi provides a interactions, make the existing system automatic enough to
seamless user experience, accompanied by the universally ease the human effort and save time, all the while consid-
high coverage rate. The existing 5G core network inculcates ering the exponential surge in the data traffic, operating the
unprecedented flexibility and intelligence in the upcoming millions of devices that are connected to the internet. The
6G system with an improved spectrum regulation, along with key technologies in this framework incorporate the spectrum
energy and cost efficiencies. sharing [19] with the cognitive radio [20], the interference

30640 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 1. List of commonly used acronyms.

management [21], the ultra-dense network (UDN) [22], the The smart applications constitute the entire automated
mm-wave [23], 5G/B5G cloud and RAN [24], [25] and smart city, comprising of the smart infrastructures in our
SDNs [26]–[28]. The existing 5G network, therefore, rides day-to-day lives, ranging from a smart traffic monitoring
on the coattails of an explicit New Radio (NR) interface system, with efficient V2V/V2X interactions, provid-
accompanied by a considerable number of virtualization tech- ing a competent sensor-based collision/accident detec-
nologies like the Network Function Virtualization (NFV), tion system to an equally competent smart parking
Software-Defined Networking (SDN), and Software-Defined system. The other applications may vary from the
RAN (SD-RAN) [29]–[32]. The market-driven allocation and smart health and education facilities like remote health
reallocation of bandwidth are the few efficacious parameters counselling, online smart classroom with the teacher-
in the 5G/B5G system [33], [34]. student interaction, the smart grid system, and the smart

VOLUME 11, 2023 30641

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

homes, collectively forming the smart neighbourhood area teleoperation with haptic communication like remote surgery.
network. It has therefore been an onset towards revolutionizing every
The virtualization facilitates an advanced computation of fragment of the society right from education and healthcare
the network resources and their allocation. These, reasoned with their prospected future applications varying from incul-
with their indispensable applications, facilitate a profitable cating sensations and sentiments using the intelligent robotic
proposition like network slicing [35]–[37]. The virtualization applications to the smart health facilities like remote surgery
in the core network (CN) [38]–[40] with the billions of mis- to the virtual shopping experience at the user end.
cellaneous IoT devices [14], [41], improves the integration of The applications evolving the modernized WSN like the
the past and present cellular and Wi-Fi standards. It, therefore, smart and automated homes and appliances, vehicles, facto-
provides a ubiquitous high rate, low latency, giving a smooth ries, remote sensing, and monitoring, augmented and virtual
experience to all the users in the network. reality(AR/VR) and quantum computing based applications
The prime objective of the existing communication stan- have an IoT as a common backbone, altogether forming an
dard has always been to fulfil the demands of increase in Internet of Everything (IoE) [48], [49]. The touch-enabled
capacity, the improvement in the data rate, the reduction in sensation and actuation are expected to be one of the most
the latency, to provide a better quality of service and experi- fundamental applications of the B5G/6G communication
ence (QoS and QoE). Therefore to meet these requirements, technology due to its potential, simplicity, and convenience,
drastic improvements have been made and are still ongo- taking into consideration the real-time scenario.
ing, to update the existing cellular architecture. Hence the For this reason, the ultra-responsive internet thus helps
next-generation network is said to be agile enough to revert enable real-time control of the physical tactile-based haptic
the intensified network complexity, regardless of handling devices, bringing in a paradigm shift toward an intelligent
diverse scenarios. and touch-enabled technology. ‘Why do we need a touch
The existing 5G/B5G networks thus have to rely on technology?’ is the most anticipated question here, taking
the self-organization and virtualization approaches, to deal into consideration all the previously-known aspects. The TI
with the disproportionate heterogeneousness and complex- although being the most researched domain of the B5G/6G
ity of the network, associated with the massive amount of framework, its exploration towards the onset of intelligent
devices [42]–[44].In the forthcoming years, intelligence in touch-based technology is still in its infancy and has thus been
the system is required to yield with such massive connectivity emphasized in this survey.
of IoT devices in the existing network with minimum latency
and network complexity. The B5G infrastructure thus has to A. SCOPE OF THIS SURVEY
focus on the considerable scaling and enhancement of the The opening gambit for such technological advancement
mobile network by incorporating open interfaces to support takes into account the subsequent technical update of the
vertical segments in the network [45]. wireless communication network with the contemporarily
Such vertical segments are most often the third parties that researched 6G system [50], [51]. It, therefore, encourages
do not own particular network infrastructure but require the a real-time interaction of humans with their environment,
networking services with their specific requirements, along with few instances like the actuation of sensors causing the
with their latest business solutions. Automotive manufac- tactile sensation and the real-time control/interface in our
turing has been one of the most notable vertical segments body system resulting from touching such surfaces.
in the existing communication system, requiring competent It, therefore, defines a new human-machine(H2M) inter-
networking capabilities combined with the IoT and edge- action framework enabling a physiological latency of human
cloud services. This in turn helps in the progress of several beings to build a real-time interactive system, with their appli-
applications like autonomous driving, bird-eye view, real- cations ranging from robotics to healthcare to autonomous
time assessment of road conditions, to name a few [46]. driving including the use of virtual/augmented reality
The mobile internet in B5G/6G will thus make provisions (AR/VR/XR/MR) [52]. For this reason, the ‘Tactile Inter-
for human-to-human (H2H) interactions with the primary net’ is regarded as an impetus and a cornerstone for the
goal of connecting the machines and gadgets to construct an deployment of touch technology and is expected to influence
IoT interface that is often built on D2D and H2H interactions. the development, innovation as well as revolution of health-
Therefore, to solve the drawbacks such as latency, poor data care, education, entertainment, manufacture, automation, and
rate, compatibility, high complexity, privacy, and security, smart grids.
the next generation reconfigurable IoT allows for a real-time This survey, therefore, paves a concrete path towards the
control labelled as the ‘Tactile Internet’ (TI). According to initiation of the intelligent and touch-enabled technology in
the ITU,1 it is a network that combines extremely low latency the B5G/6G and IoT-based wireless communication network.
with a high degree of reliability, scalability, and security [47].
TI here provides an improved and virtualized environment B. MOTIVATION AND CONTRIBUTION OF THIS SURVEY
which is most likely feasible for commercial applications like The main purpose of this paper is to present a comprehensive
and the state of art proposal motivated towards deploying
1 ITU: International Telecommunication Union. an intelligent and touch-enabled technology interface in the

30642 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 2. Comparative analysis of proposed survey with the existing surveys.

B5G/6G and IoT based wireless network. We first throw 3) We introduce intelligence in the B5G/6G network
light on the next generation 6G WCN describing its key with a descriptive mention of a few of the intelli-
requirements and applications. The survey then subsequently gence learning techniques like AI/ML/DL in wireless
emphasizes the role of network slicing in the B5G/6G net- communication.
work to sustain the high connectivity of a massive number of 4) We further provide a backdrop of the intelligent touch
devices without any latency and buffering at the user end. configured technology involving the orchestration of
The network may be sliced into Cloud and RAN services the network slicing with the tactile internet allied with
and applications at the customer end. It, therefore, enables the intelligence and IoT connectivity in the B5G/6G
easy access, storage facility, and virtualization, with each wireless domain.
slice serving different applications on the same physical net- 5) We propose a layered and an end to end (E2E) inter-
work. As a result, the network slicing and the Tactile Internet facing architecture enabling the touch technology inter-
in the B5G/6G are therefore the backbones for incorporating face in the B5G/6G domain.
a touch-enabled infrastructure with an induced intelligence 6) We further discuss the research challenges and further
by the AI/ML/DL implementation in the proposed system. explore the research areas of the next generation (6G)
Through this paper we propose here an intelligent deployment and ultra-low latency-based applications in
touch configured system that would be applicable in the 6G and IoT.
B5G/6G WCN. We hence summarize our contributions as
follows: C. COMPARATIVE ANALYSIS WITH THE EXISTING
1) We give a brief description of the B5G/6G commu- SURVEYS
nication system while throwing light on some of its Table 2 indicates a complete summary of the existing survey
key parameters and prominent use cases with their papers on network slicing and the tactile internet with some
applicability in the next-generation(6G) networks. of the intelligent technologies like ML/DL implemented in
2) We provide a discussion on the Network Slicing in the wireless cellular communication network in 5G/B5G sce-
B5G/6G and the Tactile Internet along with its appli- narios. In contrast to the other surveys, our survey provides
cations in the B5G/6G networks, which will act as an a comprehensive overview of the proposed intelligent touch-
enabler of intelligent touch-based systems. enabled technology in the B5G/6G and IoT configured WCN.

VOLUME 11, 2023 30643

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

Farris et al. [6], describes the essentiality of mobile edge application of the B5G/6G requiring the 1ms latency is Tactile
computing (MEC) for supporting a wide range of user- Internet (TI) which is quite efficient in establishing bilat-
centric applications. These have an important role in the smart eral communication between humans and machines, forming
city scenarios presented in Taleb et al. [53] where a smart an HSI, which ultimately enables a haptic communication
MEC-based architecture is significant for reducing the core system.
network traffic while guaranteeing an ultra-short latency for The term ‘Tactile Internet’ coined by Fettweis [46], has
the existing network. been a key enabler in fulfilling the need for higher data space,
Thus MEC here acts as a key factor in enhancing QoS and essentially resulting in a continuous increase in the storage
attaining the 1ms requisite latency for the 5G/B5G mobile and computation among various cellular devices connected
systems. It is accompanied by a considerable number of vir- to the internet. Therefore the TI is centered on H2M interac-
tualization technologies like NFV, SDN, and SD-RAN. The tions with the devices using the B5G and IoT connectivity,
virtualization in turn promotes an advanced computation and enabling haptic and tactile sensations at both the transmitting
allocation of the network resources. Myriad existing studies and receiving end forming a bilateral communication and
have provided a comprehensive overview of the technical feedback network. The works of Maier et al. [52] highlight
challenges and applications associated with network slicing, the commonalities and various subtle differences between the
TI, and the emerging intelligence in B5G/IoT systems. TI, IoT, and the B5G vision.
The works [39], [45], [54] provide comprehensive reviews The authors in Simsek et al. [47] highlight the critical
of the literature on network slicing in B5G networks. requirements and architectural approaches for TI, as well
Richart et al. [39] discusses resource slicing and their allo- as the technical issues and challenges associated with the
cation in virtual networks powered by SDN and NFV, as well resource management, core networking, and edge cloud/ AI
as how these can be distributed appropriately to network capabilities. Aijaz et al. [60] help to examine a number of
slices without impairing the efficiency of other slices. While the stringent design challenges to revolutionize the tactile
Afolabi et al. [45] describe the state of art network slice internet, providing an enhanced haptic perception with a 1ms
life cycle architecture operating across the multiple domains round trip delay. Authors Antonakoglou et al. [61] explain
thereby enabling an effective network programmability and the evaluation of methodology and technology assessments
flexibility with the creation, management and orchestration of for the necessary haptic communication infrastructure.
the network slices, utilizing the massive IoT and multimedia The examination of the advancements in teleoperation over
broadband connectivity. long distances affects haptic communication while using the
Foukas et al. [54] review the concept of network slicing Tactile Internet. Therefore as per [62], the Tactile internet is
and propose a generalized layered architecture consisting of a key enabler for realizing the industrial revolution(4.0) by
an infrastructure layer, a network function layer, and a service users and devices in real-time. The work by Sun et al. [63]
layer, along with their associated benefits and challenges. The explains how machine learning may help with resource con-
existing progress in the B5G network slicing with its key trol at the MAC layer, network and mobility management in
trends along with their corresponding potential challenges the network layer, and application layer localization.
are presented in [55]. The authors in [56] throw light on the While Wang et al. [64] give an overview of ML/AI
various concepts of network slicing and softwarization7 in technologies while addressing issues like channel model-
the B5G technology with their applicability across the RAN ing, estimation, network management, and optimization in
and core network, altogether establishing an E2E slicing B5G wireless communication networks. Following which
infrastructure. Bega et al. [65] encourage exercising the ML approach
Intelligent tools like AI/ML/DL play a significant role in towards market optimization while maximizing infrastruc-
outlining automation, deployment, and disposition of dif- ture provider monetization. Hence to thoroughly optimize
ferent applications the existing as well as the next gener- the availability of the computing resources, authors in [66]
ation networks (B5G/6G) [57]. The work of Mohammadi present a novel tactile-based flexible next-generation internet
and Al-Fuqaha [58] provides an intensive facet of the most architecture (FlexNGIA) that capitalizes on the coexistence
prevalent deep reinforcement learning in the structuring of of transport and network layers to provide better congestion
the cognitive smart cities and its applicability concerning the control and reliability services via cross-layered network
energy consumption with garnering of water and agricultural computing.
utilization. You et al. [67] give an insight into the next generation
Kafle et al. [59] successfully highlight the various ways 6G communication highlighting its parameters specifically
of applying AI/ML techniques for the automation of network emphasizing its coverage requirements for the functional-
functions in different configurations, ranging from develop- ity in terrestrial as well as non-terrestrial environments.
ment organizations to industrial forums. One such intelligent The paradigm aspect of this work is the integration of
the space-air-ground and sea-based communication network.
7 Network softwarization is the notion of designing, architecting, deploy-
Alwis et al. [68] discuss several 6G use cases, including holo-
ing and administrating network components, largely based on software
programmability properties. It enables flexibility, adaptability, and even total graphic telepresence, e-health, and in-body networks that
reconfiguration of a network on the spot [37]. require extremely high data rates, ultra-low latency, and

30644 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

autonomous agents in virtual environments will distinguish

future communication and industrial systems.
The remainder of the paper follows the structure depicted
in Fig. 2. In Section II and III, we discuss the next-generation
6G system in detail with its main parameters, service, and
applications along with network slicing and virtualization
concepts in the 6G system. Section IV delves into the archi-
tecture and state-of-art uses of the tactile internet, which is a
crucial component of the touch interface system. Section V
introduces the intelligence that is to be incorporated into the
next-generation wireless network.
Section VI explains the proposed intelligent touch tech-
nology, including its layered and E2E architecture in
Section VII and VIII. Section IX discusses some basic major
components of the proposed system, while Section X
presents an interfacing architecture with the existing network.
Section XI contributes some potential and applicable use-
cases of the proposed system, while Section XII highlights
some of the research challenges and upcoming future aspects.
Section XIII summarizes the recent research and a few of the
ongoing projects on 6G before concluding in Section XIV.

II. INTRODUCTION TO NEXT-GENERATION

TECHNOLOGY (6G)
The 6G communication era anticipates how humans will
engage with digital virtual worlds beyond 2030, as the pro-
jected digital transition with the existing B5G networks has
already begun and will continue to evolve over the next
decade. Although 5G/B5G is recognized for network cloudi-
fication via microservice architecture, the next generation 6G
network is strongly linked to intelligent network orchestration
and management. New digital virtual worlds with connected
intelligence must have novel technologies that support these
communication and networking challenges beyond 2030.
The 6th generation wireless communication network is
anticipated to consolidate terrestrial, aerial, and maritime
communication into a robust network that would be more
reliable, faster, and capable of supporting a large number of
devices with ultra-low latency requirements while remaining
cost-effective. Due to the exponential growth in the number
of IoT devices, the next generation systems must achieve
high spectral and energy efficiency (SEE), low latency, and
massive connectivity to provide for services like smart traffic
monitoring, VR navigation, telemedicine at the user end,
along with digital sensing using a full HD video transmission
in connected autonomous devices like drones and robots [69].
The timeline in Fig. 3 shows the evolution of the wireless
communication network from 1G to the recent 6G. The next
section explains why the shift from 5G to 6G is necessary.

A. TRANSITION FROM 5G/B5G TO 6G

FIGURE 2. Structural organization of the paper.
As 5G/B5G networks are consistently deployed, the inherent
limitations of this system are being exposed, in comparison
to its original assertion as a platform for IoE applications. It is
high reliability. As a result, the continuous penetration of gradually more difficult for the existing multiple access tech-
mobile platforms, robotics, human-computer interaction, and niques, to cope with the exponentially growing IoT devices.

VOLUME 11, 2023 30645

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

FIGURE 3. Timeline of wireless communication technology evolution from 1G to 6G.

As a result, the 5G communication systems, already being Therefore, to successfully operate these IoE services,
implemented in the world today are incapable of supporting a wireless system must simultaneously provide a high level of
these many IoT devices. The requirement for faster data reliability, low latency, and a high data rate for a wide range
rates has fueled the evolution of wireless networks, which of heterogeneous devices. These new services necessitate the
has necessitated a continuous 1000-fold increase in network resolution of novel and distinct challenges, unprecedented
capacity [70]. in terms of their complexity including the tradeoff between
As the demand for wireless capacity continues to surge, the latency, throughput, and reliability. Not only do these ser-
emerging IoT system, which connects millions of people to vices help entail new approaches for effective regulation and
billions of machines, has resulted in a radical paradigm shift handling of performance and challenges but also aid in the
from the rate centric eMBB services from the previous eras exploration of frequencies beyond the 6GHz range to create
towards URLLC and intensified mMTC services, as per the a self-sustaining and intelligent wireless network.
3GPP, which is working on the implementation of 5G/B5G This aforementioned network is capable of provisioning
standard [71]. Although it can be asserted that the evolution- and orchestrating communication, computing, control, local-
ary aspects of the existing 5G supporting the data-hungry ization, and sensing of the scenarios that are best suited
eMBB services have gained significant momentum, while the for IoT needs [73]. So to address these issues, a game-
promised revolutionary disposition systems, operating exclu- changing 6G wireless system is required, with a design that
sively at high mm-wave frequencies have yet to materialize. is organically tuned to the performance requirements of IoE
Even though today’s linked 5G systems are easily capable applications and associated technical advancements.
of supporting the most fundamental IoE and URLLC services
(such as factory automation), it is still debatable whether or B. AN OVERVIEW OF THE B5G/6G COMMUNICATION
not they will be able to deliver the smart city services based SYSTEM
IoE applications in the future. Conversely, the initial and the The 6G wireless communication network, which is cur-
existing B5G deployments are most likely to rely on the low rently being researched, is expected to integrate terres-
frequencies (sub 6GHz) to support mobile data transmissions. trial, aerial, and maritime communication systems into a
While on the other hand, an enormous influx of new IoT robust network that is more reliable, fast, and capable of
services such as XR (including AR/VR/MR), flying vehicles, supporting a large number of devices with ultra-reliable
and connected autonomous systems would most likely derail and low-latency requirements. The AI, ML/FL, quantum
5G’s original purpose of supporting small packet and sensing- communication, blockchain/DLT, beyond 6GHz and towards
based URLLC applications [72]. Terahertz communication, TI, swarm UAVs, Zero-touch

30646 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 3. Comparison of performance attributes between 5G, B5G, and 6G communication systems.

network and service management (ZSM), large intelligent with cell less architecture, decentralized networking, and
surfaces (LIS), NTN and 3D networking, VLC, compressive resource allocation with 3D radio interoperability. The next-
sensing with an efficient energy transfer and harvesting are generation(6G) wireless network comprises a large number of
just a few of the currently proposed by the ongoing researches linked devices with numerous base stations (BSs) and access
worldwide [74]. points (APs), each of which will serve multiple devices at the
Owing to massive growth in the number of IoT devices, same time, forming a coordinated multipoint (CoMP) [76].
a realization of advanced services like smart traffic mon- Hence more data would be transmitted via future wireless
itoring, VR-based navigation, smart medical facilities like communication networks since most of the value-added apps
telemedicine, and HD video transmission in drones and and services rely significantly on the data exchanges. There-
robots is possible. Hence the B5G/6G communication sys- fore large devices will generate a massive quantity of data,
tems aim to achieve high SEE, low latency, and massive con- which will need high-performance processing units and back-
nectivity. The ever-increasing number of IoT devices makes it hauling connections. The following subsections go through
difficult for the existing multiple access strategies to handle the new 6G enabled driving trends, metrics, and use service
such a huge number of devices, therefore requiring a more cases.
extensive network to make use of the massive bandwidth
capabilities offered by B5G/6G communication systems. C. 6G ENABLED SERVICE CASES, METRICS, AND DRIVING
According to the work by Zhang et al. [75], a specu- TRENDS
lated vision and functionality of the 6G networking scenario With the new performance metrics, new technological trends
provides a technological framework and requirements for step up to redefine the prevalent B5G applications by morph-
industries in the future generation communication system ing the classical URLLC, eMBB, and mMTC into something

VOLUME 11, 2023 30647

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 4. Comparison of various 6G Service cases with parameters and application areas.

entirely new and innovative. As a result, Table 3 gives a communications (ERLLC) [83], ultra-massive machine-type
comparison of some of the key performance attributes of the communications (umMTC) [44], long-distance and high-
5G, B5G, and 6G wireless communication systems [73]. Var- mobility communications (LDHMC) [84] and extremely low
ious countries have initiated projects aiming at the research power communications (ELPC) [85]. Detailed descriptions of
and deployment of B5G/6G communication networks, as each of these are provided in the preceding Table 4, along with
discussed in Section XIII. The research on 6G is accelerating their parameters and application areas.
and has been documented in recent works like [77]–[79].
According to the recent research, a variety of possible III. NETWORK SLICING IN B5G/6G BASED IOT
6G applications have been classified as mobile broad band- NETWORKS
width and low latency (MBBLL) [80], massive broad band- The forthcoming B5G/6G networks aim to serve a wide
width machine type (mBBMT) [78], massive low latency range of applications, thus recognizing the 5G epoch as the
machine type (mLLMT) [81], further-enhanced mobile century of mobile telecom networks, all the while promot-
broadband (FeMBB) [82], extremely reliable and low-latency ing dedicated use-cases and endowing unequivocal services,

30648 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

FIGURE 4. Network slice functionality in B5G/6G network.

to meet diverse user requirements. Hence the technologies A set of homogenous APIs is made available for creat-
like UHD, multimedia, AR/VR/XR, therefore, need a faster ing an abstraction layer to facilitate the slice management
speed and a relatively higher capacity and connectivity, com- while controlling its virtual resources during its operation.
pared to the mission-critical applications like IoT/MIoT and These slices can therefore be accessed by different tenants
autonomous systems that require ultra-low latency and ultra- or third parties using these APIs. Here SLA acts as the slice
reliable operation. blueprints, using which the tenant specifies its requisite slice
The 6G cellular framework formulation anticipates being characteristics ranging from topology, management, control,
accomplished on the existing and researched 5G/B5G tech- and so on. The slice lifecycle is regulated by the service life-
nology, thus supporting a surplus of network services with cycle management loop, openly accessed by all the function-
miscellaneous performance requirements. The advancement ing slices [45]. The management and orchestration (MANO)
of cellular networks and resulting generation-wide improve- offer an integrated and holistic approach towards the regula-
ments is motivated primarily by the desire to enable better tion of network slicing and NFV management.
data-based services. A variety of aspects render 5G important, It offers a standardized level of data abstraction followed
including the mm-wave spectrum distribution and realloca- by the adapt specification of its network infrastructure
tion of bandwidth. Virtualization with a billion individual net- together with its service management and implementation
works in the CN and IoT facilitates the convergence between process [31]. This section discusses the concept of net-
previous and present cellular and Wi-Fi requirements [83]. work slicing in B5G/6G communication networks with its
This in turn offers a pervasive high rate and low latency functionality, management, and orchestration in RAN and
experience for network customers. The most significant part CN and finally its application in the 6G with the proposed
of the B5G/6G infrastructure comprises the network service architecture.
and its development platform. It is highly capable of improv-
ing network scalability while fulfilling the user requirements A. SLICING PROCESS IN THE B5G/6G NETWORK
by utilizing existing services. The network slicing function- The existing B5G is most likely to consider a variety of
ality in the B5G/6G domain has been depicted in Fig. 4. business and service quality requirements like the enhanced
The virtualized infrastructure here has provisions for slice capacity coupled with the intelligent traffic and offloading
instances and collectively functions with the infrastructure techniques accompanied with a highly complex and hetero-
resources from another slice instance [37]. geneous network. All of these fulfil the required performance

VOLUME 11, 2023 30649

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

FIGURE 5. Network slice orchestration and management.

criterion together with autonomous network management. A These applications may range from automated vehicles,
high data rate guarantees a high level of end-user service tactile applications, smart devices, and so forth [87]. Fig. 5
quality with unlimited mobile broadband connectivity in jam- diagrammatically illustrates the network slice orchestration
packed areas like stadiums, concerts, and shopping centers, process applicable in B5G/6G networks. At the back end,
employing the terminals having AI capabilities [86]. the resource allocation takes place in the infrastructure layer
The reduced latency with high data rates is capable of where resources are provided to the individual slices. The
supporting the UHD streaming from the cloud technology slicing applications are managed by the network orchestrator
and improvised VR devices and other wearable computing in the service layer, to enable effective application slice man-
gadgets. It, therefore, provides a faster web downloading agement and orchestration. It can be undoubtedly claimed
while enabling a premium user experience in services like that the most defining feature and in other words the ‘secret
YouTube streaming, Netflix, and so on with a high video sauce, for the 5G/B5G success is the E2E network slicing,
resolution. Network slicing is a fundamental key for the which will be applicable in 6G networks as well.
B5G/6G technology. Thus the network slices here are an end- Hence network slicing is responsible for the optimal
to-end concept of the next-generation technology where the resource efficiency and flexibility in the network. It, there-
slice operator supports a massive amount of customers. Here fore, enables the implementation of new business models as
each of them in the long run requires multiple end-to-end NaaS, supporting various mission-critical use cases includ-
individual and logical networks referred to as network slices. ing industrial automation(4.0) and availing of remote health
They are categorized into three components, each of them facilities [50], [88]. The network slicing architecture of the
governing the RAN, core, and transport domain [56]. B5G/6G system has been illustrated in Fig. 6. The slicing
The RAN and core slices consist of the application context process is therefore described as the three sub-processes,
and personalities respectively with the transport slices being interlinked with each other: the intelligent cloud slicing, the
connectivity between the RAN and core. Each domain has RAN slicing, and the application slicing, all of which are
a controller, i.e. the RAN controller, core controller as well functional in the B5G/IoT enabled networks.
as transport controller, all of these supported by an end- Commencing from the lowermost stratum we have various
to-end orchestrator. To realize the B5G/6G networks, 5G applications that are furnished at the consumer end ranging
network slicing plays a crucial role in guiding slice utiliza- from real-time online gaming with a UHD streaming to the
tion for automation, assurance, and optimization of trans- live online classroom teaching sessions, efficiently making
port slices involving various low latency and high-reliability use of the AR/VR technology to deliver the required informa-
applications. tion. The other applications may also involve the use of robots

30650 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

The following section describes the tactile mode of com-

munication in the B5G/6G system that is to be incorporated
with the aforementioned slicing techniques, to deploy and
configure the proposed touch interfacing wireless system
with incorporated intelligence which is discussed in detail in
the subsequent sections.

IV. TACTILE INTERNET IN B5G/6G SYSTEM

The internet initially was designed and intended to be a
reliable and interoperable means of communication across
the globe. With time, not only has it evolved to convey a large
amount of content, but it also helps enhance the real-world
experience of the users. The tactile internet thus improves
remote real-time physical engagement with existing and vir-
tual items. Whereas integrating TI with a touch interface
will provide a two-way interactive experience that blurs the
boundaries between the real and the virtual world. The works
in [89]–[92] examine the technical requirements of TI and its
ability to support future smart city applications. It, therefore,
provisions for a real-time control with the physical and haptic
sensations to be experienced by the users remotely.
Unlike the traditional internet applications, the new tac-
tile internet (TI) intends to serve as a medium for remote,
real-time physical interactions with actual, physical objects,
including humans, machines, and processors. The tactile
internet enables enhanced virtualized remote classroom
instruction with the participants in collaboration with the
remote environment collectively amid the presence of remote
and virtual resources. Thus, the data rates of wireless com-
munications have been increasing, which is primarily due to
innovation in electronics and the latest communications tech-
nology, including text messaging, video streaming, emails,
and file sharing.
The TI is centered on the H2M interactions with the haptic
devices. To facilitate haptic communication, the transmission
of data via tactile internet creates a network that is both
FIGURE 6. Network slicing in next-generation networks (B5G/6G). extremely reliable and extremely responsive. Haptic interac-
tion is a type of interaction that includes the use of remote
touch. It refers to the kinesthetic perception of information
in real-time actuation and tactile and touch-based haptic com- conveyed by the muscles and joints of the body via force,
munications, which may be put to practical use in industrial torque, position, and velocity, as well as the tactile perception
operations, automation in the robotics and machinery, vehi- of information conveyed by the mechanoreceptors of the
cles, and UAV fleet to accomplish the services requested by human skin via surface texture and friction. Sharing infor-
the users at the customer end in the form of application slices. mation through kinesthetic mode facilitates a global control
The succeeding layer is that of RAN slicing, taking place loop with stringent latency requirements while containing
in between the CN and RAN, i.e., at the backend of the feedback of generally audio/video type [93].
network, routing the clients with the final applications. It is Now with the enabling of haptic data, the TI enables a
capable of enabling an effective resource and spectrum allo- networked control system (NCS) supporting the connected
cation with power and energy-efficient cognitive radio net- sensors and actuators while controlling highly dynamic pro-
work system. All of this is in the form of RAN slices and cesses. Therefore it helps in digitally transmitting the sense
successfully connects the edge devices with the cloud net- of touch from one place to another, facilitating the URLLC
work. Cloud computing and storage enable intelligent cloud network in the B5G/6G system. In other words, the tactile
slicing techniques, where the cloud-enabled applications are internet’s purpose is to provide a remote and dynamic way
accessed in the form of cloud slices. These support different for people to experience physical haptic or touch-based con-
edge devices at the user end while facilitating the requested trol while exchanging closed-loop information between the
services. virtual and physical worlds [90].

VOLUME 11, 2023 30651

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

FIGURE 7. Tactile internet E2E architecture with a bilateral feedback system.

Wireless communications can thus be a medium for con- interface, provided by the gyroscope and the modern touch
trolling and directing real and virtual objects using such a screen technology, which has been a welcome step that could
platform. This revolutionary technology continues to trans- drastically transform the way we connect.
form healthcare, transportation, education, logistics, smart Additionally, we have an instinctive (or innate) sense of
grid systems, and many more, hence covering a major por- our surroundings, which has a tactile understanding of the
tion of the economic sector in society. It thus provides sub- real-time connection with our world. The tactile feedback
millisecond connectivity for the healthcare applications like and a phone give our hands the sensations, which in turn
remote surgery. enable our whole system to be modulated with its proximity.
This section draws attention to one of the most popular Whereas inside the vehicle it modulates several sensors and
applications of the B5G/6G communication system: ‘Tactile controls elucidating a real-time human-machine communica-
Internet’ and helps review its parameters, applications, and tion interface. Thus all TI needs is highly responsive, smart,
the basic architecture while focusing on its application in the and reliable connectivity to provide a medium for intelligent,
intelligent 6G and IoT systems. real-time touch, control, sensing, and operation.
With the high availability of TI, accompanied by its very
A. INTRODUCTION fast reaction time and reliability, the human interaction with
The ‘Tactile Internet’, as coined by Fettweis, has been a the machines enables a new dimension by creating an inter-
catalyst for economic development and creativity and in active, real-time system, which revolutionizes almost every
bringing a new stage of maturity in adapting technologies for segment of society [94]. Taking into account the industrial
a changing global environment [46]. Given that cellular com- dimension, TI is an interconnected system of specialized
munications connect a vast majority of people worldwide, components and applications used in industrial environments
it is therefore imperative to connect the technology as well. to monitor and operate physical equipment. Hence the works
According to the IEEE P1918.1 working group, the Tactile in [62], [89], [92] contribute to the role of TI in industrial
Internet may be defined as a network of networks that can be systems by examining its potential in emerging and future
accessed, perceived, and manipulated by people or machines industrial sectors.
on a remote, real or virtual basis in real-time perceptions [94]. Against this backdrop, the goal of this study is to identify
The tactile internet application offers the standard required and address the cutting-edge challenges to implementing an
latency needed to guide and control real and virtual objects intelligent touch-enabled system via tactile internet in the
without causing cybersickness, revolutionizing education, perspectives of B5G/6G-based wireless and IoT communi-
accessibility and traffic, healthcare, athletics, culture, games, cations networks. Thus, tactile internet serves as a key to
and the smart grids, thereby profoundly shaping our com- realizing the vision of ‘‘Touch Technology’’. The TI has an
munity. The i-phone for instance has an astounding haptic expected potential and future scope through the increasing

30652 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

penetration of mobile and cell platforms with the robotic- 3) THE NETWORK OR CONTROLLER: The network
human and computer interactions in the virtual control envi- domain kinesthetically integrates the person with its
ronments, for interconnecting people, machines, appliances, surroundings and distant environment.
and processes in real-time. 4) TACTILE SUPPORT ENGINE Being on the edge net-
To accomplish this, we present our discussion with an work, it effectively offers AI capabilities that are crucial
overview of haptic communication via tactile internet archi- in system stabilization.
tecture, with an emphasis on its potential development in the 5) HAPTIC DEVICE: Enables tactile communication,
touch-enabled framework in 6G and IoT mobile networks. which means a user may touch, feel, and engage with
virtual or real-world things.
B. TACTILE INTERNET ARCHITECTURE As a result, the most common design for a tactile haptic
The haptic or the touch sensation helps ascertain a connection device is a linkage-based system that consists of a robotic arm
between the humans and the peripheral environment in a way connected to a stylus and is capable of applying force on its
analogous to the auditory and visual senses. It, therefore, tip. Thus, the growing degree of freedom (DOF) is an essen-
occurs bilaterally as a touch, sensed by imposing a motion tial phenomenon for the envisaged applications that integrate
or a movement in an environment by some reacting force. the network interface in a direct and indirect connection with
Haptic communication thus provides an additional dimension a cellular network. Since they both include sensations rather
and advantage over traditional audiovisual communication than conventional multimedia, it is essential to differentiate
for real-time control and accessibility in distant and remote between the tactile internet and haptic communication, which
environments. The tactile internet architecture consists of a are comparable to traditional multimedia like speech, data,
radio access network (RAN) and a core network (CN), both and video.
of which are expected to meet critical requirements for TI To summarize, haptic communication networks include
functionality realization. communications across the wired and mobile internet, as well
Each of the three domains in the tactile E2E architecture as applications that operate on the tactile internet. This means
can be separated into three sections: the master domain, that the haptic communications and the tactile internet have a
the network domain, and the controlled (slave) domain. The service and medium connection with each other, respectively.
master framework comprises an operator, either human or
machine, and an operating system interface. Using vari- V. NEED FOR INTELLIGENCE
ous coding techniques, this interface acts as a master robot The existing and the future B5G/6G wireless network are
or as a controlling device, converting the operator’s input expected to endow the users with improved coverage and
into tactile input. If the controlling device is haptic, it lets high data rates with better cost efficiency, resource uti-
humans interact with objects in virtual or real environments lization, scalability, adaptability, and security. Hence the
through physical means such as touching, feeling, manipulat- B5G/6G wireless communication system is anticipated to be
ing, or controlling. a backbone of the digital revolution in the next generation net-
It is primarily responsible for the controlled domain’s oper- work providing a ubiquitous reliable and practically instan-
ation. In the case of a network-controlled system, the master taneous connectivity for humans and machines. Artificial
domain includes a controller that issues command signals Intelligence (AI) may be regarded as the ‘processing and sim-
to the sensor or actuator system. A domain, where both ulation of the human intelligence by machines’ and therefore
robotic machines and other objects are at distant locations in a has a great potential in working out several intractable and
controlled environment, and directly accessed by the master unstructured problems containing a large amount of data [84].
domain via command signals is referred to as a controlled In other words, AI may be defined as a science of con-
domain. When remote operations are carried out via haptic structing computers that are capable of performing tasks
feedback, energy is transferred between the master and the requiring human-like cognitive intellect [95]. All the while
controlled domains, and the global control loop is completed. AI has therefore been a widened approach for the machines
As apparent from Fig. 7, the components in the E2E tactile to be able to smartly carry out assigned tasks. ML on the
internet architecture are explained as below: other hand is presently the widely accepted application of
AI, empowering the machines to train and learn from large
1) MASTER DOMAIN: Generally, an HSI/HMI8 is a
datasets and perform tasks without any need for explicit
robot where the user may touch, feel, and move virtual
programming. Next in order is deep learning (DL), a subset
and real-world items while directing the actions in the
of ML that analyses the artificial neural networks (ANNs).
slave domain.
These have more hidden layers to replicate the human brain,
2) SLAVE DOMAIN: The slave domain is controlled
making it one of the most widely used ML techniques. DL,
by a teleoperator, through different command sig-
therefore, has a lucrative application in fields like computer
nals from the master domain, interacting with its
vision, bioinformatics, and speech recognition.
surroundings.
Such induced intelligence in the wireless communica-
tion network not only reduces the manual effort in net-
8 HSI/HMI: Human System Interface or Human Machine Interface. work deployment, configuring, and management but also

VOLUME 11, 2023 30653

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

helps in improved system performance. It also asserts the combined with the incorporation of machine learning (ML)
adaptability and reliability of the communication network techniques with reasoning and decision-making mechanisms.
by taking robust decisions in real-time according to the Accordingly, the development of an intelligent touch-based
prediction and behavior of the users and network. Hence due system calls for a promising development in facilitating the
to the recent advances and research in intelligence (AI/ML), efficient resource distribution in the cloud, fog, and edge
a wide range of novel technologies like self-driving cars, nodes, aiming to put together system intelligence and data
voice assistants, holographic telepresence, e-health and well- processing abilities in close proximity with the original data
ness applications, pervasive connectivity in smart environ- source.
ments, industry 4.0 applications, massive robotics with the
unmanned mobility in 3D, AR/VR have become possible. A. MACHINE LEARNING IN B5G/6G WCN
6G wireless networks thus offer a broadband network, ML as a member of the quintessential AI technology has
which is fast, instantaneous, and safe, to enable mass data nowadays become a key component of the 6G wireless com-
exchange at various frequencies using a wide range of tech- munication network. ML is said to be of a plausible advantage
nologies. In addition, these technologies are moving towards in the communication system since the dynamic nature of the
intelligent devices in IoT that will demand more reliable, wireless communication channels complicates the channel
stable, efficient, and secure connectivity. Hence the complex interference models in the B5G scenarios. Therefore ML
connected devices, therefore, require a dynamic communica- techniques are capable of extracting information from the
tion network to address their inherent complexity. The future unknown channel by learning from the communication data
wireless networks will eventually need a self-organizing and while taking into account previous learning experiences. Fur-
configuring capability alongside their cooperation and coor- thermore, the rapidly growing number of wireless access
dination between the different nodes and communication points necessitates a global optimization of communication
layers. resources as well as a fine-tuning of the system design [73].
It enables us to effectively meet challenges like coverage, However, the existing approaches together with the mas-
spectrum, and energy efficiency. The continuous acceleration sive amount of resources complicate the tasks concerning
of the machine type communication (MTC) devices adds the optimization and correlation of the system parameters.
to the existing ultra-dense network’s complexity. Therefore In contrast, advanced ML techniques like deep learning and
many such applications supported by the B5G network must probabilistic learning can represent highly nonlinear relation-
attain a short transit time and low latency with high reliability, ships and aid in the determination of optimal system parame-
availability, and security. The majority of these are resource- ters. Sequentially, ML aids in the realization and installation
constrained and unable to rely on their bounded resources of learning-based adaptive network configurations by iden-
and thus call for an uninterrupted and safe operation as their tifying and evaluating their behavioral patterns in advance
main concern. Consequently, these applications owing to rather than reacting to unanticipated outcomes [63].
their delay and bandwidth constraints cannot be moved to the As a result, the current cellular networks that were built
cloud or network controller [96]. and managed based on the preceding premise may be unable
Furthermore, these devices generate a diverse range of to keep up with the growing complexity of data produced and
datasets with a large scale of erroneous or missing val- therefore fail to provide the necessary capacity, dependability,
ues. Many wearables with VR/AR, intelligent products and and flexibility. Now as a response, the network may be unable
support systems, and other data-hungry use cases have a built- to respond fast enough to expected occurrences, thereby
in end infrastructure to afford and deliver content-based ser- compromising real-time communication services. Because
vices. Thus to incorporate an intelligent system, an intelligent the majority of AI/ML algorithms are not purpose-built for
and content-aware approach must be implemented for the wireless communication networks, it is difficult to apply them
planning, design, analysis, and optimization of such a net- directly to the B5G/6G networks.
work. This necessitates an integration of the network systems All the above arguments call for an intelligent communi-
with their data sources, decision making, and cyber-physical cation interface in real-time, facilitating stable and efficient
infrastructures, as well as sensing and communication net- connectivity within the network. The 6G wireless communi-
works [97]–[99]. cation system guarantees a wide range of frequency bands,
Conclusively the favourable conditions for the implemen- including sub-6GHz, mm-wave, THz, and optical bands,
tation of the intelligent learning techniques in 6G wireless while also increasing the computational complexity in the
networks range from: channel model. As a consequence, comprehending new chan-
nel characteristics for modeling new channel scenarios is a
1) Network interoperability with the distribution, afford-
lengthy process.
ability, and accessibility of computing resources.
Owing to the significantly high computational channel
2) The predictive nature of the network characteristics.
complexity in many situations, traditional techniques may
3) Accessibility of a considerable amount of data.
aid in certain approximations and assumptions to help sim-
Therefore a densely integrated wireless network may be plify the channel modeling and processing methodology.
engineered by adopting artificial intelligence (AI) principles This ensures the balance between accuracy and complexity

30654 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

tradeoffs of both channel modeling and processing methods

is beneficial.

B. GENERALIZED WORKFLOW PROCESS FOR MACHINE

LEARNING
For most prevailing and functional machine learning algo-
rithms, the generalized workflow process in basic steps is
described and shown in Fig. 8 [100]:
1) PROBLEM FORMULATION: Since the ML training
process is time-consuming, the problem must be appro-
priately formulated at the start of the process. More-
over, there should be a strong correlation between the
problem and the information gathered. Classification,
clustering, and decision making are three important
types of machine learning. The proposed model should
also be considered within these three categories to aid
in the identification of the learning model as well as
data collection. Improper problem formulation results
in an unsatisfactory learning model and performance.
2) DATA COLLECTION: There are two types of data
collection: offline and online data collection, where
data collected in real-time is used as model feedback
in online data collection and is also used for re-training FIGURE 8. Machine learning basic workflow process [100].
of models. In contrast, offline data may be retrieved
from the source without an Internet connection [101].
trade-offs and stability are monitored to ensure accu-
By utilizing monitoring and measurement tools, online
racy and to determine the optimal sequence of steps to
and offline data can be collected efficiently, securely,
be taken.
and stored for model adaptation. Data collection marks
the beginning of training and learning. Validation and
testing are set in motion after that. C. MACHINE LEARNING TECHNIQUES IN B5G/6G
3) DATA ANALYSIS: It is divided into two types: prepro- WIRELESS COMMUNICATION SYSTEM
cessing and feature extraction where preprocessing is A few existing works containing a comprehensive overview
used to reduce noise from the gathered data. The data’s of the existing intelligent techniques with their application in
features are then extracted, which is a prerequisite for the B5G/IoT wireless communication systems are discussed
learning and training [102]. The types of characteristics here. Authors in [99], [101] talk about different learning
that can be extracted from the network include packet- techniques in IoT applicable scenarios while taking into con-
level features and flow-level features. The packet size, sideration their input information and computational com-
mean, root, and variance are extracted at the packet plexity. While the works in [103], [104] provide fundamental
level, and mean flow length and mean number of packet concepts about the state of art AI-based technologies applied
flow features are extracted at the flow level. in the existing network, making the system competent enough
4) MODEL CONSTRUCTION While iterative process to accomplish self-configuration, self-optimization, and self-
selection, training, and tuning are all necessary aspects healing of the concerned system.
of the model selection process, they are applied dif- Additionally, authors in [58] discuss the framework and
ferently and a suitable learning model must be chosen application of the popular deep reinforcement learning (RL)
depending on the dataset size. The training stage entails technique that is vital in the engineering of cognitive smart
training the model with the dataset that will be collected cities. Furthermore, the articles in [105], [106] further provide
at the start of the stage, whereas the tuning stage has a comprehensive survey of the existing RL algorithms and
the model learn for itself by comparing it to the trained their stability and behavioral adaptability with the other learn-
data. ing agents in the implemented network. Speech recognition,
5) MODEL VALIDATION: Cross-validation of the test- bioinformatics, and computer vision are just a few of the
ing process is used to check the model’s accuracy, applications for machine learning that make efficient use of
which aids in optimizing the model and preserving the the technology.
overall efficiency of the system. Machine learning is primarily used for prediction and clas-
6) DEPLOYMENT AND INFERENCE: Throughout sification, but it also plays a role in performance prediction
the deployment and inference stages, the model’s and intrusion detection in networking systems. Therefore to

VOLUME 11, 2023 30655

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 5. Machine learning techniques and their applications in B5G/6G System.

make decisions directly, machine learning constructs mod- system set and learns how to map all information about the
els that can learn from data without adhering to a set of action, without any training data [109].
rules [107]. ML hence allows a model to enter a self-learning This section provides a brief overview of some of the most
mode without having to be explicitly trained. To learn system commonly used artificial intelligence and machine learning
characteristics that cannot be represented by an explicit math- techniques, with Table 5 illustratively defining and highlight-
ematical model, ML models, therefore, are used as computing ing various machine learning techniques and their function-
systems. ality in B5G and 6G wireless communication systems. The
These models are employed in tasks including catego- works in [110]–[113] encompass a few of the researched
rization, regression, and intelligent agent-environment inter- investigations concerning the ML application on the B5G/6G
actions. Using basic arithmetic calculations, the model can wireless communication channel. However, most of the exist-
efficiently complete the task once the system characteristics ing works exercise intelligence on a limited part of the
are learned [108]. Models may be trained by providing them existing wireless communication channel. Some of the wire-
with data sets, and when they are exposed to fresh data, they less channel characteristics influenced by intelligent learning
can learn, forecast, and develop. There are three types of techniques are discussed below:
machine learning algorithms: supervised learning, unsuper- 1) CHANNEL MODELING USING ML: The ML helps
vised learning, and reinforcement learning. deal with the channel modeling problem by imple-
In supervised learning, a model is trained on labeled menting the model-based approach, extracting the
datasets and then learns on its own by comparing the training wireless channel features from the existing data.
dataset to the predicted output. This method is commonly ML has efficacy in predicting channel features, esti-
used for classification and regression issues. Unsupervised mating channel parameters, CIR18 modeling, MPC,19
learning is a type of machine learning that uses an unlabeled and classification of scenarios about the concerning
dataset to detect patterns and relationships in the data. It is
mostly used to solve clustering and association problems. 18 CIR: Channel Impulse Response.
19 MPC: Multipath Component Clustering.
During reinforcement learning, an agent interacts with a

30656 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

channel environment, derived out from the above-cited to achieve complete automation of the systems that surround
works. us, taking into account the feasibility, interoperability, and
2) CHANNEL MEASUREMENT: A channel model functionality of the 6G and IoT-based wireless communi-
based on a feed-forward neural network(FNN20 ) and cation networks. An intelligent touch-infused technological
RBFNN21 is shown in [114] which is functional in pre- framework is proposed in the next section, which incorpo-
dicting channel properties like received power, RMS rates intelligence from existing smart IoT infrastructure and
delay, and angle spread(DS/AS). Therefore in addition interfaces it with next-generation systems by creating a tac-
to the transmitter and the receiver coordinates, their tile/haptic feeling as we interact.
input parameters also influence the distance and fre-
quency of the Tx-Rx link. VI. PROPOSED TOUCH TECHNOLOGY IN B5G/6G AND
3) NOISE: ML in [115] makes use of ANN to remove IOT NETWORK
noise from CIR, while [116] makes use of CNN to
The proposed and futuristic anticipated system infrastructure
identify the relevant wireless channel. In CNN and
is expected to encompass an intelligent and reconfigurable
wireless communication channels, MPC characteristics
touch-enabled system that is pertinent in an IoT interfaced
like amplitude, latency, and Doppler frequency serve as
B5G/6G system. The intelligent Touch-based IoT paradigm
input and output parameters.
can be made up of a variety of functional elements that
4) CHANNEL ESTIMATION: To obtain accurate chan-
help smart objects perform various functions such as sensing,
nel estimation, the work in [117] uses a 2D non-
actuation, identification, management, and communication.
linear complex support vector regression (SVR) in
The touch-based IoT system’s functional elements [3] can be
a rapidly fading and time-varying multipath channel.
summarized as follows:
On the other hand, the work in [118] refers to a deep
learning-based channel estimation method for beam 1) SMART DEVICES: The primary components of the
space mm-wave massive MIMO systems that can learn IoT-based Touch System, performing sensing, actua-
the channel structure from a huge number of training tion, and control functions are capable of sharing data
datasets. with other applications and servers. To connect to other
5) MASSIVE RADIO INTERFACE: ML algorithms help smart devices, each IoT device has to be prepared
in analyzing the enormous amount of data pro- with numerous interfaces including internet access, I/O
duced by the massive MIMO arrays, where the interfaces for sensors, audio and video, storage, and
conventional channel estimation and detection algo- memory interfaces.
rithms are rendered incapable. Deep learning meth- 2) FUNCTIONALITY: The device functionality ranges
ods and techniques, particularly image processing and from smart-watches, wearable sensors, automatic cars,
video analytics, provide the most exciting algorithmic industrial machines, LED lights, and so on. From office
approaches [119]. automation and household appliances to manufacturing
6) WIRELESS NETWORK LOCALIZATION: The con- lines and commodities tracking, intelligent IoT tech-
tinuous development and updating of wireless chan- niques are used in a wide range of applications. As a
nel locations have been made feasible via automated result, IoT services must be used to improve IoT appli-
learning from crowd-sourced data employing a large cation development and accelerate installation.
number of mobile devices, yielding precise localization 3) SERVICES These are dedicated to identity and device
results. In exchange, it enables consumers to benefit modeling and are commonly grouped under the
from improved location-based services. umbrella term identity services. Additional subcate-
7) NETWORK MANAGEMENT Machine learning and gories include information aggregation, discovery, con-
artificial intelligence can optimize a variety of tasks trol, collaborative awareness, ubiquitous services, and
such as fault detection and user tracking over a wire- data analytics and publishing.
lessly linked network. 4) REMOTE ACCESS: As opposed to devices that use
8) RESOURCE MANAGEMENT: The resource manage- mechanical switches or buttons to remotely manage,
ment mechanism is only able to function once the IoT devices have either no human involvement or can
system has memorized the states and conditions of the be remotely managed without the need for human
network users and their real-time wireless environment. intervention.
This, therefore, helps improve the system performance 5) SECURITY Taking into consideration the security
with time and in turn, helps the system incorporate an aspect, as far as the data on wireless networks is con-
intelligent and dynamic decision-making phenomenon. cerned, especially with regards to denial of service
Following consideration of all relevant factors, we intend to (DoS), spoofing, and eavesdropping, the information
incorporate intelligence into the tactile internet infrastructure is vulnerable to an array of attacks. Thus, IoT systems
use many security features, such as privacy, authoriza-
20 FNN: Feed Forward Neural Network. tion, authentication, data security, content integrity, and
21 RBFNN: Radial Basis Function Neural Network. message integrity, to attempt to thwart these attacks.

VOLUME 11, 2023 30657

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

6) IOT APPLICATION: To provide IoT users with inter- to set up a global network of connected smart objects and
faces, the application layer supplies IoT users with var- devices, all of which can connect without human intervention.
ious interfaces that enable them to monitor and control Each object that has been embedded with a smart interface
the various aspects of IoT applications. and connected to the user platforms and digital environments
Thus the projected system is to be most likely based on the is assigned a virtual identity and interfacing, allowing it to
IoT functionality and is to be expected to be implemented in connect and communicate with other embedded objects [3].
the B5G/6G networks, where the system intelligence is of an As we build our IoT network, our physical and virtual entities
utmost need to be compatible with the accelerating high data turn into virtual things in a cyber world, each with specific
rate and in turn satisfy the low latency requirements of the abilities that all IoT entities can use to realize personalization,
next generation system [120], [121]. specialization, and autonomy based on the communications
protocols used to make the smart entities unique and provide
A. INTELLIGENT TOUCH-BASED SYSTEM them with virtual personalities.
The problem statement here describes the need for the intel- The combination of B5G/6G network slicing technology
ligence in the B5G system and therefore can be elaborated as and the TI application will thus prove to be a driving factor
per the works in [122]–[125] stating that ever since the evolu- in the realization of this suggested reconfigurable and intel-
tion of the wireless network from 1G to the existing 5G/B5G ligent touch-enabled framework, whose research is still in its
and subsequently the 6G networks, have led to a tremendous infancy. Consequently, it can be concluded that the main goal
ascend in the billions of connected devices bound together by of the IoT-based Touch technology system is to improve the
IoT, forming an integrated IoE network. The increased traffic, lives of people, where all objects around us can figure out
due to a growing number of devices, therefore requires high what we want, what we require, and what we like, as well as
energy efficiency and lower latency. serve it accordingly without us having to explicitly command
The growing new use cases in the evolving B5G net- them.
work incorporate the AR/VR/XR based smart systems includ- Fig. 9 presents the layered architecture of the proposed
ing smart road monitoring, smart cities, consisting of smart intelligent touch technology in B5G/IoT networks. Hence it
homes and IoT governed smart appliances. These systems requires an AI/ML technology, coupled with the B5G/6G
are externally controlled and therefore greatly lack intelli- based Network Slicing and Tactile Internet; to implement
gence. Hence for the efficient functioning of these devices and interface the intelligent touch-based system. Hence the
in the next generation 6G network, an intelligent system Network slicing and tactile internet explained in the previous
is required to govern the existing AR/VR based 5G/B5G sections prove to be the cornerstone in the incorporation of a
network effectively. reconfigurable intelligent touch system.
It, therefore, requires an interfacing mechanism with the
existing network infrastructure. To implement this in real- VII. LAYERED ARCHITECTURE FOR TOUCH TECHNOLOGY
time, the B5G/6G enabled tactile internet proves to be a
The Internet of Things connects millions of smart objects,
promising technology, which in turn can provide a new
increasing data traffic and necessitating the use of large
dimension to H2M interaction by enabling haptic sensa-
processors and storage systems. This emergence of IoT sys-
tions and therefore a touch-enabled communication interface,
tems together with the rising demand for wireless capac-
in real-time. The touch-governed IoT system is expected to
ity ultimately paves way for the futuristic intelligent Touch
permeate many facets of contemporary daily living, including
Technology system. Many factors, including interoperability,
the ability to sense, process, analyze, and infer environmental
scalability, QoS, and reliability, must be considered while
parameters from natural resources and ecologies to human
designing such an IoT-based intelligent infrastructure. As a
environments.
result, the required intelligent IoT architecture based on touch
The main aspect of this touch-based IoT network is its abil-
technology should have the following characteristics:
ity to intelligently connect to the other existing and future net-
works, to all the resources that are utilized by these networks, 1) DISTRIBUTIVENESS The IoT model for the pro-
and to help accomplish that through the advancement of the posed system should enable data collection from var-
prior networks and communication protocols. For this vision ious sources and their processing by various smart
to succeed, we will need to advance beyond conventional entities in a distributed manner.
mobile computing technologies and design an IoT system in 2) INTEROPERABILITY: IoT devices from different
which everything we can touch is connected and capable of vendors must communicate to achieve common goals.
acting as a smart and intelligent extension of ourselves. Additionally, protocols and systems must be designed
In-depth knowledge of IoT issues spans from an awareness in such a way that smart devices from a variety of
and a better understanding of the IoT concerns and the com- manufacturers can exchange sensed data in an interop-
plexities involving the size and depth of the pervasive com- erable manner.
munication network, software architectures, and data transfer 3) SCALABILITY:Any such IoT network is expected to
and processing. This knowledge is used to create autonomous have billions of objects connected at any time. Because
and intelligent devices in IoT systems. The primary goal is these platforms run many applications and systems,

30658 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

FIGURE 9. Proposed layered architecture for intelligent touch technology.22

hence such applications should be able to handle and form the HSI network. Thus, this layer contains various end
process a huge amount of data. devices and sensors, which aid in touch-based sensory data
4) LIMITED RESOURCES:Both computing units and transmission. All of this is located on the perception layer as
energy are considered highly rare units regarding the per the layered architecture presented before. The perception
limited resource availability. layer is responsible for transmitting the generated sensory
5) SECURITY: The feelings of helplessness and vulnera- data to its required destination, over the network.
bility that users have when they are under the control The reliable and timely transmission of data from the per-
and dominance of an unknown external device could be ception layer devices to the edge computing layer is ensured
detrimental to the given system deployment. by the efficient connectivity between the layers. The connec-
This section presents the proposed layered architecture for tivity layer is therefore responsible for processing and com-
the intelligent and touch-enabled technology which is to be munication between the master devices accompanied by their
functionally implemented in the B5G/6G and IoT configured effective routing and switching, enabling reliable delivery of
wireless communication network. This architecture, which information across the network. This layer also ensures the
is defined by the tactile internet and network slicing as its security of the network.
cornerstone components, can thus be broadly classified into The third layer is edge computing or the Fog layer, where
master, controller, and network domains. the data evaluation is done and is processed at higher lev-
The master domain is comprised of user-facing edge els. The information is then processed and transferred by
devices such as a robotic arm, gesture-based glove, or a gam- the tactile support from the end-user to the base station.
ing console for real-time online gaming, which collectively Thus the fog and edge network, therefore, facilitates dis-
tributed computing, storage; control, and communication of
22 The architecture presents the IoT Layered Infrastructure compris- the network functions with a reduction in the latency, system
ing of: Perception, Connectivity, Edge computing(network), Cloud Stor- response, and cloud workload of the transmitted data. It is
age(transport) and Application layers, providing tactile-based communica- further connected to the RAN layer which is responsible
tion and device interaction, as well as efficient network slicing between
the connectivity and application levels in order to transfer and completely
for introducing intelligence in the system through the radio
support requested services and user requirements. network.

VOLUME 11, 2023 30659

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

The B5G/6G network slicing, virtualization, and frag- homes, intelligent transportation, smart health, and smart
mentation take place in this layer, the details of which are cities, among others. Furthermore, it is in charge of receiving
explained in the earlier sections. Hence to further avail the the data that has been processed from the network layer.
tactile facilities, the RAN slicing is accomplished, fragment- The IoT application layer provides appropriate protocols
ing it into the isolated frameworks, each of which is further and services needed to transmit messages at the applica-
meant to furnish various high data rate applications simulta- tion level. When choosing a communication protocol for a
neously, while maintaining the latency of the requisite ≤1ms. certain application, many elements should be considered,
The routed network endows with a provision for network including power consumption, necessary BW, transfer and
slicing and storage of data, enabling access to the various connection time, delivery guarantee, data security, and packet
cloud services, for effective data storage and acquisition. size. A description of a few protocols that are commonly used
The data acquisition process, before further furnishing at the application layer is provided below and summarized in
the requisite applications, follows the data accumulation and Table 6:
abstraction. Data accumulation involves the data capture and 1) MQTT: It uses middleware and apps to enable communi-
storage, to be used by the applications later. It also deals cation between IoT devices and the network in a variety
with query-based data processing. Data abstraction further of ways, including M2M, server to server, and machine to
virtualizes and consolidates the data at a place before its slic- server, and it runs on the TCP/IP protocol. Additionally,
ing and cloud storage. The setup of the signalling procedure it supports communication over low-bandwidth and unre-
and protocol stack along with the physical layer optimization liable links and is used for publishing and subscribing to
allows the accessible devices to direct and control the power lightweight message exchanges [127].
of the sensors, edge, fog, and cloud-based platforms. 2) XMPP: It allows for low latency communication and
The application layer further furnishes the applications minimal message delivery, making it ideal for video calls,
due to application slicing. Here the applications required are instant messaging, and chats, as well as publish/subscribe
entirely based on the TI and are intelligent and touch-enabled. systems, gaming, and IoT applications. Because of its
These may vary from the haptic communication by the haptic simplicity and adaptability, it makes it possible to com-
and IoT sensors, the automated remote industrial operation municate between heterogeneous systems [127].
and monitoring, which is an example of the industrial revolu- 3) REPRESENTATIONAL STATE TRANSFER (REST-
tion 4.0. It has great potential in the evolution of healthcare, FUL): The REST protocol is a collection of best prac-
education, entertainment as well as edutainment. Since the tices, rules, and constraints for developing web services
application layer is in charge of monitoring, controlling, and that enable data exchange and communication between
analyzing the data, it must be at the heart of all systems. various devices, as well as for developing distributed
The business layer, in the end, provides these services at hypermedia systems with desirable properties such as
the consumer end i.e. at the controller domain at the receiver modifiability and scalability. RESTful is a request-
at the destination. For an easy understanding, let’s go over response and client-server architecture that allows clients
each layer of the architecture with its associated protocols and to access server resources in IoT contexts. It is based
functionality. Following the bottom-up approach, the five- on the HTTP protocol. Because they are lightweight and
layered IoT configured architecture of the proposed touch straightforward protocols, RESTful APIs are considered
technology comprises of the following layers: to be a good choice for a variety of IoT applications [128].
1) Perception layer. 4) CONSTRAINED APPLICATION PROTOCOL (COAP):
2) Connectivity (Data Link) layer In IoT applications, it enables resource-constrained and
3) Fog/Edge computing (Network) layer unsynchronized devices to communicate while provid-
4) Data Accumulation (Transport) layer ing flow control, reliable delivery, and simple congestion
5) Application layer control. It uses multicast and unicast requests to enable
a publish-subscribe communication strategy. CoAP uses
As a result, we will go over each of these layers one by UDP due to its small message size, low code footprint,
one, explaining their functionality and interoperability as we and lack of TCP handshake overhead before transmis-
go. This section describes each layer one by one with its sion [129].
functionality and interoperability beginning with the appli- 5) AMQP It is widely used in commercial and business
cation layer up to the perception layer following a top-down domains because it offers reliable and secure communica-
approach [126] in Fig. 9. tion between heterogeneous devices and supports publish-
subscribe architecture based on an efficient and reliable
A. APPLICATION LAYER messaging queue. In addition, it uses the TCP protocol
As the front end of the IoT architecture, the application for increased reliability. The message queue and exchange
layer is where the majority of the IoT potential will be queue models are used to transfer data over AMQP. In the
realized. This is because the application layer provides IoT message queue model, messages are retained until they
developers with the user interfaces, platforms, and tools that are transmitted to the receiver, whereas in the exchange
are required to implement IoT applications such as smart queue model, messages are routed in proper order [129].

30660 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 6. Some commonly used application layer protocols.

B. DATA ACCUMULATION/STORAGE LAYER data packet into multiple packets, each with an ordering
It interacts with the application layer to send and receive data number and source and destination IPs [130].
without mistakes in a typical way. The transmitting side here 2) USER DATAGRAM PROTOCOL (UDP): It is a con-
is in charge of breaking down messages received from the nectionless protocol that attempts to give protocols
application layer into segments and sending them to the net- and applications that run over IP, an unreliable, min-
work layer. Following that, the segmented messages received imum message queuing, message passing, and best-
will be reassembled into messages that will be passed directly effort transport. There is no requirement for end-to-end
into the application layer by the receiving side of the commu- connectivity between communicating entities, which
nication channel. enables efficient communication for some applications
This layer is responsible for ensuring the integrity and that require real-time performance with low latency,
reliability of transmitted data by providing features such as such as video and voice. UDP does not provide a port
packet delivery order, congestion avoidance, multiplexing, with any attribute for addressing the source and destina-
byte orientation, and data integrity. Hence known as the tion functions and provides a data integrity check [130].
routing layer because it is in charge of routing data packets 3) DATAGRAM CONGESTION CONTROL PROTO-
through the network area, where its protocols are in charge of COL (DCCP): It establishes unicast bidirectional
packet ordering, error detection, and correction. This section connections for datagrams with unreliable dynamic
provides an overview of a few protocols that are commonly congestion control. These characteristics make DCCP
used at the data accumulation layer. The protocols are sum- ideal for applications that transmit large amounts of
marized in Table 7: data and require a trade-off between reliability and
1) TRANSPORT CONTROL PROTOCOL (TCP) It is timeliness, such as VoIP and media streaming. Due
a heavyweight, connection-oriented protocol, where to its unreliability and absence of a receiving win-
the connection is not established until all the required dow, the flow rate of DCCP can be progressively
data has been sent between each end device. TCP increased [130].
is suited for reliable communication since it requires 4) STREAM CONTROL TRANSMISSION PROTO-
acknowledgement messages to ensure each sending COL (SCTP): It is a connectionless, message-oriented,
and receiving procedure, as well as support for retrans- IP transport layer protocol similar to UDP that enables
mission of lost or corrupted packets and a flow control SCTP-based peer-to-peer (P2P) communication and
mechanism. Hence, this protocol’s packet overhead is reliable transmission for applications communicating
extremely high, resulting in increased power consump- over an IP network. As a result, it inherits the majority
tion from devices and thus making it impossible to of TCP’s functionality, such as packet recovery and
operate on power-constrained devices. TCP divides the congestion management [130].

VOLUME 11, 2023 30661

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 7. Some commonly used data accumulation layer protocols.

5) TRANSPORT LAYER SECURITY (TLS): It was cre- paradigm, cache management routers, and receiver-
ated to provide security channels among communicat- initiated reservation [133].
ing peers and to give authentication, data secrecy, data
integrity, and encryption to applications by preventing C. EDGE/FOG COMPUTING LAYER
eavesdropping, message forging, and tampering. It runs
It is responsible for providing data with routing paths so that
on top of several transport layer protocols. It is com-
the packets can be transmitted across the network area. This
posed of two components: the handshaking protocol,
layer creates logical connections, sends out error messages,
which is responsible for authenticating communica-
and maintains the data transmission routing path. Further-
tion ends, agreeing on shared keys, and negotiating
more, this layer contains all network devices such as switches,
cryptographic parameters and modes; and the record
firewalls, bridges, and routers that are required to work with
protocol, which divides the traffic into multiple records
appropriate communication protocols such as 3G-4G-5G,
and protects them using the traffic keys [131].
Wi-Fi, infrared, ZigBee, and Fibre to the X.
6) DATAGRAM TRANSPORT LAYER SECURITY
This layer is in charge of forming, addressing, and rout-
(DTLS): It was created to secure datagram applica-
ing data packets, as it receives datagram packets from the
tions that do not require or provide dependable data
transport layer and converts them to data packets before
delivery, such as datagram online gaming, internet
transmitting them to the destination side. This section gives an
telephony, and media streaming, which are deemed
overview of some protocols commonly used in the edge/fog
delay-sensitive. DTLS is an enhancement of the TLS
computing layer with the protocols summarized in Table 8:
protocol that prevents message forgery, tampering,
and eavesdropping when transmitting data streams. 1) ROUTING PROTOCOL FOR LOW POWER AND
Therefore, it should be able to cope with and resolve LOSSY NETWORK (RPL): It is a tree-based, IPv6
a variety of datagram difficulties, such as packet loss, proactive distance vector routing protocol developed by
packet reordering, and delay [132]. the routing-over-low-power and lossy networks work-
7) RESOURCE RESERVATION PROTOCOL (RSVP): It ing group to run commercial appliance networks with
is multicast and unicast control transmission protocol insecure connectivity, poor data rates, and substan-
that was created to enable data stream transmission tial losses. It has a storing and non-storing mode to
with a flexible, robust, scalable, and heterogeneous reduce memory requirements and eliminate loops in
resource reservation setup at each router. There is also low-resource applications. It is prone to high packet loss
support for resource reservations in each node along owing to congestion, has a long delay, and is vulner-
the data path, multipoint to multipoint communication able to assaults since it lacks end-to-end encryption.

30662 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

Overhead packets for control are flooded into the net- In addition, it converts IP addresses into internet linking
works as a result [134]. addresses (ILA) and when a node wants to be connected
2) COGNITIVE ROUTING PROTOCOL FOR LOW to the internet, the IP associated with the desired internet
POWER NETWORK (CORPL): It is an extension of is connected to ILA so that the search function can be
the RPL protocol, which was designed to be compatible utilized. If there is no internet node in ICT, the source
with cognitive networks to improve performance. This node will broadcast the requested packet to update both
feature ensures a high packet delivery ratio (PDR) and tables until an optimal route to an internet node is dis-
keeps the network from colliding. The CORPL routing covered [139].
mechanism takes advantage of an opportunity to select
the most effective forwarding next hop from a pool of D. CONNECTIVITY LAYER
eligible neighboring nodes. There are minimum colli-
The IoT connectivity layer in the touch system architecture is
sions and duplication of data packets [135].
comprised of a variety of communication protocols that are
3) CHANNEL AWARE ROUTING PROTOCOL (CARP):
primarily responsible for providing services to the network
It is a distributed protocol with a lightweight data
layer. Hence, it is in charge of connecting and transmitting
package that was created for underwater and IoT appli-
signals from end devices to higher layers via routers and
cations. Network initialization and data forwarding are
gateways.
two steps in the routing operation performed by CARP.
The IoT connectivity layer consists of a variety of commu-
The receiving node updates its distance to the sink
nication protocols that, depending on the transmission range
node, broadcasting the welcome messages with their ID
and coverage area, provide services to the network layer.
and hop count. In data transmission, the sender sends
Some of the most commonly used protocols are reviewed
a ping message to its neighbors, determining the best
further down this section.
relaying node based on the link quality and information
received from their ping messages, and then forwards 1) NFC PROTOCOL A short-range protocol that allows
data. When selecting a relaying node, residual energy, mobile objects to communicate with one another over
network quality, and buffer space are all taken into a few cm of distance and allows data to be transferred
account [136]. in seconds between the connected NFC devices that are
4) COLLECTION TREE PROTOCOL (CTP): A tree- close to one another. It is RFID-based and thus uses
based routing system that was created to give the great- an alternate magnetic field to connect devices that are
est effort for anycast communication in networks with either active or passive. In the active mode, all of the
low energy demands. An early form of networking is connecting devices generate a magnetic field, whereas,
where a source node (sink node) announces itself as the in passive mode, one device generates a magnetic field
root node, where minimal cost is paid to deliver data while the others use load modulation to transmit data.
to the root. Other nodes connect to the root tree via Passive mode is energy-saving and is widely used in
lightning ads and then send their collected data into the today’s smartphones [140], [141].
sink node with the minimum amount available. CTP, 2) 6LOWPAN: It allows smart devices to connect to the
on the other hand, does not permit reverse routing from internet via the IPv6 protocol while also taking into
the sink to the sensors [137]. account the nature of wireless IoT networks by creating
5) LIGHTWEIGHT ON-DEMAND AD HOC DIS- a very small header message format. It also removes
TANCE VECTOR ROUTING PROTOCOL-NEXT obstacles to using IPv6 addressing protocol in IoT
GENERATION (LOAD-NG): When compared to the devices with limited processing power, data rate, and
on-demand distance vector (AoDV) protocol, it is a power [142]–[144].
more lightweight distance vector and reactive protocol 3) BLUETOOTH LOW ENERGY (BLE) PROTOCOL:
that is designed to provide a secure, scalable, and effi- BLE is a low-power alternative to short-range wireless
cient routing in lossy and low power networks. As a communication developed by the Bluetooth Special
reactive protocol, LOAD-ng generates on-demand route Interest Group. Additionally, it allows for fast data
requests to discover a path to the target node and when packets to be transmitted at speeds up to 2Mbps in the
data has to be delivered, the receiving unicast replies ISM band [145].
hop by hop from the destination node back to the sender 4) ZIGBEE The objective is to develop a low-cost, scal-
node. If a route is found to be broken, attempts to fix it able, and power-sipping wireless connectivity that is
are made, or an error message is sent to the requested suitable for a wide range of controlling and monitor-
node [138]. ing purposes. With intelligent routing and setup proce-
6) AD-HOC ON DEMAND MULTIPATH VECTOR FOR dures, this protocol builds on IEEE 802.15.4’s features
IOT (AOMDV-IOT): It seeks to discover and estab- to enable high failure resilience and easy installation.
lish connections between nodes and the internet. For It also works well with other wireless communication
each node, AoMDV-IoT generates two routing tables: technologies due to its strict security and listening
an internet connecting table (ICT) and a routing table. techniques [146].

VOLUME 11, 2023 30663

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 8. Some commonly used edge/fog computing layer protocols.

5) Z-WAVE Smart light controllers and other sensors in- 7) LONG RANGE WIDE AREA NETWORK (LORA
home devices use this low-power wireless communi- WAN) It is a physical layer communication protocol
cation technology. With low latency transmissions and that uses very little power and has a battery life of up
data rates of up to 200kbps, this technology operates to ten years. LoRaWAN specializes in M2M, industrial
over 900 MHz ISM bands [147], [148]. and smart city applications, which require long-range
6) LOW POWER WIDE AREA NETWORKS (LPWAN) communication, ranging from 2 to 5 km in urban areas
PROTOCOLS: LPWAN protocols are low-power, low- and up to 15km in suburban areas. The process of
bandwidth, and low-cost protocols that are particularly communication through large networks, which contain
useful for long-distance communications. Furthermore, billions of intelligent devices, also promotes the data
the devices that implement these protocols have trans- rates of this protocol in the complete duplex wireless
mission ranges ranging from 1m to 50 km. The general medium, from 0.3 to 50Kbps [149].
characteristics of LPWAN protocols are as follows, 8) NB-IOT: It is a narrowband radio technology that was
followed by a brief discussion of each protocol: developed and standardized by the 3GPP to support
a. These LPWAN protocols are implemented by IoT applications with low data rates and high com-
low-power devices. plexity. It proposes a new radio access method based
b. These protocols are limited to the transmission of on LTE standards but with fewer capabilities to lower
small smart packets, typically 100 bytes or less. the power consumption of IoT devices with limited
c. Devices that implement LPWAN protocols are resources [150].
composed of extremely low-cost components, 9) SIGFOX It is a technology of narrowband and
typically costing less than a few dollars. ultra narrowband for connecting a large number of
d. Within and outside their domains, these devices power-controlled devices. To operate, the protocol
are designed to provide good coverage and must operate on a frequency band of 868MHz, where
reliability. the spectrum is split into 400 channels of 100Hz. Rural

30664 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 9. Some commonly used short ranged connectivity layer protocols.

areas can receive signals from IoT devices that can and its application, TI as the building blocks of the touch
transmit up to 140 packets per day at a data rate of system.
up to 100bps, and urban areas can receive signals that One of the most promising aspects of this proposal is that
can reach distances of (30-50)km in rural areas and it is based on the assumption that 6G applications will have a
(3-10)km in urban areas [149]. high data rate and latency requirement of less than 1ms in the
majority of cases. The application of the TI-based intelligent
Table 10 compares the characteristics of the LPWAN proto-
systems is still in the nascent stage and therefore requires a
cols discussed in this section [149]:
well-defined technology to deploy the touch-enabled technol-
ogy in the next generation networks, combining the network
E. PERCEPTION LAYER slicing with the TI.
The major goal of this layer is to feel the physical characteris- Thus an IoT-enabled interface acts as a backbone of the
tics of the entities that surround us and within the dominating proposed reconfigurable and intelligent tactile-based touch
IoT network, which relies on sensing technologies like RFID, communication wireless system. Summarizing the important
WSN, and GPS. It’s also in charge of translating sensory data requisites for the proposed touch-based technology can there-
into digital signals that may be transmitted via a network. fore be listed below as:
Indeed, embedded intelligence and nanotechnology play an 1) An interfacing architecture with the existing B5G/IoT
important role in this layer, as they improve the processing framework is required that can link the present B5G
capabilities of any object by inserting small chips or micro- network with the next generation 6G network through
controllers into everyday smart devices. This layer consists IoT-based intelligent and sensory devices and sensors.
of all user end devices (smart devices, wearables, sensors, 2) The network slicing and TI prove to be a strong back-
actuators, etc.) that are connected to the IoT and cellular bone in implementing the given framework. Also, the
network and capable of accessing and transmitting tactile intelligence may be incorporated into the system using
sensations over the network. ML/AI-based algorithms, to increase the system com-
Additionally, some fundamental attributes are an integral putational capacity.
part of the end devices, as detailed in Table 11 [151]. For 3) The 6G applications are largely based on the
the system to enable an intelligent and reconfigurable touch- AR/VR/MR/XR, having large power consumption and
based system, both the master domain and the controller real-time simulation and therefore need proper energy
domain, need to have bilateral connectivity with the intel- efficiency and power optimization techniques for its
ligent RAN network. It is further governed by the AI/ML effective implementation. It can therefore prove to be
algorithms, responsible for intelligent resource allocation and an important future aspect that needs to be considered.
data transmission. Hence the motivation for the proposed 4) Also, the minimum infrastructure costs, with an
model lies entirely in the fact that to incorporate an entirely energy-efficient performance are important parameters
intelligent system in the 6G and IoT framework requires the that need to be considered while implementing the
AI/ML technology implemented on the B5G network slicing proposed system.

VOLUME 11, 2023 30665

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 10. Comparison of LPWAN protocols. TABLE 11. Fundamental attributes of connected user end devices.

architecture in Fig. 10. This end to end architecture is, there-

fore, to be realized in three phases which include:
1) The TI enabled real-time touch communication net-
work establishments at the transmitting end complete
with its bilateral feedback system.
2) Similar TI enabled real-time touch reception system at
the receiver end that is operated by the remote industrial
machinery and robotic system.
3) Lastly, it requires an intelligent interfacing algo-
rithm to simultaneously operate both the ends and
a suitable slicing mechanism to fulfil almost all the
URLLC and IoT-enabled intelligent Touch-based tac-
tile applications.
Hence it can be concluded by saying that, all the above-
enlisted parameters, will eventually pave the way towards
the establishment of an intelligent and configurable touch- VIII. END TO END TOUCH SYSTEM ARCHITECTURAL
enabled system, interfacing with the B5G/6G Wireless MODEL
Communication Network. We have further proposed an The E2E system architectural model of the proposed intelli-
end-to-end touch interfacing architecture to be implemented gent touch technology has been represented in Fig. 10. The
in the real-time B5G/IoT-based wireless communication net- real-time applications at the user ends satisfying the desired
work as described ahead in this section. output at the destination may range from AR/VR-based trans-
The IoT infrastructure includes physical objects integrated mission and communication systems, real-time online gam-
with the WCN, which are designed to provide intelligent in- ing, remote medical service accessibility like telesurgery and
house services to users. Thus the IoT system comprises five remote health consultation, and remote industrial operation
layers perception, connectivity, network, transport, and the involving the major role of robots and automatic machinery.
application layers starting from the device end. The network The system model has been divided into three components
layer in turn includes the connectivity with the edge/fog and therefore may be analyzed in three phases. These are
computing as well, as the data storage functionality within categorized as:
itself as given in Fig. 9, altogether forming the middleware 1) PHASE1: The bilateral communication at the transmit-
component of the system. ter comprising of the tactile based user devices like
The proposed Touch-based intelligent and IoT config- the robotic arm, AR/VR gear and gesture-based haptic
ured system may therefore be realized as an end-to-end devices, real-time online gaming console, and many

30666 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

FIGURE 10. Touch interfaced end-to-end architecture.

more, interacting with the BS, the local server (LS), Fig. 11 presents the touch-enabled flow process at the trans-
transmitting the data to the middleware through the mitter end. The process may begin with the abovementioned
gateway. devices at the user end requesting diverse high data rates
2) PHASE2: It represents the bilateral communication at and IoT based URLLC applications varying from the remote
the receiver end, accomplishing the intelligent, touch- medical consultancy or telesurgical operation to the assem-
based applications ranging from intelligent healthcare bling and modeling of the machinery of the remote industrial
facilities like remote surgery, remote industrial opera- function. The user may request for a particular service by
tions, real-time online classrooms, and many more. the means of a touch sensor or an AR/VR device, virtually
3) PHASE3: It acts as the mediating controller, connect- furnishing the data towards the server, to be routed further.
ing the above phases and is therefore responsible for The further the convenience may be at the individual view-
implementing intelligence in the proposed system. The point, making use of a tactile enabled robotic system for
AI/ML algorithms are therefore routed in this phase. household function. In addition, it may have a future appli-
The aforementioned phases may be separately analyzed cation as in virtual e-commerce or virtualized or holographic
as the transmitter end, the receiver end, and the process- shopping where the customer may virtually access or try
ing end. the product by tactile sensing and a touch-enabled interface
between the user end and the linking server forming a bilateral
A. TOUCH-BASED TRANSMITTER SYSTEM communication feedback system at the user/customer end.
The transmitter end of this end-to-end architecture of the The TI-based applications place a request to the local
intelligent touch interfaced system may be represented as server to check for channel availability. The request is for-
phase 1, comprising of the tactile user devices like the warded when the channel is available and if the channel
robotic arm, augmented or virtual reality gear, gesture- appears to be busy, the entire process starts all over again from
operated tactile devices, real-time online gaming modules. the first step, where the user must ask for a service. Following

VOLUME 11, 2023 30667

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

FIGURE 11. Touch-enabled flow process at the transmitter end.

the channel access granted to the user, and before sending in the network slicing and providing the requisite virtual
the data it tries to establish a connection with the server. This platform for supporting different tactile applications. The
is followed by the handshaking request-response procedure slices may be controlled, scheduled, and allocated using dif-
between the channel and server. ferent ML/DL-based algorithms, successfully implementing
The server gives acknowledgement on the channel avail- the proposed touch-based network.
ability and thus firmly establishes the connection between the
customer and the server, virtually accessing, forwarding, and B. TOUCH-BASED RECEIVER SYSTEM
routing the data further to the gateway. Here ‘A’ represents The flow process at the receiver end of the touch interfaced
the data to be forwarded to the gateway. A similar kind of system is presented in Fig. 12. The point ‘A’ which rep-
procedure is conceded at the receiver end explained in the resents the user end devices at the transmitter end, further
next section. routes the information signal in the channel via BS, router,
This middleware is to be positioned between the edge and gateway. The gateway connects the transmitter with the
and the cloud computing layers, successfully participating middleware. The process therefore briefly illustrates the role

30668 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

FIGURE 12. Touch-enabled flow process at the destination end.

of the middleware infrastructure with its connectivity at both to the server after establishing a connection with the server
the transmitter and receiver end devices. via handshaking mode. The LS at the receiver end in turn
This middleware exists at the junction of edge and cloud indicates whether or not the channel is available for signal
computing layers. Middleware performs the virtualization transmission utilizing the request and acknowledgement pro-
process in conjunction with network slice abstraction and cess. The signal is processed as the output after the channel
allocation, fragmenting the existing physical network into availability is declared at the LS. The receiving end of the
multiple virtual networks. The techniques and algorithms proposed system may perhaps be represented by point ‘B’.
required for the proficient implementation of the slices bear- This point ‘B’ may further endow the requested TI enabled
ing n number of virtual networks may perhaps be applied in touch-based applications at the destination end. Hence the
the middleware layer. The middleware further connects to the user/customer successfully gets the desired output. It may be
LS on the receiver side and establishes a connection. a successful telesurgical operation or it may a remote indus-
The process is similar to that in the transmitter and is trial operation. It may also consist of the successful model-
followed by the middleware forwarding the information data ing and assembling of robotic and machinery components.

VOLUME 11, 2023 30669

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

in a sequence considered and governed by factors like opti-

mization, priority, power consumption, and security.
The next subsection further elaborates on the architecture
and functioning of the middleware to be used in the proposed
touch-based system. Hence at the output end, these factors
may be implemented as their corresponding techniques and
algorithms like the layer optimizing algorithm, the priority-
based algorithm, the power usage algorithm, and the security-
based algorithm, to obtain the desired output result. The
AI/ML/DL algorithmic implementation of the above-listed
factors, in wireless IoT and B5G/6G systems, has been a
research interest of many scientists and researchers.
FIGURE 13. Middleware component system.
Thus for the real-time implementation of the intelligent
touch-enabled system, we call for an interfacing architecture
Additional applications vary from the smart and virtual class- that integrates the existing B5G technology with the 6G
room to the virtual holographic shopping and e-commerce, technological interface, enabling the TI enabled touch con-
with successfully accessing the desired product without actu- nectivity through intelligent network slicing. Consequently,
ally having to lay a hand on it. It is possible to virtually access the subsequent sections describe the interfacing architecture
and analyzes the product through holographic imagery and of the present scenario with the next-generation system along
the transmission and reception of the information may be with its future scope and applications towards the implemen-
established by implementing the TI based touch configured tation of the intelligent touch system.
system, which will prove to be an impacting factor in the 6G
system.
IX. TOUCH TECHNOLOGY IOT MIDDLEWARE
The IoT is a vast network of connected smart devices that
C. TOUCH-BASED MIDDLEWARE SYSTEM
aim to make the surrounding environment intelligent and
The middleware system is an integral and the most important
autonomous. Thus most vendors do not care about the com-
component of the proposed intelligent touch configured sys-
patibility of their products with other competitive brands,
tem. It, therefore, helps in integrating and configuring both
which is one of the major challenges that IoT paradigms
the transmitter and the receiving user/application ends. Hence
face in the M2M communication. This problem has been
both its connecting ends must have the same configuration
addressed in several ways, one of which is the enforcement
and dimensionality for the system to act smoothly and with-
of universal standards, which is extremely difficult to imple-
out any undesired latency. The LS at the user end routes
ment. Another approach has been suggested, which is the
the requested information via BS, router and gateway to
implementation of middleware software to facilitate commu-
the middleware infrastructure. At this juncture, the informa-
nication between these devices.
tion requested is virtualized in numerous network functions
Middleware can be defined as the software that offers inter-
enclosed in several network slices.
operability between incompatible applications and devices;
This system also comprises a layered infrastructure, and we
as well as shielding customers from the smart object’s com-
have considered a 4-layered centralized middleware network
plexity. Hence, it serves as a software bridge between appli-
in Fig. 13. These layers may be categorized as the input layer,
cations and things, allowing IoT systems to communicate and
processing layer, delivery layer and output service layer. The
collaborate more effectively with one another. There are a
input layer forwards the input data from the transmitter end;
plethora of middleware solutions available, whether propri-
the processing layer helps analyze the data from the trans-
etary or open-source, that is provided by various companies,
mitter end; the processing layer helps analyze the data using
with the majority of these solutions being very similar to one
the virtualization and network slicing process. The delivery
another.
layer categorizes the slices as per their use cases and data
However, there are no benchmarks, performance indica-
rate and bandwidth requirements, whereas the output service
tors, or performance measurements that allow us to compare
layer arranges the final application slices to be delivered at
various systems. Following our examination of several arti-
the destination end.
cles [152]–[154], we have come to the following conclusions
Therefore the proposed requisite intelligence in the sys-
about some of the issues faced by IoT middleware:
tem is to be established via this middleware infrastructure.
There is a bilateral feedback system at both the transmit- 1) ABSTRACTION AND INTEROPERABILITY: IoT
ter and receiver end which helps in transmitting the feed- middleware aids in allowing various smart devices to
back and incorporating the intelligent techniques/algorithms interface easily with each other to facilitate collabora-
at both the connecting ends. The input to this system is tion and data exchange among heterogeneous devices.
through the transmitter user end represented by ‘A’. Conse- 2) DEVICE DISCOVERY AND MANAGEMENT: This
quently, the input data fed to the middleware may be arranged attribute allows IoT devices and services to be located

30670 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

in their network domain where the IoT environment

infrastructure is primarily dynamic because all newly
joined devices must announce their existence and
services.
3) SCALABILITY: The IoT middleware must be scalable
and must provide APIs to list all IoT devices, their
capabilities, and their services, among other things.
4) DEVICE CATEGORIZATION: APIs must also allow
users to categorize devices based on capabilities, man-
age devices based on remaining energy, report IoT
device problems to users, and perform problem load-
balancing procedures.
5) BIG DATA AND ANALYTICS Because of the fragile
nature of wireless sensor networks, part of the detected
data may be incomplete, requiring the middleware to
take this into account and extrapolate incomplete data
using a suitable machine learning method.
6) PRIVACY: Since this majority of data coming from
IoT applications and services is related to human daily
life, security and privacy issues must be considered
when transferring and processing it, necessitating the
development of middleware that addresses these issues.
7) CLOUD SERVICES: Cloud computing is the vital
layer of any IoT system. The data captured through
sensors will be stored and analyzed in a centralized FIGURE 14. Cloud and IoT enabling technologies.
cloud, and, as a result, IoT middleware should run well
in various types of clouds as shown in Fig. 14.
8) CONTEXT DETECTIO: Ambient data collection
applications and real-time reactive applications are the
two types of IoT applications. In the first, sensors col-
lect data that will be processed offline later to obtain
reasonable information that will be used for similar
scenarios in the future, while in the second, systems
must make a real-time decision based on the sensed
data.

A. ARCHITECTURE OF TOUCH-BASED IOT MIDDLEWARE

Following are types of IoT middleware architectures that are
currently available, which are classified based on the services
that they provide [155]:

1) SERVICE-ORIENTED ARCHITECTURE (SOA) OR

SERVICE-BASED SOLUTION
Users and developers are given the ability to employ or add
different types of IoT devices that can be used as services
in the service-oriented middleware architecture (SOA). Three
layers make up SOA architecture: the physical layer, which
contains actuators and sensors, the virtualized layer, which
contains cloud and infrastructure servers that are responsible
for performing various computational operations, and the
Applications layer, which contains all services and utilities as FIGURE 15. Service-oriented IoT middleware architecture.
shown in Fig. 15. Access control, storage management, and
event management are just a few of the general intermediate
layer services accessible. powerful gateway that sits between the cloud levels and the
SOA is a powerful middleware that may be deployed IoT layer. As a result, this sort of middleware is incompat-
on nodes that communicate with cloud servers or on a ible with resource-constrained devices and does not allow

VOLUME 11, 2023 30671

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 12. Comparison of commonly used service-based middleware attributes.

for device-to-device communication. The most widely used

service-based middleware is discussed here and summarized
in Table 12.
a) Link Smart (Hydra): A web service platform that
intends to reduce the heterogeneity of different devices
and entities in the IoT ecosystem, as well as control all
types of smart devices regardless of their communica-
tion protocols, such as Zigbee, RFID, Bluetooth, Wi-
Fi, and so on. This middleware is unique and it allows
IoT devices to be used as services by embedding the
required services. Health care, agriculture, and home
automation are a few of the IoT applications that can
be managed with Link Smart [156].
b) Kaa: It is an open-source platform that enables the
creation of IoT solutions and is administered by Cyber
vision Inc and Kaa IoT technologies. By utilizing a
web-based GUI built on the Apache platform, it is
possible to create data delivery schemes, support multi-
tenancy on servers, and generate endpoint software
development kits (SDK). Kaa enables direct or indirect FIGURE 16. Cloud-based IoT middleware architecture.
communication with endpoint devices while encrypt-
ing their data using AES29 and RSA30 encryption series of wrappers to feed the system with collected line
methods [154]. data that is later processed using XML specification
c) Global Sensor Network (GSN): Its objective is to files [157].
provide a standardized platform that enables adaptable d) Thing Speak IoT: It is a MATLAB-developed ana-
deployments, sharing, and integration of heterogeneous lytic open-source platform service that allows people
Internet of Things (IoT) objects. This platform is built and things to communicate. To collect, visualize, and
to meet the requirements of physical and virtual sensors analyze real-time data from devices and sensors, Thing
and actuators, whether they are connected via a wired Speak provides users with tools that allow them to use
connection or wirelessly. GSN is a Java platform that the HTTP protocol over the internet to store and retrieve
can be placed on IoT cloud or servers, and it allows a data from them [158].
e) Aura: This middleware facilitates the development
29 AES: Advanced Encryption Standard. of pervasive mobile IoT applications by abstracting
30 RSA: Rivest Shamir Adleman. device differences and allowing them to communicate

30672 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 13. Comparison of commonly used cloud-based middleware attributes.

freely. Aura makes an effort to optimize the screen decisions without the need for infrastructure manage-
backlight and CPU to improve performance while also ment by utilizing AWS services such as Amazon Kine-
reducing power consumption. In engaging with events, sis and Amazon Cloud Watch. AWS IoT customers can
Aura uses two concepts: proactive and reactive. In a also always monitor all devices that communicate with
proactive concept, system layers respond immediately their applications [160].
to the higher layer, whereas in a reactive concept, all b) Azure IoT Hub: It is a central platform developed by
layers adapt their resource and performance based on Microsoft for managing bidirectional communication
demand [159]. between IoT applications and the devices to which they
are linked. Because of Azure’s extensive capabilities,
2) CLOUD-BASED MIDDLEWARE it enables clients to develop full-featured, scalable IoT
User options are limited in the cloud-based middleware solutions that provide secure and reliable communica-
framework due to the limited number and variety of smart tion between the hosted cloud and connected devices.
devices connected to IoT applications. In addition, because Azure is a Microsoft product. When it comes to control-
different use cases can be programmed and then determined ling IoT connected devices, Azure IoT Hub supports
in advance, the sensed data can be collected and interpreted a variety of messaging patterns, including request-
with relative ease and accuracy. The operational component response, file upload from devices, and device to cloud
of this middleware is restricted by the resources available in telemetry [161].
the cloud computing environment. c) IBM Watson IoT: This platform, which is built on top
Although IoT functions have a general presence in the IT of the IBM Cloud, allows users to connect and control a
architecture, like storage systems and computation engines, variety of IoT appliances, sensors, industries, and home
these functions are represented and controlled by APIs where appliances. Using IBM Watson, its clients can create
IoT services are accessed by either cloud-based RESTful and manage their own IoT applications and appliances.
APIs or by vendor-provided applications as shown in Fig. 16. They can also extract KPIs from their data and use
The most widely used cloud-based IoT middleware is listed them to control their tools and applications, as well as
ahead and summarized in Table 13. process their collected data using historical and real-
a) AWS IoT: To manage cloud services, such as allowing time analytics. IBM Watson also offers a blockchain
millions of connected devices to communicate securely service [162].
and easily with other devices and cloud applications, d) Google Cloud IoT Essentially, it is a fully managed
Amazon developed this platform. AWS IoT enables device that is composed of a set of tools that pro-
customers to build IoT applications that collect, pro- vide a comprehensive solution for secure and easily
cess, analyze, and sense data to make appropriate connecting and processing of data generated, whether

VOLUME 11, 2023 30673

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 14. Comparison of commonly used actor-based middleware attributes.

they are located in the cloud or at the network edge. a) Calvin: It is an open-source IoT platform developed
Google Cloud IoT aspires to develop models capable of by Ericsson to be used on energy-constrained smart
optimizing client business, anticipating problems, and devices because it offers a portable and lightweight uni-
increasing operational efficiency [154]. fied programming architecture with input and output
e) Oracle IoT: A cloud-based service platform that lets ports that define the interfaces. Furthermore, Calvin
users create a real-time IoT solution to be linked with can also be used at the edge of IoT ecosystems to
enterprise applications while leveraging rigorous secu- reduce long-distance connections, lowering latency and
rity cloud capabilities and cutting-edge edge analytics. reducing the power consumption of IoT devices. Their
Furthermore, it integrates IoT data quickly and easily main advantage is their ability to move from one envi-
into customer business. Oracle IoT enables clients to ronment to another [164].
connect their devices to the cloud, which will aid them b) Node-RED It is an open-source IoT platform devel-
in making critical strategies and decisions in their busi- oped by IBM and built on the node.js31 programming
nesses [163]. language. Because of its small footprint, this platform
can be used at the edge of an IoT network, while on the
3) ACTOR BASED MIDDLEWARE FRAMEWORK server-side, a JavaScript platform with an event-driven
In terms of functionality, it is a lightweight middleware that module and non-blocking I/O is used. It allows users
can be implemented at the sensory, gateway, and cloud com- to build IoT applications by dragging and dropping
puting layers. Unlike other middleware, the computational connected blocks that represent IoT components. The
operations of this middleware are distributed across multiple platform drawbacks include the fact that it does not
layers, including the sensory layer and mobile access layer. support service discovery and only allows for password
The sensory swarm is the outermost layer, made up of sensors authentication for security [165].
and actuators, while the mobile access layer, made up of c) Ptolemy Accessor Host: Professor Edward Lee cre-
gateways, smartphones, Raspberry Pi, and laptops, is the ated this open-source platform in 1996 to design, sim-
intermediate layer, and the cloud is the innermost layer. The ulate, and model embedded and real-time devices. The
middleware which is also the actor host is designed to be underlying concept of this platform is that an IoT
lightweight and can be embedded in any layer of the appli- system is constructed from software components that
cation stack as shown in Fig. 17. interact and communicate with one another through
A storage device, for example, may not be included in messages sent through interconnected ports on a com-
the actor-based middleware used on a smartwatch. If the puter network [166].
storage device is provided by an actor, it can be downloaded d) Akka: It is a collection of open-source libraries and
from a cloud repository whenever it is required. The most a free actor-based platform that was created to allow
widely used actor-based middleware is discussed here and 31 Node.js: It is an open source, cross platform and run time environment
summarized in Table 14. for executing JavaScript code on the server side.

30674 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 15. Comparison of commonly used event-based middleware attributes.

developers to create distributed and run-time appli- computational units they support, and the locations where the
cations in either the Java or Scala programming lan- IoT middleware can be embedded or deployed.
guages. It enables users to meet business requirements IoT middleware based on SOA is implemented on servers
without having to write large low-level code, result- and in the cloud. Because the middleware can be implemented
ing in high performance, fault tolerance, and reli- in all tiers and IoT devices can perform computation where it
ability. Akka also allows multi-threading, isolates is most advantageous, actor-based delivers the best latency
communication between applications and their devices, and scalability for large scale linked IoT devices. While
and provides a clustered architecture with excellent these architectures provide some level of security and privacy,
availability [167]. cloud-based architecture requires users to place their trust in
the cloud provider to protect the privacy and integrity of their
4) EVENT-BASED MIDDLEWARE FRAMEWORK data.
Since a middleware cannot be implemented within the
To improve the development of distributed systems, middle-
physical device and the data exchanged between physical
ware that supports the publish/subscribe paradigm is being
devices and the middleware can be compromised, there is
developed and implemented. According to this definition,
a weak security link between the physical devices and the
this paradigm is a communication infrastructure that aims
middleware in both service and cloud-based architecture.
to provide clients with general-purpose services by assisting
To ensure QoS, the middleware must have a service discovery
them in dealing with the heterogeneity and complexity of
component that allows new services to be made accessible on
large-scale and distributed environments as shown in Fig. 18.
demand and failed services to be dynamically replaced. The
The event-based middleware hides some of the complex-
most widely used event-based middleware is discussed below
ity of distributed applications from the programmer, which
and summarized in Table 15.
will make it easier to develop and program much different
functionality in the future. The most significant differences a) Hermes: An event-based, scalable middleware
between these architectures are their openness to supporting designed to make distributed and large-scale applica-
new IoT device types, the types of middleware services or tions easier. To address big size and dynamic situations,

VOLUME 11, 2023 30675

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

applications and the creation of a robust, redundant,

publish/subscribe and content-based multi-broker. This
middleware is completely secured and offers simple
yet efficient routing and event handling. It also uses
a messaging information flow paradigm (BKS+99)
to specify communication between publisher and sub-
scriber [168].
c) Rebeca: This middleware is based on publish/subscribe
technology and focuses on the design of efficient
routing algorithms and the use of professional soft-
ware engineering methodologies to implement large-
scale business applications. To avoid network flooding,
Rebeca employs advanced routing techniques. It incor-
porates interoperability and subscription merging capa-
bilities into its services to facilitate location mobility
and reduce the size of the routing table. The event scope
function abstracts away the implementation details of
a service, such as transmission policies, security, data
transmission methods, external and internal interfaces,
and notification representation [169].
d) Fi WARE: It enables distributed IoT devices and appli-
cations to communicate in an efficient, flexible, secure,
and scalable manner. It was created to facilitate the
control and monitoring of a variety of IoT applications,
including logistics, retail, and smart cities. This plat-
form is comprised of numerous components, including
APIs, reusable modules, and massive code, all of which
enable a user to create an IoT application. A collec-
FIGURE 17. Actor based IoT middleware architecture. tion of sensed data from IoT sensors is captured via
REST API and later sent to a dedicated server called
the broker. FiWare has developed an API for querying
and storing various IoT contents, which enables any
application registered as a content consumer to retrieve
the necessary data from the broker. This platform has
a component called an adapter that is in charge of
transferring a certain type of material to subscriber
applications 170].

X. INTERFACING ARCHITECTURE WITH THE EXISTING

NETWORK
The architecture in Fig. 19 represents the real-time interfac-
FIGURE 18. Event-based IoT middleware architecture.
ing of the existing 5G/B5G network scenario with the next
generation 6G through the network slicing phenomenon. The
Hermes provides self-managed event brokers based on slicing aspect comprises intelligent cloud slicing, RAN slic-
a P2P routing layer. It features an adaptive solution ing and application slicing. Cloud slicing intends to facilitate
that takes into account failed event-broker events and cloud computing and storage access to the edge devices and
routing stacks, all while maintaining the event-broker users via different slices catering to different users simul-
network. Hermes middleware has two versions, both taneously. The RAN slicing is beneficial in connecting and
of which share the majority of the codebase and are routing the end users and devices to their receiving ends,
intended for use in distributed and large-scale sys- accordingly making use of techniques like spectrum sharing,
tems as well as in communication and implementation beamforming, cognitive antenna and radio transmission, tac-
among event brokers [155]. tile support and transmission system through different slices.
b) Gryphon: It is a highly scalable publish/subscribe All the applications accessed at the user end owing to the
middleware designed to distribute large amounts of slicing phenomenon are furnished by the application slic-
real-time data over the network. Gryphon is a Java- ing technique. The applications accessed may range from
based interface that enables the development of web robotics, tactile, haptic and touch-based communications

30676 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

to real-time online gaming using the AR/VR/XR with the than their restricted instruction library. The introduc-
UHD video streaming. Other applications may range from tion of touch-based sensation/actuation into the frame-
the remote industrial application (IIoT), telemedicine and work enables them to automatically respond and act
telesurgery, smart classroom with UHD video streaming. The in response to the sensory simulations provided by the
autonomous vehicular and UAV system may be operated environment, thereby saving time that would otherwise
by application slice providing the IoT connectivity along be spent on user instructions, programming, and the
with the URLLC of 1ms. The complete autonomy of the interface itself.
system and devices connected to the internet with the min- In this way, the touch-enabled robotic-human interaction sys-
imum latency and high data rate facilitate an intelligent and tem with emotional intelligence is illustrated in Fig. 20.
autonomous functioning system.
The smart cities comprising of smart homes, smart traf-
B. CASE 2-AR/VR BASED
fic monitoring systems and the rest instil another level of
ENTERTAINMENT/WWWWW/SHOPPING EXPERIENCE
intelligence in the existing wireless network system. The
subsequent section highlights some of the future aspects AR and VR technology are becoming the two important
concerning the application of intelligent technologies in the innovation factors that promote technological progress. Here
wireless 6G communication system followed by some of the AR glasses provide a realistic perspective for viewing
the collaborative and ongoing projects implemented in the augmented reality content, while VR headsets provide an
B5G/IoT and 6G wireless communication systems. immersive sensory experience. Therefore AR and VR tech-
nology are becoming the two important innovation factors
that promote technological progress. The virtual reality head-
XI. APPLICABLE USE CASES OF INTELLIGENT TOUCH
sets provide users with an immersive sensory experience by
TECHNOLOGY
allowing them to view AR content from a realistic perspec-
This section explores the touch technology framework and
tive. At the same time, AR/VR technology has opened up a
future application elements, with a focus on tactile-based
slew of new advertising and marketing possibilities. Around
haptic communication integrated with learning approaches
75 percent of business owners anticipate adopting AR/VR
to enable efficient network transmission. The touch-enabled
technology in the next two years, and global AR/VR market
telediagnosis, robotic interaction, and haptic sensation-based
spending will be more than double. Moreover, AR/VR users
shopping experience are just a few of the potential research
are becoming more common in the world. As consumers
aspect areas to look into.
increasingly rely on e-commerce and online shopping, there
is a high demand for augmented reality content. WIMI plays
A. CASE 1- ROBOTIC INTERACTION a significant role in the AR shopping market and also uses
Another application of touch technology could be tactile- ‘‘AI+AR’’ to support other industries such as advertising,
based robotic interaction between humans and robots, with entertainment, and e-commerce [174].
the bot system being able to recognize and respond to Immersive AR/VR solutions help bridge the gap between
human emotions [171], [172]. Emotions are interfaced into online and offline purchasing for consumers, and more and
the machinery by creating a database in the system using more people are seeing AR/VR as a valuable tool for dis-
AI/ML algorithms. The following are some examples of such covering products and getting brand services. AR/VR can
applications: currently improve the shopping experience through virtual-
1) ROBOTIC PETS: It entails the integration of AI algo- ization, with technology such as a virtual mirror assisting in
rithms and functionality into robotic systems, as well a virtual garment trial before purchase is implemented [175].
as the introduction of emotional intelligence (EI) into Hence it acts as a future aspect that complements the inte-
them. The robotic pet dog [173] is a dog-shaped robot gration of tactile/haptic-based intelligence with AR/VR tech-
capable of learning and detecting human gestures, nologies to further enrich our virtual shopping experience
face and eye movement, and responding appropriately. using touch technology.
By introducing a touch-based interface into the system, In this regard, the illustration below shows how the pro-
these robots are capable of reacting to human touch posed methodology can be integrated with existing AR/VR
feelings. Those who are elderly or mentally ill and are technology to further enhance the real-time user experience.
looking for companionship to relieve their loneliness The Fig. 21 illustration above depicts a graphical representa-
will find these useful, as they will aid in their emotional tion of the virtualized shopping experience provided to users
development as a result of the interaction. via intelligent touch technology.
2) CLEANING/DOMESTIC ROBOTS: These robotic
systems are capable of following user instructions and C. CASE 3-TACTILE AND HAPTIC SENSATION BASED
performing household chores. These might be driven TELEDIAGNOSIS FOR CONTACT-FREE COVID-19 CASES
by popular AI-based home automation systems like EXAMINATION
Alexa to provide complete automation, allowing them Tactile internet-enabled wireless communication systems
to hear and act on the user’s vocal commands rather have been integrated into the conventional healthcare system

VOLUME 11, 2023 30677

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

FIGURE 19. Intelligent interfacing architecture of the existing communication system with the next-generation technologies (B5G/6G).

to create a smart healthcare system. Tactile-enabled health- intelligence (AI) into the system while providing instructions
care systems, such as telemedicine, telesurgery, and remote through the tactile communication network. Additionally, if a
telediagnosis, could all benefit from the suggested touch tech- sense of touch or emotion is introduced in the form of emo-
nology. In today’s technological environment, let us consider tional intelligence (EI), the robot will be able to comprehend
a practical scenario in which a medical surgeon is working and execute the user’s instructions without delay. This method
from a smart surface or console that is connected to a telecom- could be used for both remote and local diagnoses.
munications network in Chennai, India, while the patient is In the current covid-19 pandemic, localized telediagnosis
lying on an operation table at Fortis Hospital in New Delhi employing robots could be used to allow on-duty clinicians to
thousands of miles away [176]. undertake contactless diagnosis and testing on infected indi-
The medical surgeon can remotely control the movement viduals. That, in turn, will lessen their chances of contracting
of a multi-armed surgical robot to perform gall bladder the virus from the patients and thus lessen the strain on them
surgery on the patient by utilizing the smart surface and other during such a pandemic crisis or emergency.
communication technologies. Through the use of a tactile
communication network, the doctor can communicate with XII. RESEARCH CHALLENGES AND FUTURE ASPECTS
and instruct the robot, while at the patient’s end, a multi- This section comprehends some of the conclusive future
machinery robot performs operations on the patient under the aspects concerning the incorporation of the reconfigurable
doctor’s instructions. Introducing intelligence into the robotic and intelligent technologies like AI/ML/DL in the B5G/6G
system, which allows the robot to learn on its own while and IoT governed wireless communication system. In the
performing operations, can further enhance the technological end, intelligent technology machine learning has become
benefits of the proposed touch technology as illustrated in one of the promising tools of artificial intelligence for the
Fig. 22. intelligent integration of wireless communications in the next
The procedure will be made easier if the machine is capable generation. Their futuristic scope may be steered towards
of sensing and responding to the user’s touch sensations and their relevance in channel estimation and detection, infer-
interactions. Thus, an attempt is made to incorporate artificial ring user location and behavior, resource allocation, iterative
30678 VOLUME 11, 2023
M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

FIGURE 20. Touch induced human to robotic interactions with emotional intelligence.

learning, computational intelligence like neural networks and 4) SERVICE LEVEL AGREEMENT: To provide cus-
the decision-making process. Some of these have been dis- tomers with an agreed level of service, three
cussed below: components should be considered: a model that pre-
cisely defines all functional and nonfunctional services
A. RESEARCH CHALLENGES required by consumers, automatic service to ensure a
1) INTELLIGENT CHANNEL ESTIMATION: The high level of QoS and adaptation, and a monitoring
futuristic scope in the ML-based channel estimation tool for SLA services. human intervention in current
technique lies in the fact that it may be directly put autonomous networks will be phased out in favor of
to use in the scenarios without any need for training. the development of intelligent IoT devices.
the only way to be able to learn the channel features 5) QOS LEVELS: Since there is no mechanism in place
of various user environments can be that a system is to guarantee a specific level of QoS for non-functional
as smart to understand its parametric needs or in other IoT services, researchers should develop procedures for
words that such a generalized system requires a vast optimizing and monitoring QoS levels.
amount of pre-collected communication data to be used 6) PRIVACY AND SECURITY: As a result of the
by ML/DL algorithms. resource-constrained devices in IoT environments, the
2) NON/SEMI-AUTONOMOUS DEVICE DISCOV- majority of autonomous and semi-autonomous applica-
ERY: Human intervention in IoT components, such tion services restrict security, authorization and authen-
as device discovery, renders these applications non- tication mechanisms, among other things. Hence for the
scalable and prone to error. because of this limita- intelligent network to communicate between the cloud,
tion, interacting IoT devices like IoT middleware is gateway, and sensors securely and efficiently, security
unsuitable for self-adaptive applications such as M2M and privacy must be both end-to-end and lightweight.
communication systems.
3) HETEROGENOUS ENVIRONMENTS: Since most B. FUTURE RESEARCH ASPECTS
smart IoT devices and middlewares only handle one or 1) ALGORITHMIC MODELING: Almost all relevant
two types of heterogeneous components, this is consid- and physical modeling/construction should be seen as
ered a major issue that must be addressed. since non- an integral step toward algorithmic implementation
autonomous and inflexible services and devices restrict using applicable ML tools, and DNN is a key technol-
the support of iot applications, new approaches must ogy component in this regard.
address and resolve the heterogeneity of IoT environ- 2) ROBOTIC FEELINGS: The implementation of tac-
ments, particularly in large scale networks. tile based haptic communication in conjunction with

VOLUME 11, 2023 30679

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

FIGURE 21. Touch technology-enabled virtualized shopping experience.

touch-enabled gesture recognition is another applica- 6) BEHAVIORAL LEARNING: Predicting user behav-
tion that could benefit from this technology. This could iors will result in better network resource utilization
be used in the inculcation of feelings and emotions in and will allow us to optimally allocate end-to-end
robotic systems. network sources in an online fashion, which would
3) INTELLIGENT HEALTH MEASURES DURING be impossible without the assistance of AI and ML
PANDEMIC LIKE COVID-19: Telemedicine, remote techniques.
surgery, and treatment using intelligence and haptic
sensations may also benefit from it, requiring a suit-
XIII. RECENT RESEARCH AND PROJECTS CONCERNING
able algorithm with appropriate ML tools for effi-
6G WIRELESS COMMUNICATION SYSTEMS
cient implementation and risk-free operation. So it may
Numerous 6G research and development activities have
also be useful in the contactless administration to the
already begun on a global scale and this section summarizes
patients during pandemics such as the one that occurred
the most significant 6G research activities underway at the
in COVID-19. In this way, the risk of doctors becoming
moment [68].
infected with the virus is reduced because the diagnosis
will be done by a robotic system that will use haptic and
touch sensations that have been introduced. A. 6G FLAGSHIP (MAY 2018-APRIL 2026)
4) EFFICIENT NETWORK MANAGEMENT: As a The 6G Flagship [177] is an eight-year research initiative that
result, a paradigm shift is required for the efficient focuses on the wireless smart society and ecosystem enabled
design of B5G/6G networks to leverage AI/ML and by 6G technology. Being funded by the Academy of Finland,
take advantage of big data analytics to improve the this project intends to realize B5G networks from the very
overall performance of future networks. outset towards its phase of commercialization and to develop
5) PROACTIVE NETWORKING: Future networks the new 6G standards for future digital societies. It aims
require a prediction mechanism to help predict and to develop essential technology components of 6G mobile
anticipate the future while allocating network resources networks in areas such as wireless connectivity, distributed
proactively. As a result, it aids in the prediction and intelligent computing, security, and privacy.
analysis of traffic patterns while determining off-peak In addition to human-to-human communication, the
times on various spectrum bands so that incoming research focuses on communication between devices, pro-
traffic demands can be properly allocated over a given cesses, and objects. This, in turn, enables a highly automated,
window. smart society that will permeate all aspects of life. Ultimately,

30680 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

FIGURE 22. Touch technology-enabled contactless telediagnosis and surgery.

the 6G flagship project is to conduct large-scale pilots with a TABLE 16. Industrial avenues on 6G developments.
test network using industry and academic support.

B. HEXA-X: A FLAGSHIP FOR 6G VISION AND

INTELLIGENT FABRIC OF TECHNOLOGY ENABLERS
CONNECTING HUMANS, PHYSICAL AND DIGITAL
WORLDS (JAN 2021-JUNE 2023)
HEXA-X [178] is the first flagship project of the European
Commission for implementing the 6G vision and establishing
an intelligent fabric of technology enablers for integrating the
human, physical, and digital worlds. HEXA-X is a European
industry-academic collaboration that intends to prepare the
way for the next generation of wireless networks through
exploratory research. Its objective is to connect humans, with
the physical and digital worlds through a fabric of 6G essen-
tial enablers. To achieve this goal and vision, the HEXA-X
project is concentrating on building critical technical enablers
in the following areas:
1) High-frequency and high-resolution long-range access
using New Radio access technologies.
2) Connected intelligence to future networks via frameworks. It will also provide new features for traffic flow
AI-powered air interface. aggregation, service layer management, infrastructure layer
3) Network disaggregation and dynamic dependability are network equipment integration, AI/ML-based security and
enabled by 6G architectural drivers. forensic evidence for multi-tenancy networks.

C. TERA FLOW: SECURED AUTONOMIC TRAFFIC D. DAEMON: NETWORK INTELLIGENCE FOR ADAPTIVE
MANAGEMENT FOR A TERA OF SDN FLOWS (JAN AND SELF LEARNING MOBILE NETWORKS (JAN 2021-
2021-JUNE 2023) JUNE 2023)
Tera flow [179] is working on developing a cloud-native The major goal of the DAEMON project [180] is to enable
SDN controller for B5G/6G networks. This novel SDN con- high-quality Network Intelligence (NI) for 6G systems,
troller is compatible with the contemporary NFV and MEC which will completely automate network administration.

VOLUME 11, 2023 30681

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 17. Collaborative and ongoing projects in B5G and IoT wireless communication systems.

The project includes an end-to-end B5G/6G NI architecture, F. SOUTH KOREA MSIT 6G RESEARCH PROGRAM
which can be fully coordinated with the NI-assisted features. The Ministry of Science and ICT (MSIT) [182] in
DAEMON performs a systematic analysis of each NI task that South Korea is working on a bold strategy to be the first
is solved using AI models and also provides a solid set of country to deploy 6G networks. 6G services are expected
guidelines for incorporating machine learning into network to be commercially available in Korea between 2028 and
functionalities. A major goal of the DAEMON project is 2030, according to the South Korean government. The ini-
to focus on existing B5G network-specific AI methods that tial deployment is expected in 2028, followed by a mass
go beyond the current trend of integrating AI into network commercial deployment in 2030, with a total investment of
controllers and orchestrators. $169 million in R&D for 6G technology.
The preliminary goal is to deploy a 6G pilot in five
E. 6G BRAINS: BRINGING REINFORCEMENT LEARNING important areas, including digital healthcare, immersive con-
INTO RADIO LIGHTWEIGHT NETWORK FOR MASSIVE tent, self-driving cars, smart cities, and smart manufacturing,
CONNECTIONS (JAN 2021-JUNE 2023) by 2026 [183]. The 6G research program’s objectives are as
The 6G BRAINS [181] project is focused on implementing follows.
multi-agent DRL for 6G radio links using AI. To improve 1) Attain a data rate of 1Tbps.
massive connection over D2D assisted highly dynamic cell- 2) Latency reduction of 0.1ms for wireless networks
free networks, a novel comprehensive cross-layer DRL driven 3) Increases the range of connectivity coverage to up to
resource allocation solution will be required to perform 10 km from the ground.
resource allocation for Sub 6GHz/mm-wave/THz/optical 4) AI must be integrated into the network to ensure that
wireless communication (OWC) medium. This significantly everything is covered.
improves the capacity, reliability, and latency of future indus- 5) By implementing security by design, it is possible to
trial, intelligent transportation, and e-health networks. protect the network.

30682 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

TABLE 18. Global-level ongoing projects on 6G alongside their associated applications.

G. JAPAN 6G PROMOTION STRATEGY improve the fusion of virtual and physical environ-
Japan invests approximately 50 billion dollars in the 6G ments. This will enable a new level of 6G digital ser-
development project. This initiative aims to strengthen col- vices by better connecting human senses with ambient
laboration between the public and private sectors in the field and remote data.
of 6G research and development. Furthermore, by 2025, this 2) UBIQUITOUS COVERAGE: It has a role in empha-
6G promotion strategy seeks to establish and exhibit the 6G sizing increasing the quality and range of 6G com-
system’s basic technologies, as well as put new technologies munication network coverage. The research will focus
into practice by 2030 [184]. on expanding coverage indoors, utilizing intelligent
surfaces, and satellite technology to enable 6G services
H. 6TH GENERATION INNOVATION CENTER to be available globally.
With the continuation of the 5th Generation Innovation Cen-
ter (5GIC), the University of Surrey in the United Kingdom I. INDUSTRIAL 6G PROGRAMS
launched the 6GIC [185] in 2020 to focus on 6G-related Table 16 summarizes several industrial research programs
research activities across two themes. focused on the development and implementation of 6G.
1) AMBIENT INFORMATION: There is a use of While Table 17 and Table 18 give an overall projection of
advanced wireless technologies, high-resolution the ongoing recent projects in B5G/IoT and 6G respectively
sensing, and highly accurate geo-location methods to in Appendix I

VOLUME 11, 2023 30683

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

XIV. CONCLUSION B. GLOBAL LEVEL ONGOING PROJECTS FOR 6G

Wireless technologies have had a drastic influence on the DEPLOYMENT
various facets of the human lifestyle, via interactions between Table 18 summarizes all the ongoing research works and
the individuals at the virtual level throughout most of the projects about the 6G wireless communication system along
applications. It has thus transformed many elements of life with their application.
including business, living standards and infrastructure. In a
never-ending quest for elegant solutions to various prob- ACKNOWLEDGMENT
lems, society is always on the lookout for new avenues of The authors gratefully acknowledge the support provided by
progress. This is therefore the motivation behind the seamless 5G and IoT Lab, SoECE, SMVDU, TBIC-SMVDU, TEQIP-
connectivity that has resulted in the evolution of wireless III, and the 6G and IoT Innovation Laboratory, IIIT Jabalpur.
communication from 1G to 5G. Despite the recent devel-
opment, work in this field is still underway all over the REFERENCES
world intending to deploy the 6G communication network [1] A. Goldsmith, Wireless Communications. Cambridge, U.K.: Cambridge
by 2030, thus making it one of the most in-demand research Univ. Press, 2005.
[2] (Mar. 2020). Cisco Annual Internet Report—Cisco Annual Internet
fields, with the potential to revolutionize the personal life Report (2018–2023) White Paper. CISCO. Accessed: Apr. 22, 2021.
of individuals and society, business and communication [Online]. Available: https://www.cisco.com/c/en/us/solutions/
collateral/executive-perspectives/annual-internet-report/white-paper-
systems. c11-741490.html
The real-time applications from the individual user’s point [3] N. Sharma, M. Shamkuwar, and I. Singh, ‘‘The history, present and
of view are perceptible in both professional as well as domes- future with IoT,’’ in Internet of Things and Big Data Analytics for Smart
Generation, vol. 154. 2019, pp. 27–51, doi: 10.1007/978-3-030-04203-
tic fields. It may vary from e-health, smart appliances to smart 5_3.
classroom-based enhanced learning as well as the certain [4] J. G. Andrews, S. Buzzi, W. Choi, S. V. Hanly, A. Lozano, A. C. K. Soong,
amount of intelligence incorporated in industrial automation, and J. C. Zhang, ‘‘What will 5G be?’’ IEEE J. Sel. Areas Commun.,
vol. 32, no. 6, pp. 1065–1082, Jun. 2014.
manufacturing and transportation of goods. The currently [5] A. Gupta and R. K. Jha, ‘‘A survey of 5G network: Architecture and
existing technologies make the IoT concept viable but do not emerging technologies,’’ IEEE Access, vol. 3, pp. 1206–1232, 2015, doi:
10.1109/ACCESS.2015.2461602.
necessarily fit well with the expected scalability and profi- [6] I. Farris, T. Taleb, H. Flinck, and A. Iera, ‘‘Providing ultra-short latency
ciency criteria of the 6G system. Therefore many problems do to user-centric 5G applications at the mobile network edge,’’ Trans.
exist in the technical process and therefore, a societal reform Emerg. Telecommun. Technol., vol. 29, no. 4, p. e3169, Apr. 2018, doi:
10.1002/ett.3169.
is vital in the technology to universally conceptualize the IoT [7] S. Parkvall, E. Dahlman, A. Furuskar, and M. Frenne, ‘‘NR: The new 5G
connectivity. radio access technology,’’ IEEE Commun. Standards Mag., vol. 1, no. 4,
Hence complete interoperability of the integrated devices pp. 24–30, Dec. 2017, doi: 10.1109/MCOMSTD.2017.1700042.
[8] A. Dogra, R. K. Jha, and S. Jain, ‘‘A survey on beyond 5G network with
is required, provided with a high percentage of smartness, the advent of 6G: Architecture and emerging technologies,’’ IEEE Access,
maintaining the trust, safety and protection in the system vol. 9, pp. 67512–67547, 2021, doi: 10.1109/ACCESS.2020.3031234.
[9] P. Gandotra, R. K. Jha, and S. Jain, ‘‘Sector-based radio resource
for the creation of technologies that emphasize technical allocation (SBRRA) algorithm for better quality of service and
requirements. However, the most intriguing aspect of the experience in device-to-device (D2D) communication,’’ IEEE
6G is the incorporation of an intelligent interface link- Trans. Veh. Technol., vol. 67, no. 7, pp. 5750–5765, Jul. 2018, doi:
10.1109/TVT.2017.2787767.
ing the existing communication standards with the recently [10] H. Wang, J. Wang, G. Ding, L. Wang, T. A. Tsiftsis, and P. K. Sharma,
researched ones, to competently fulfil the high data rate and ‘‘Resource allocation for energy harvesting-powered D2D communica-
the stringent latency requirements. tion underlaying UAV-assisted networks,’’ IEEE Trans. Green Commun.
Netw., vol. 2, no. 1, pp. 14–24, Mar. 2018.
In other words, all we lack is an intelligent interface [11] M. Shafi and R. K. Jha, ‘‘Half-duplex attack: An effectual attack mod-
that is capable enough to integrate and enable a tactile elling in D2D communication,’’ in Proc. Int. Conf. Commun. Syst. Netw.
based haptic (touch-based) communication in the existing (COMSNETS), Jan. 2020, pp. 879–881.
[12] S. Akbar, Y. Deng, A. Nallanathan, M. Elkashlan, and
B5G network right from its source to its destination, com- G. K. Karagiannidis, ‘‘Massive multiuser MIMO in heterogeneous
plete with its feedback mechanism, altogether incorporating cellular networks with full duplex small cells,’’ IEEE Trans. Wireless
Commun., vol. 65, no. 11, pp. 4704–4719, Nov. 2017.
intelligence in the system to satisfy the stringent latency [13] F. Al-Turjman, E. Ever, and H. Zahmatkesh, ‘‘Small cells in the forth-
requirement of less than 1ms. This paper, therefore, provides coming 5G/IoT: Traffic modelling and deployment overview,’’ IEEE
preliminary insight to and answers the above-mentioned Commun. Surveys Tuts., vol. 21, no. 1, pp. 28–65, 1st Quart. 2018.
[14] V. Hassija, V. Chamola, V. Saxena, D. Jain, P. Goyal, and B. Sikdar,
challenges by providing a comprehensive survey of the ‘‘A survey on IoT security: Application areas, security threats, and solu-
touch-based intelligent communication system using network tion architectures,’’ IEEE Access, vol. 7, pp. 82721–82743, 2019.
[15] (2017). Cloud VR-oriented Bearer Network White Paper. Accessed:
slicing and TI combining intelligence like AI and ML, in a Aug. 25, 2020. [Online]. Available: https://www-file.huawei.com/-/
6G network. media/corporate/pdf/ilab/cloud_vr_oriented_bearer_network_white_
paper_en_v2.pdf
[16] S. Sharma, R. Miller, and A. Francini, ‘‘A cloud-native approach to 5G
APPENDIX I
network slicing,’’ IEEE Commun. Mag., vol. 55, no. 8, pp. 120–127,
A. COLLABORATIVE AND ONGOING PROJECTS IN Aug. 2017.
B5G/IOT [17] R. El Hattachi and J. Erfanian, ‘‘NGMN 5G white paper,’’
NMGN Alliance, NGMN 5G Initiative, Feb. 2015, pp. 1–125.
Table 17 illustrates the recent collaborative projects in B5G Accessed: Aug. 25, 2020. [Online]. Available: https://www.ngmn.org/
and IoT based wireless communication networks. uploads/media/NGMN_5G_White_Paper_V1_0.pdf

30684 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

[18] A. Ghosh, A. Maeder, M. Baker, and D. Chandramouli, [37] F. Granelli, ‘‘Network slicing,’’ in Computing in Communication Net-
‘‘5G evolution: A view on 5G cellular technology beyond 3GPP works. Amsterdam, The Netherlands: Elsevier, 2020, pp. 63–76, doi:
release 15,’’ IEEE Access, vol. 7, pp. 127639–127651, 2019, doi: 10.1016/B978-0-12-820488-7.00014-1.
10.1109/ACCESS.2019.2939938. [38] J. Prados-Garzon, J. J. Ramos-Munoz, P. Ameigeiras,
[19] A. Ali and W. Hamouda, ‘‘Advances on spectrum sensing for cognitive P. Andres-Maldonado, and J. M. Lopez-Soler, ‘‘Modeling and
radio networks: Theory and applications,’’ IEEE Commun. Surveys Tuts., dimensioning of a virtualized MME for 5G mobile networks,’’ IEEE
vol. 19, no. 2, pp. 1277–1304, 2nd Quart., 2016. Trans. Veh. Technol., vol. 66, no. 5, pp. 4383–4395, May 2017, doi:
[20] W. Lu, P. Si, G. Huang, H. Han, L. Qian, N. Zhao, and Y. Gong, 10.1109/TVT.2016.2608942.
‘‘SWIPT cooperative spectrum sharing for 6G-enabled cognitive IoT [39] M. Richart, J. Baliosian, J. Serrat, and J.-L. Gorricho, ‘‘Resource
network,’’ IEEE Internet Things J., vol. 8, no. 20, pp. 15070–15080, slicing in virtual wireless networks: A survey,’’ IEEE Trans.
Oct. 2021. Netw. Serv. Manage., vol. 13, no. 3, pp. 462–476, Sep. 2016, doi:
[21] B. Soret, A. D. Domenico, S. Bazzi, N. H. Mahmood, and 10.1109/TNSM.2016.2597295.
K. I. Pedersen, ‘‘Interference coordination for 5G new radio,’’ IEEE [40] T. Taleb, A. Ksentini, and A. Kobbane, ‘‘Lightweight mobile core
Wireless Commun., vol. 25, no. 3, pp. 131–137, Jun. 2018, doi: networks for machine type communications,’’ IEEE Access, vol. 2,
10.1109/MWC.2017.1600441. pp. 1128–1137, 2014, doi: 10.1109/ACCESS.2014.2359649.
[22] G. Chopra, R. K. Jha, and S. Jain, ‘‘Rank-based secrecy rate improve- [41] Q. Zhang, M. Jiang, Z. Feng, W. Li, W. Zhang, and M. Pan, ‘‘IoT
ment using NOMA for ultra dense network,’’ IEEE Trans. Veh. Technol., enabled UAV: Network architecture and routing algorithm,’’ IEEE Inter-
vol. 68, no. 11, pp. 10687–10702, Nov. 2019. net Things J., vol. 6, no. 2, pp. 3727–3742, Apr. Apr. 2019, doi:
[23] M. Giordani, M. Polese, A. Roy, D. Castor, and M. Zorzi, ‘‘A tutorial 10.1109/JIOT.2018.2890428.
[42] P. K. Agyapong, M. Iwamura, D. Staehle, W. Kiess, and
on beam management for 3GPP NR at mmWave frequencies,’’ IEEE
A. Benjebbour, ‘‘Design considerations for a 5G network architecture,’’
Commun. Surveys Tuts., vol. 21, no. 1, pp. 173–196, Sep. 2019, doi:
IEEE Commun. Mag., vol. 52, no. 11, pp. 65–75, Nov. 2014, doi:
10.1109/COMST.2018.2869411.
10.1109/MCOM.2014.6957145.
[24] S. Bakri, P. A. Frangoudis, A. Ksentini, and M. Bouaziz, ‘‘Data- [43] A. Imran, A. Zoha, and A. Abu-Dayya, ‘‘Challenges in 5G: How to
driven RAN slicing mechanisms for 5G and beyond,’’ IEEE Trans. empower SON with big data for enabling 5G,’’ IEEE Netw., vol. 28, no. 6,
Netw. Service Manage., vol. 18, no. 4, pp. 4654–4668, Dec. 2021, doi: pp. 27–33, Nov. 2014.
10.1109/TNSM.2021.3098193. [44] N. H. Mahmood, S. Böcker, A. Munari, F. Clazzer, I. Moerman,
[25] N. Nomikos, E. T. Michailidis, P. Trakadas, D. Vouyioukas, H. Karl, K. Mikhaylov, O. Lopez, O. S. Park, E. Mercier, H. Bartz, and
J. Martrat, T. Zahariadis, K. Papadopoulos, and S. Voliotis, ‘‘A UAV- R. Jäntti, ‘‘White paper on critical and massive machine type commu-
based moving 5G RAN for massive connectivity of mobile users and nication towards 6G,’’ 2020, arXiv:2004.14146.
IoT devices,’’ Veh. Commun., vol. 25, Oct. 2020, Art. no. 100250, doi: [45] I. Afolabi, A. Ksentini, M. Bagaa, T. Taleb, M. Corici, and
10.1016/j.vehcom.2020.100250. A. Nakao, ‘‘Towards 5G network slicing over multiple-domains,’’
[26] A. Manzalini, C. Lin, J. Huang, C. Buyukkoc, and M. Bursell, ‘‘Towards IEICE Trans. Commun., vol. 100, no. 11, pp. 1992–2006, 2017, doi:
5G software-defined ecosystems: Technical challenges, business sus- 10.1587/transcom.2016NNI0002.
tainability and policy issues,’’ IEEE SDN, IEEE, New York, NY, [46] G. P. Fettweis, ‘‘The tactile internet: Applications and challenges,’’
USA, Work./Discussion Paper, White Paper, 2016. [Online]. Available: IEEE Veh. Technol. Mag., vol. 9, no. 1, pp. 64–70, Mar. 2014, doi:
https://discovery.ucl.ac.uk/id/eprint/10043678/ 10.1109/MVT.2013.2295069.
[27] P. P. Ray and N. Kumar, ‘‘SDN/NFV architectures for edge-cloud oriented [47] M. Simsek, A. Aijaz, M. Dohler, J. Sachs, and G. Fettweis, ‘‘5G-enabled
IoT: A systematic review,’’ Comput. Commun., vol. 169, pp. 129–153, tactile internet,’’ IEEE J. Sel. Areas Commun., vol. 34, no. 3, pp. 460–473,
Mar. 2021. Feb. 2016.
[28] K. K. Karmakar, V. Varadharajan, S. Nepal, and U. Tupakula, ‘‘SDN- [48] M. Geller and P. Nair, ‘‘5G security innovation with Cisco,’’ Cisco,
enabled secure IoT architecture,’’ IEEE Internet Things J., vol. 8, no. 8, 5G Infrastruct. Public Private Partnership (5G PPP), Cisco Public,
pp. 6549–6564, Apr. 2021. White Paper, 2018. [Online]. Available: https://infotech.report/
[29] A. A. Barakabitze, A. Ahmad, R. Mijumbi, and A. Hines, ‘‘5G net- Resources/Whitepapers/eed3eb22-b342-443c-99c1-41a96eb43512_5g-
work slicing using SDN and NFV: A survey of taxonomy, architec- security-innovation-with-cisco-wp.pd
tures and future challenges,’’ Comput. Netw., vol. 167, Feb. 2020, [49] S. J. Nawaz, S. K. Sharma, S. Wyne, M. N. Patwary, and
Art. no. 106984. M. Asaduzzaman, ‘‘Quantum machine learning for 6G communication
[30] A. Cardenas and D. Fernandez, ‘‘Network slice lifecycle networks: State-of-the-art and vision for the future,’’ IEEE Access, vol. 7,
management model for NFV-based 5G virtual mobile network pp. 46317–46350, 2019.
operators,’’ in Proc. IEEE Conf. Netw. Function Virtualization [50] K. David and H. Berndt, ‘‘6G vision and requirements: Is there any need
Softw. Defined Netw. (NFV-SDN), Nov. 2020, pp. 120–125, for beyond 5G?’’ IEEE Veh. Technol. Mag., vol. 13, no. 3, pp. 72–80,
doi: 10.1109/NFV-SDN50289.2020.9289883. Sep. 2018.
[31] A. El-Mekkawi, X. Hesselbach, and J. R. Piney, ‘‘Novel NFV aware [51] M. Latva-aho, K. Leppänen, F. Clazzer, and A. Munari. (2020). Key
network service for intelligent network slicing based on squatting- Drivers and Research Challenges for 6G Ubiquitous Wireless Intel-
kicking model,’’ IEEE Access, vol. 8, pp. 223041–223068, 2020, doi: ligence. University of Oulu. [Online]. Available: http://urn. fi/urn:
10.1109/ACCESS.2020.3044951. Isbn:9789526223544
[52] M. Maier, M. Chowdhury, B. P. Rimal, and D. P. Van, ‘‘The tactile
[32] G. Nencioni, R. G. Garroppo, A. J. Gonzalez, B. E. Helvik, and
internet: Vision, recent progress, and open challenges,’’ IEEE Commun.
G. Procissi, ‘‘Orchestration and control in software-defined 5G networks:
Mag., vol. 54, no. 5, pp. 138–145, May 2016.
Research challenges,’’ Wireless Commun. Mobile Comput., vol. 2018, [53] T. Taleb, S. Dutta, A. Ksentini, M. Iqbal, and H. Flinck, ‘‘Mobile edge
pp. 1–18, Aug. 2018, doi: 10.1155/2018/6923867. computing potential in making cities smarter,’’ IEEE Commun. Mag.,
[33] Z. Hou, C. She, Y. Li, T. Q. Quek, and B. Vucetic, ‘‘Burstiness-aware vol. 55, no. 3, pp. 38–43, Mar. 2017.
bandwidth reservation for ultra-reliable and low-latency communica- [54] X. Foukas, G. Patounas, A. Elmokashfi, and M. K. Marina, ‘‘Network
tions in tactile Internet,’’ IEEE J. Sel. Areas Commun., vol. 36, no. 11, slicing in 5G: Survey and challenges,’’ IEEE Commun Mag, vol. 55, no. 5,
pp. 2401–2410, Nov. 2018. pp. 94–100, May 2017, doi: 10.1109/MCOM.2017.1600951.
[34] S. Yin, Y. Zhao, and L. Li, ‘‘Resource allocation and basestation [55] A. Kaloxylos, ‘‘A survey and an analysis of network slicing in 5G
placement in cellular networks with wireless powered UAVs,’’ networks,’’ IEEE Commun. Standards Mag., vol. 2, no. 1, pp. 60–65,
IEEE Trans. Veh. Technol., vol. 68, no. 1, pp. 1050–1055, Mar. 2018, doi: 10.1109/MCOMSTD.2018.1700072.
Jan. 2019. [56] I. Afolabi, T. Taleb, K. Samdanis, A. Ksentini, and H. Flinck, ‘‘Net-
[35] Description of Network Slicing Concept, ‘‘NGMN 5G projects work slicing and softwarization: A survey on principles, enabling tech-
requirements & architecture work stream end-to-end architecture,’’ nologies, and solutions,’’ IEEE Commun. Surveys Tuts., vol. 20, no. 3,
NGMN Alliance, Tech. Rep. 1.0.8, Sep. 2016. [Online]. Available: pp. 2429–2453, 3rd Quart., 2018, doi: 10.1109/COMST.2018.2815638.
https://www.ngmn.org/ [57] M. E. Morocho-Cayamcela, H. Lee, and W. Lim, ‘‘Machine learning
[36] X. Foukas, G. Patounas, A. Elmokashfi, and M. K. Marina, ‘‘Network for 5G/B5G mobile and wireless communications: Potential, limita-
slicing in 5G: Survey and challenges,’’ IEEE Commun. Mag., vol. 55, tions, and future directions,’’ IEEE Access, vol. 7, pp. 137184–137206,
no. 5, pp. 94–100, May 2017. 2019.

VOLUME 11, 2023 30685

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

[58] M. Mohammadi and A. Al-Fuqaha, ‘‘Enabling cognitive smart cities [80] M. S. Elbamby, C. Perfecto, M. Bennis, and K. Doppler, ‘‘Toward low-
using big data and machine learning: Approaches and challenges,’’ IEEE latency and ultra-reliable virtual reality,’’ IEEE Netw., vol. 32, no. 2,
Commun. Mag., vol. 56, no. 2, pp. 94–101, Feb. 2018. pp. 78–84, Mar./Apr. 2018.
[59] V. P. Kafle, Y. Fukushima, P. Martinez-Julia, and T. Miyazawa, ‘‘Con- [81] T. Park and W. Saad, ‘‘Distributed learning for low latency machine type
sideration on automation of 5G network slicing with machine learning,’’ communication in a massive Internet of Things,’’ IEEE Internet Things J.,
in Proc. ITU Kaleidoscope, Mach. Learn. 5G Future (ITU K), Santa Fe, vol. 6, no. 3, pp. 5562–5576, Jun. 2019.
Argentina, Nov. 2018, pp. 1–8, doi: 10.23919/ITU-WT.2018.8597639. [82] M. Alsenwi, N. H. Tran, M. Bennis, S. R. Pandey, A. K. Bairagi, and
[60] A. Aijaz, M. Dohler, A. H. Aghvami, V. Friderikos, and M. Frodigh, C. S. Hong, ‘‘Intelligent resource slicing for eMBB and URLLC coexis-
‘‘Realizing the tactile internet: Haptic communications over next gen- tence in 5G and beyond: A deep reinforcement learning based approach,’’
eration 5G cellular networks,’’ IEEE Wireless Commun., vol. 24, no. 2, IEEE Trans. Wireless Commun., vol. 20, no. 7, pp. 4585–4600, Jul. 2021.
pp. 82–89, Apr. 2017, doi: 10.1109/MWC.2016.1500157RP. [83] M. A. Siddiqi, H. Yu, and J. Joung, ‘‘5G ultra-reliable low-latency
[61] K. Antonakoglou, X. Xu, E. Steinbach, T. Mahmoodi, and M. Dohler, communication implementation challenges and operational issues
‘‘Toward haptic communications over the 5G tactile internet,’’ IEEE with IoT devices,’’ Electronics, vol. 8, no. 9, 2019, doi: 10.3390/
Commun Surv Tutor., vol. 20, no. 4, pp. 3034–3039, 2018, doi: electronics8090981.
10.1109/COMST.2018.2851452. [84] H. Yang, A. Alphones, Z. Xiong, D. Niyato, J. Zhao, and K. Wu,
[62] A. Aijaz and M. Sooriyabandara, ‘‘The tactile internet for industries: ‘‘Artificial-intelligence-enabled intelligent 6G networks,’’ IEEE Netw.,
A review,’’ Proc. IEEE, vol. 107, no. 2, pp. 414–435, Feb. 2019, doi: vol. 34, no. 6, pp. 272–280, Nov./Dec. 2020.
10.1109/JPROC.2018.2878265. [85] C. Huang, A. Zappone, G. C. Alexandropoulos, M. Debbah, and C. Yuen,
[63] Y. Sun, M. Peng, Y. Zhou, Y. Huang, and S. Mao, ‘‘Application of machine ‘‘Reconfigurable intelligent surfaces for energy efficiency in wire-
learning in wireless networks: Key techniques and open issues,’’ IEEE less communication,’’ IEEE Trans. Wireless Commun., vol. 18, no. 8,
Commun. Surveys Tuts., vol. 21, no. 4, pp. 3072–3108, 4th Quart., 2019, pp. 4157–4170, Aug. 2019.
doi: 10.1109/COMST.2019.2924243. [86] N. Janbi, I. Katib, A. Albeshri, and R. Mehmood, ‘‘Distributed
[64] C.-X. Wang, M. D. Renzo, S. Stanczak, S. Wang, and artificial intelligence-as-a-service (DAIaaS) for smarter IoE and 6G
E. G. Larsson, ‘‘Artificial intelligence enabled wireless networking environments,’’ Sensors, vol. 20, no. 20, p. 5796, Oct. 2020, doi:
for 5G and beyond: Recent advances and future challenges,’’ IEEE 10.3390/s20205796.
Wireless Commun., vol. 27, no. 1, pp. 16–23, Feb. 2020. [87] L. U. Khan, I. Yaqoob, N. H. Tran, Z. Han, and C. S. Hong, ‘‘Net-
[65] D. Bega, M. Gramaglia, A. Banchs, V. Sciancalepore, and X. Costa-Perez, work slicing: Recent advances, taxonomy, requirements, and open
‘‘A machine learning approach to 5G infrastructure market optimization,’’ research challenges,’’ IEEE Access, vol. 8, pp. 36009–36028, 2020, doi:
IEEE Trans. Mobile Comput., vol. 19, no. 3, pp. 498–512, Mar. 2020, doi: 10.1109/ACCESS.2020.2975072.
10.1109/TMC.2019.2896950. [88] (Jun. 2016). Perspectives on Vertical Industries and Implications
[66] M. F. Zhani and H. ElBakoury, ‘‘FlexNGIA: A flexible Internet archi- for 5G. Accessed: Aug. 25, 2021. [Online]. Available:
tecture for the next-generation tactile internet,’’ J. Netw. Syst. Manag., https://www.ngmn.org/fileadmin/user_upload/160610_NGMN_Perspe
vol. 28, pp. 1–45, Oct. 2020. ctives_on_Vertical_In dustries_and_Implications_for_5G_v1_0.pdf
[67] X. You, C. X. Wang, J. Huang, X. Gao, Z. Zhang, M. Wang, Y. Huang, [89] C. Li, C.-P. Li, K. Hosseini, S. B. Lee, J. Jiang, W. Chen, G. Horn,
C. Zhang, Y. Jiang, J. Wang, and M. Zhu, ‘‘Towards 6G wireless com- T. Ji, J. E. Smee, and J. Li, ‘‘5G-based systems design for tactile
munication networks: Vision, enabling technologies, and new paradigm Internet,’’ Proc. IEEE, vol. 107, no. 2, pp. 307–324, Feb. 2019, doi:
shifts,’’ Sci. China Inf. Sci., vol. 64, no. 1, pp. 1–74, Nov. 2021. 10.1109/JPROC.2018.2864984.
[68] C. D. Alwis, A. Kalla, Q.-V. Pham, P. Kumar, K. Dev, W.-J. Hwang, [90] S. K. Sharma, I. Woungang, A. Anpalagan, and S. Chatzinotas, ‘‘Toward
and M. Liyanage, ‘‘Survey on 6G frontiers: Trends, applications, require- tactile Internet in beyond 5G era: Recent advances, current issues, and
ments, technologies and future research,’’ IEEE Open J. Commun. Soc., future directions,’’ IEEE Access, vol. 8, pp. 56948–56991, 2020, doi:
vol. 2, pp. 836–886, 2021. 10.1109/ACCESS.2020.2980369.
[69] (Jun. 2020). White Paper on 6G Networking. 6G Res. [91] J. Arshad, M. A. Azad, K. Salah, R. Iqbal, M. I. Tariq, and T. Umer,
Visions. Accessed: Oct. 22, 2021. [Online]. Available: ‘‘Performance analysis of content discovery for ad-hoc tactile networks,’’
http://jultika.oulu.fi/files/isbn9789526226842.pdf Future Gener. Comput. Syst., vol. 94, pp. 726–739, May 2019.
[70] M. Giordani, M. Polese, M. Mezzavilla, S. Rangan, and M. Zorzi, [92] I. Budhiraja, S. Tyagi, S. Tanwar, N. Kumar, and J. J. P. C. Rodrigues,
‘‘Toward 6G networks: Use cases and technologies,’’ IEEE Commun. ‘‘Tactile Internet for smart communities in 5G: An insight for
Mag., vol. 58, no. 3, pp. 55–61, Dec. 2020. NOMA-based solutions,’’ IEEE Trans. Ind. Informat., vol. 15, no. 5,
[71] A. Al-Dulaimi, X. Wang, and I. Chih-Lin, 5G Networks: Fundamen- pp. 3104–3112, May 2019.
tal Requirements, Enabling Technologies, and Operations Management. [93] K. Antonakoglou, X. Xu, E. Steinbach, T. Mahmoodi, and M. Dohler,
Hoboken, NJ, USA: Wiley, 2018. ‘‘Toward haptic communications over the 5G tactile Internet,’’ IEEE
[72] L. U. Khan, I. Yaqoob, M. Imran, Z. Han, and C. S. Hong, ‘‘6G wireless Commun. Surveys Tuts., vol. 20, no. 4, pp. 3034–3059, 4th Quart., 2018,
systems: A vision, architectural elements, and future directions,’’ IEEE doi: 10.1109/COMST.2018.2851452.
Access, vol. 8, pp. 147029–147044, 2020. [94] The Tactile Internet, ITU-T Technol. Watch Rep., Aug. 2014. Accessed:
[73] W. Saad, M. Bennis, and M. Chen, ‘‘A vision of 6G wireless systems: Sep. 23, 2020. [Online]. Available: https://www.itu.int/dms_pub/
Applications, trends, technologies, and open research problems,’’ IEEE itu-t/opb/gen/T-GEN-TWATCH-2014-1-PDF-E.pdf
Netw., vol. 34, no. 3, pp. 134–142, Oct. 2019. [95] S. Bi, R. Zhang, Z. Ding, and S. Cui, ‘‘Wireless communications in the
[74] (Feb. 2021). NTT DOCOMO. White Paper 5G Evolution era of big data,’’ IEEE Commun. Mag., vol. 53, no. 10, pp. 190–199,
and 6G. White Paper. Accessed: Jun. 12, 2021. [Online]. Oct. 2015.
Available: https://www.nttdocomo.co.jp/english/binary/pdf/corporate [96] M. Chiang and T. Zhang, ‘‘Fog and IoT: An overview of research oppor-
/technology/whitepaper_6g/DOCOMO_6G_White_PaperEN_v3.0.pdf tunities,’’ IEEE Internet Things J., vol. 3, no. 6, pp. 854–864, Dec. 2016.
[75] Z. Zhang, Y. Xiao, Z. Ma, M. Xiao, Z. Ding, X. Lei, [97] S. Han, I. Chih-Lin, G. Li, S. Wang, and Q. Sun, ‘‘Big data enabled mobile
G. K. Karagiannidis, and P. Fan, ‘‘6G wireless networks: Vision, network design for 5G and beyond,’’ IEEE Commun. Mag., vol. 55, no. 9,
requirements, architecture, and key technologies,’’ IEEE Veh. Technol. pp. 150–157, Jul. 2017.
Mag., vol. 14, no. 3, pp. 28–41, Sep. 2019. [98] Y. Sun, H. Song, A. J. Jara, and R. Bie, ‘‘Internet of Things and big data
[76] Y. Al-Eryani and E. Hossain, ‘‘The D-OMA method for massive multiple analytics for smart and connected communities,’’ IEEE Access, vol. 4,
access in 6G: Performance, security, and challenges,’’ IEEE Veh. Technol. pp. 766–773, 2016.
Mag., vol. 14, no. 3, pp. 92–99, Sep. 2019. [99] X. Wang and Y. He, ‘‘Learning from uncertainty for big data: Future
[77] B. Zong, C. Fan, X. Wang, X. Duan, B. Wang, and J. Wang, ‘‘6G technolo- analytical challenges and strategies,’’ IEEE Syst., Man, Cybern. Mag.,
gies: Key drivers, core requirements, system architectures, and enabling vol. 2, no. 2, pp. 26–31, Apr. 2016.
technologies,’’ IEEE Veh. Technol. Mag., vol. 14, no. 3, pp. 18–27, [100] G. Villarrubia, J. F. De Paz, P. Chamoso, and F. D. la Prieta, ‘‘Artifi-
Sep. 2019. cial neural networks used in optimization problems,’’ Neurocomputing,
[78] G. Gui, M. Liu, F. Tang, N. Kato, and F. Adachi, ‘‘6G: Opening new vol. 272, pp. 10–16, Jan. 2018.
horizons for integration of comfort, security, and intelligence,’’ IEEE [101] M. Chen, U. Challita, W. Saad, C. Yin, and M. Debbah, ‘‘Artificial
Wireless Commun., vol. 27, no. 5, pp. 126–132, Oct. 2020. neural networks-based machine learning for wireless networks: A tuto-
[79] S. Dang, O. Amin, B. Shihada, and M.-S. Alouini, ‘‘What should 6G be?’’ rial,’’ IEEE Commun. Surveys Tuts., vol. 21, no. 4, pp. 3039–3071,
Nature Electron., vol. 3, no. 1, pp. 20–29, Jan. 2020. 4th Quart., 2019, doi: 10.1109/COMST.2019.2926625.

30686 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

[102] G. Alnwaimi, S. Vahid, and K. Moessner, ‘‘Dynamic heterogeneous learn- [125] (Jun. 2020). White Paper on Machine Learning in 6G
ing games for opportunistic access in LTE-based macro/femtocell deploy- Wireless Communication Networks,6G Research Visions.
ments,’’ IEEE Trans. Wireless Commun., vol. 14, no. 4, pp. 2294–2308, White Paper. Accessed: Mar. 25, 2021. [Online]. Available:
Apr. 2015. https://www.6gchannel.com/items/6g-white-paper-machine-learning/
[103] X. X. Wang Li and V. C. M. Leung, ‘‘Artificial intelligence-based tech- [126] M. R. Abdmeziem, D. Tandjaoui, and I. Romdhani, ‘‘Architecting the
niques for emerging heterogeneous network: State of the arts, opportuni- Internet of Things: State of the Art,’’ in Robots and Sensor Clouds,
ties, and challenges,’’ IEEE Access, vol. 3, pp. 1379–1391, 2015. A. Koubaa and E. Shakshuki, Eds. Cham, Switzerland: Springer, 2016,
[104] R. Li, Z. Zhao, X. Zhou, G. Ding, Y. Chen, Z. Wang, and pp. 55–75, doi: 10.1007/978-3-319-22168-7_3.
H. Zhang, ‘‘Intelligent 5G: When cellular networks meet artificial [127] A. A. O. Bahashwan and S. Manickam, ‘‘A brief review of messaging
intelligence,’’ IEEE Wireless Commun., vol. 24, no. 5, pp. 175–183, protocol standards for Internet of Things (IoT),’’ J. Cyber Secur. Mobil.,
Oct. 2017. vol. 8, no. 1, pp. 1–14, 2019.
[105] L. Busoniu, R. Babuska, and S. B. De, ‘‘A comprehensive survey of [128] H. V. Nguyen and L. L. Iacono, ‘‘RESTful IoT authentication protocols,’’
multiagent reinforcement learning,’’ IEEE Trans. Syst., Man, Cybern. C, in Mobile Security and Privacy. Amsterdam, The Netherlands: Elsevier,
Appl. Rev., vol. 38, no. 2, pp. 156–172, Feb. 2008. 2017, pp. 217–234.
[106] K. Arulkumaran, M. P. Deisenroth, M. Brundage, and A. A. Bharath, [129] N. Naik, ‘‘Choice of effective messaging protocols for IoT systems:
‘‘Deep reinforcement learning: A brief survey,’’ IEEE Signal Process. MQTT, CoAP, AMQP and HTTP,’’ in Proc. IEEE Int. Syst. Eng. Symp.
Mag., vol. 34, no. 6, pp. 26–38, Nov. 2017. (ISSE), Oct. 2017, pp. 1–7.
[107] C. Jiang, H. Zhang, Y. Ren, Z. Han, K.-C. Chen, and L. Hanzo, ‘‘Machine [130] A. A. Ahmed and W. Ali, ‘‘A lightweight reliability mechanism proposed
learning paradigms for next-generation wireless networks,’’ IEEE Wire- for datagram congestion control protocol over wireless multimedia sensor
less Commun., vol. 24, no. 2, pp. 98–105, Apr. 2017. networks,’’ Trans. Emerg. Telecommun. Technol., vol. 29, no. 3, p. e3296,
[108] I. Portugal, P. Alencar, and D. Cowan, ‘‘The use of machine learning Mar. 2018.
algorithms in recommender systems: A systematic review,’’ Expert Syst. [131] E. Rescorla and T. Dierks, ‘‘The transport layer security (TLS) pro-
Appl., vol. 97, pp. 205–227, May 2015. tocol version 1.3,’’ Internet Eng. Task Force (IETF), Aug. 2018, doi:
[109] T. E. Bogale, X. Wang, and L. B. Le, ‘‘Machine intelligence tech- 10.17487/RFC8446.
niques for next-generation context-aware wireless networks,’’ ITU Spe- [132] C.-S. Park and W.-S. Park, ‘‘A group-oriented DTLS handshake for
cial Issue Impact Artif. Intell. AI Commun. Netw. Services, 2018, doi: secure IoT applications,’’ IEEE Trans. Autom. Sci. Eng., vol. 15, no. 4,
10.48550/arXiv.1801.04223. pp. 1920–1929, Oct. 2018.
[110] S. M. Aldossari and K.-C. Chen, ‘‘Machine learning for wireless com- [133] M. Taillon, T. Saad, R. Gandhi, Z. Ali, and M. Bhatia, ‘‘Updates to
munication channel modeling: An overview,’’ Wireless Pers. Commun., the resource reservation protocol for fast reroute of traffic engineering
vol. 106, no. 1, pp. 41–70, May 2019. GMPLS label switched paths (LSPs),’’ Internet Req. Comments RFC, Ed.,
[111] K. Bonawitz, H. Eichner, W. Grieskamp, D. Huba, A. Ingerman, RFC, vol. 8271, 2017.
V. Ivanov, C. Kiddon, J. Konečný, S. Mazzocchi, H. B. McMahan, T. Van [134] M. B. Yassien, S. A. Aljawarneh, M. Eyadat, and E. Eaydat, ‘‘Routing
Overveldt, D. Petrou, D. Ramage, and J. Roselander, ‘‘Towards federated protocol for low power and lossy network–load balancing time-based,’’
learning at scale: System design,’’ 2019, arXiv:1902.01046. Int. J. Mach. Learn. Cybern., vol. 12, no. 11, pp. 3101–3114, Nov. 2021,
[112] R. Dong, C. She, W. Hardjawana, Y. Li, and B. Vucetic, ‘‘Deep learn- doi: 10.1007/s13042-020-01261-w.
ing for hybrid 5G services in mobile edge computing systems: Learn [135] S. A. Hashemian and V. V. Tabataba, ‘‘A multigate scheme to improve
from a digital twin,’’ IEEE Trans. Wireless Commun., vol. 18, no. 10, CORPL under traffic load in cognitive radio based smart grids with mesh
pp. 4692–4707, Oct. 2019, doi: 10.1109/TWC.2019.2927312. topology,’’ vol. 16, no. 4, pp. 229–238, 2019, Art. no. 13980120178556.
[113] D. Gunduz, P. de Kerret, N. Sidiroupoulos, D. Gesbert, C. Murthy, and [Online]. Available: https://rimag.ricest.ac.ir/en/Article/28412
M. van der Schaar, ‘‘Machine learning in the air,’’ IEEE J. Sel. Areas [136] A. Bello, ‘‘Implementation of a channel-aware routing protocol in the
Commun., vol. 37, no. 10, pp. 2184–2199, Oct. 2019. network simulator for underwater acoustic communication networking,’’
[114] J. Huang, C. X. Wang, L. Bai, J. Sun, and Y. Yang, ‘‘A big data enabled Michigan Technol. Univ., Houghton, MI, USA, Tech. Rep., 2020, doi:
channel model for 5G wireless communication systems,’’ IEEE Trans. 10.37099/mtu.dc.etdr/1119.
Big Data, vol. 6, no. 2, pp. 211–222, Jun. 2020. [137] O. Gnawali, R. Fonseca, K. Jamieson, A. M. Kazandjieva, D. Moss, and
[115] X. Ma, J. Zhang, Y. Zhang, and Z. Ma, ‘‘Data scheme-based wire- P. Levis, ‘‘CTP: An efficient, robust, and reliable collection tree protocol
less channel modeling method: Motivation, principle and performance,’’ for wireless sensor networks,’’ ACM Trans. Sensor Netw., vol. 10, no. 1,
J. Commun. Inf. Netw., vol. 2, no. 3, pp. 41–51, Sep. 2017. pp. 1–49, Dec. 2013.
[116] H. Li, Y. Li, S. Zhou, and J. Wang, ‘‘Wireless channel feature extraction [138] T. Clausen, J. Yi, and U. Herberg, ‘‘Lightweight on-demand ad hoc
via GMM and CNN in the tomographic channel model,’’ J. Commun. Inf. distance-vector routing–next generation (LOADng): Protocol, extension,
Netw., vol. 2, no. 1, pp. 41–51, Mar. 2017. and applicability,’’ Comput. Netw., vol. 126, pp. 125–140, Oct. 2017.
[117] A. Charrada and A. Samet, ‘‘Joint interpolation for LTE downlink chan- [139] P. Li, L. Guo, and F. Wang, ‘‘A multipath routing protocol with load
nel estimation in very high-mobility environments with support vector balancing and energy constraining based on AOMDV in ad hoc network,’’
machine regression,’’ IET Commun., vol. 10, no. 17, pp. 2435–2444, Mobile Netw. Appl., vol. 26, pp. 1–10, Jun. 2019.
Nov. 2016. [140] V. Coskun, B. Ozdenizci, and K. Ok, ‘‘A survey on near field commu-
[118] H. He, C.-K. Wen, S. Jin, and G. Y. Li, ‘‘Deep learning-based channel nication (NFC) technology,’’ Wireless Pers. Commun., vol. 71, no. 3,
estimation for beamspace mmWave massive MIMO systems,’’ IEEE pp. 2259–2294, 2013.
Wireless Commun. Lett., vol. 7, no. 5, pp. 852–855, Oct. 2018. [141] Y. Choi, Y.-G. Hong, J.-S. Youn, D. Kim, and J. Choi, Transmission of
[119] Y. Li, Y. Zhang, X. Huang, H. Zhu, and J. Ma, ‘‘Large-scale remote IPv6 Packets Over Near Field Communication, document Internet-Draft
sensing image retrieval by deep hashing neural networks,’’ IEEE Trans. draft-ietf-6lo-nfc-12, Internet Engineering Task Force, 2019. [Online].
Geosci. Remote Sens., vol. 56, no. 2, pp. 950–965, Feb. 2018. Available: https://datatracker.ietf.org/doc/draft-ietf-6lo-nfc/12/
[120] R. El Hattachi and J. Erfanian. (Feb. 2015). NGMN 5G White [142] C. L. Devasena, ‘‘IPv6 low power wireless personal area network (6LoW-
Paper. NMGN Alliance. Accessed: Aug. 25, 2020. [Online]. Avail- PAN) for networking Internet of Things (IoT)—Analyzing its suitability
able: https://www.ngmn.org/uploads/media/NGMN_5G_White_Paper_ for IoT,’’ Indian J. Sci. Technol., vol. 9, no. 30, pp. 6–11, Aug. 2016.
V1_0.pdf [143] P. Sethi and S. R. Sarangi, ‘‘Internet of Things: Architectures, proto-
[121] S. Zhang, C. Xiang, and S. Xu, ‘‘6G: Connecting everything by cols, and applications,’’ J. Electr. Comput. Eng., vol. 2017, Jan. 2017,
1000 times price reduction,’’ IEEE Open J. Veh. Technol., vol. 1, Art. no. 9324035.
pp. 107–115, 2020. [144] Z. Yang and C. H. Chang, ‘‘6LoWPAN overview and implementations,’’
[122] S. Antipolis, ‘‘Feasibility study on new services and markets technology in Proc. EWSN, 2019, pp. 357–361.
enablers for critical communications, relese 14,’’ 3GPP, Paris, France, [145] S. M. Darroudi, C. Gomez, and J. Crowcroft, ‘‘Bluetooth low energy mesh
Tech. Rep. TR 22.862, Jun. 2016. networks: A standards perspective,’’ IEEE Commun. Mag., vol. 58, no. 4,
[123] S. Antipolis, ‘‘Study on enhancement of 3GPP support for 5G V2X ser- pp. 95–101, Apr. 2020.
vices, release 15,’’ 3GPP, Paris, France, Tech. Rep. TR 22.886, Dec. 2016. [146] R. Singh, J. Kaur, and I. S. Gill, ‘‘Evaluation of hybrid topologies under
[124] IMT Vision–Framework and Overall Objectives of the Future mobility of ZigBee devices using different trajectories,’’ Int. J. Comput.
Development of IMT for 2020 and Beyond, Technical Report: Appl., vol. 122, no. 20, pp. 8–13, Jul. 2015.
Recommendation ITU-R M.2083-0, Mobile, Radiodetermination, [147] M. B. Yassein, W. Mardini, and A. Khalil, ‘‘Smart Homes automa-
Amateur and Related Satellite Services, Sep. 2015. [Online]. Available: tion using Z-wave protocol,’’ in Proc. Int. Conf. Eng. MIS (ICEMIS),
https://www.itu.int/pub/R-REC/en Sep. 2016, pp. 1–6.

VOLUME 11, 2023 30687

 M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

[148] S. J. Danbatta and A. Varol, ‘‘Comparison of zigbee, Z-wave, Wi-Fi, and [175] (Jul. 6, 2018). Virtual Mirror Technology—It Will Change the Way
Bluetooth wireless technologies used in home automation,’’ in Proc. 7th You Shop. Quytech Blog. Accessed: Jul. 21, 2021. [Online]. Available:
Int. Symp. Digit. Forensics Secur. (ISDFS), Jun. 2019, pp. 1–5. https://www.quytech.com/blog/how-virtual-mirror-technology-will-
[149] K. Mekkia, E. Bajica, F. Chaxela, and F. Meyerb, ‘‘A comparative study of change-the-way-you-shop/
LPWAN technologies for large-scale IoT deployment,’’ ICT Exp., vol. 5, [176] Robotic Surgery. Accessed: Jul. 27, 2021. [Online]. Available:
no. 1, pp. 1–7, Mar. 2019. https://www.womencentre.com.au/robotic-surgery.html
[150] S. Popli, R. K. Jha, and S. Jain, ‘‘A survey on energy efficient [177] (2020). 6G Flagship. Univ. Oulu, Oulu, Finland. Accessed: Jul. 18, 2021.
narrowband Internet of Things (NBIoT): Architecture, application [Online]. Available: https://www.oulu.fi/6gflagship/
and challenges,’’ IEEE Access, vol. 7, pp. 16739–16776, 2019, doi: [178] Hexa-X: A Flagship for 6G Vision and Intelligent Fabric of Technology.
10.1109/ACCESS.2018.2881533. Accessed: Jul. 18, 2021. [Online]. Available: https://hexa-x.eu/
[151] L. Atzori, A. Iera, and G. Morabito, ‘‘SIoT: Giving a social structure to the [179] TeraFlow: Secured Autonomic Traffic Management for a Tera of SDN
Internet of Things,’’ IEEE Commun. Lett., vol. 15, no. 11, pp. 1193–1195, Flows. Accessed: Jul. 18, 2021. [Online]. Available: https://teraflow-
Nov. 2011. h2020.eu/
[152] M. A. A. da Cruz, J. J. P. C. Rodrigues, A. K. Sangaiah, J. Al-Muhtadi,
[180] DAEMON: Network Intelligence for Adaptive and Self-Learning
and V. Korotaev, ‘‘Performance evaluation of IoT middleware,’’ J. Netw.
Mobile Networks. Accessed: Jul. 18, 2021. [Online]. Available:
Comput. Appl., vol. 109, pp. 53–65, May 2018.
https://h2020daemon.eu/
[153] A. Farahzadi, P. Shams, J. Rezazadeh, and R. Farahbakhsh, ‘‘Middleware
[181] Bringing Reinforcement—Learning Into Radio Light Network for Mas-
technologies for cloud of things: A survey,’’ Digit. Commun. Netw., vol. 4,
sive Connections (6G BRAINS). Accessed: Jul. 19, 2021. [Online]. Avail-
no. 3, pp. 176–188, Aug. 2018.
able: https://6g-brains.eu/
[154] A. H. Ngu, M. Gutierrez, V. Metsis, S. Nepal, and Q. Z. Sheng, ‘‘IoT Mid-
dleware: A survey on issues and enabling technologies,’’ IEEE Internet [182] Ministry of Science and ICT, South Korea. Accessed: Jul. 19, 2021.
Things J., vol. 4, no. 1, pp. 1–20, Feb. 2017. [Online]. Available: http://english.msip.go.kr
[155] W. Kassab and K. A. Darabkh, ‘‘A–Z survey of Internet of Things: [183] South Korea to launch 6G pilot project in 2026. Accessed: Jul. 19, 2021.
Architectures, protocols, applications, recent advances, future directions [Online]. Available: https://www.rcrwireless.com/20200810/asia-
and recommendations,’’ J. Netw. Comput. Appl., vol. 163, Aug. 2020, pacific/south-korea-launch-6g-pilot-project-2026-report
Art. no. 102663. [184] Japan to Earmark 50 Billion for 6G Development. Accessed:
[156] Link Smart Docs. Accessed: Jul. 24, 2021. [Online]. Available: Jul. 19, 2021. [Online]. Available: https://www.japantimes.co.jp/news/
https://docs.linksmart.eu 2020/12/10/business/japan-earmark%C%A550-billion-6g-development/
[157] Global Sensor Network (GSN). Accessed: Jul. 24, 2021. [Online]. Avail- [185] 6th Generation Innovation Center (6GIC). Accessed: Jul. 19, 2021.
able: https://www.epfl.ch/labs/lsir/global-sensor-networks/ [Online]. Available: https://www.surrey.ac.U.K./news
[158] Understand Your Things The Open IoT Platform With MATLAB Analytics. [186] NTT, Intel and Sony Establish New Global Forum Dedicated to Realizing
Accessed: Jul. 25, 2021. [Online]. Available: https://thingspeak.com the Communications of the Future. Accessed: Jul. 19, 2021. [Online].
[159] Aura Middleware. Accessed: Jul. 25, 2021. [Online]. Available: Available: https://www.ntt.co.jp/news2019/1910e/191031a.html
https://github.com/AuraMiddleware/aura-middleware [187] Huawei Starts 6G Research at its Canada Lab. Accessed: Jul. 19, 2021.
[160] Service Discovery. Accessed: Jul. 24, 2021. [Online]. Available: [Online]. Available: https://www.techspot.com/news/81457- huawei-
https://aws.amazon.com starts-6g-research-canada-lab.html
[161] Azure IoT Hub. Accessed: Jul. 24, 2021. [Online]. Available: [188] SK Telecom to Collaborate. 6G Tech With Nokia, Ericsson
https://azure.microsoft.com and Samsung. Accessed: Jul. 19, 2021. [Online]. Available:
[162] IBM Watson IoT Platform. Accessed: Jul. 24, 2021. [Online]. Available: https://www.rcrwireless.com/20190618/5g/sk-telecom-collaborate-
https://www.ibm.com/internet-of-things/solutions/iot-platform/watson- 6g-nokia-ericsson-samsung
iot-platform. [189] Samsung’s 6G White Paper Lays Out the Company’s Vision for the Next
[163] Oracle IoT. Accessed: Jul. 24, 2021. [Online]. Available: Generation of Communications Technology. Accessed: Jul. 19, 2021.
https://www.oracle.com/internet-of-things/. [Online]. Available: https://news.samsung.com/global/samsungs-6g-
[164] Calvin Middleware. Accessed: Jul. 25, 2021. [Online]. Available: white-paper-lays-out-the-companys-vision-for-the-next-generation-of-
https://github.com/EriccsonResearch/calvin-base communications-technology
[165] Node-RED. Accessed: Jul. 25, 2021. [Online]. Available: [190] LG Electronics, KRISS and KAIST Team Up for 6G Develop-
https://Nodered.org ment. Accessed: Jul. 19, 2021. [Onli Available:. [Online]. Available:
[166] Ptolemy II. Accessed: Jul. 25, 2021. [Online]. Available: http://www. businesskorea.co.kr/news/articleView.html?idxno=50389
https://ptolemy.berkeley.edu/ptolemyII/index.htm [191] NTT Successfully Demonstrates 100 Gbps Wireless Trans-
[167] Akka: Part of Lightbend Platform. Accessed: Jul. 25, 2021. [Online]. mission Using a New Principle (OAM Multiplexing) as a
Available: https://www.lightbend.com/akka-part-of-lightbend-platform World’s First. Accessed: Jul. 19, 2021. [Online]. Available:
[168] Gryphon Trading Framework 0.12 Documenta- https://www.ntt.co.jp/news2018/1805e/180515a.html
tion. Accessed: Jul. 25, 2021. [Online]. Available:
[192] Tektronix, IEMN Demonstrate 100 Gb/s Wireless Transmissions Using
https://gryphon.readthedocs.io/en/latest/
New IEEE 802.15.3d Standard. Accessed: Jul. 19, 2021. [Online]. Avail-
[169] REBECA-Publish/Subscribe Middleware. Accessed: Jul. 25, 2021.
able: http://news.tektronix.com
[Online]. Available: https://www.ava.uni-rostock.de/en/ava-research/
[193] G. Biczok, M. Dramitinos, L. Toka, P. E. Heegaard, and H. Lonsetha-
projects/rebeca/
gen, ‘‘Manufactured by software: SDN-enabled multi-operator composite
[170] FIWARE Step by Step. Accessed: Jul. 25, 2021. [Online]. Available:
services with the 5G exchange,’’ IEEE Commun. Mag., vol. 55, no. 4,
https://fiware-tutorials.readthedocs.io/en/latest/
pp. 80–86, Apr. 2017.
[171] New ‘Emotional’ Robots Aim to Read Human Feelings. Accessed:
Jul. 21, 2021. [Online]. Available: https://techxplore.com/news/2018-01- [194] (2017). MATILDA, Deliverable D1.1. MATILDA Reference Architecture,
emotional-robots-aim-human.html Conceptualization and Use Cases. Accessed: Jun. 14, 2021. [Online].
[172] Emotional Robots: Machines That Recognize Human Feelings. Available: http://www.matilda-5g.eu/
Discovery. Accessed: Jul. 21, 2021. [Online]. Available: [195] (Sep. 2020). Smart Networks in Context of NGI. European
https://www.discovery.com/science/emotional-robots–machines-that- Technology Platform Networld 2020. Networld2020, SRIA-
recognize-human-feelings 2020. Accessed: Jun. 12, 2021. [Online]. Available: https://bscw.
[173] A. Pardes. The Second Coming of the Robot Pet. Wired. Accessed: 5g-ppp.eu/pub/bscw.cgi/d367342/Networld2020%20SRIA%202020%20
Jul. 21, 2021. [Online]. Available: https://www.wired.com/story/the- Final%20Version%202.2%20.pdf
second-coming-of-the-robot-pet/ [196] (2017). Machine Learning for Future Networks including 5G (ML5G).
[174] (Jul. 20, 2021). AR Shopping Is the Future Growth Point, and ITU Focus Groups. Accessed: Jun. 14, 2021. [Online]. Available:
WIMI Hologram Cloud Focuses on the 5G Consumer Market— https://www.itu.int/en/ITU-T/focusgroups/ml5g/pages/default.aspx
ZEE5 News. Accessed: Jul. 21, 2021. [Online]. Available: [197] (2017). AI and Applied Machine Learning. TIP. Accessed: Jun. 14, 2021.
https://www.zee5.com/zee5news/ar-shopping-is-the-future-growth- [Online]. Available: https://telecominfraproject.com/artificial-
point-and-wimi-hologram-cloud-focuses-on-the-5g-consumer-market intelligence-and-applied-machine-learning/

30688 VOLUME 11, 2023

M. Gupta et al.: Tactile Based Intelligence Touch Technology in IoT Configured WCN in B5G/6G-A Survey

[198] (2018). Network Data Analytics Function (NWDAF). 3GPP. RAKESH KUMAR JHA (Senior Member, IEEE)
Accessed: Jun. 14, 2021. [Online]. Available: http://www.tech- received the B.Tech. degree (Hons.) in electron-
invite.com/3m29/tinv-3gpp-29-520.html ics and communication engineering, the M.Tech.
[199] ITU-T Study Group. Framework for Evaluating Intelligence degree (Hons.) from NIT Jalandhar, India, in 2008,
Levels of Future Networks Including IMT-2020. ITU-T Y.3173. and the Ph.D. degree from NIT Surat, India,
Accessed: Jun. 14, 2021. [Online]. Available: http://handle. in 2013. He has completed his tenth exam from
itu.int/11.1002/1000/14133 the Government High School and class 12th from
[200] M. Katz, M. Matinmikko-Blue, and M. Latva-Aho, ‘‘6Genesis flagship the Science College. He is currently an Asso-
program: Building the bridges towards 6G-enabled wireless smart soci- ciate Professor with the Department of Electronics
ety and ecosystem,’’ in Proc. IEEE 10th Latin-Amer. Conf. Commun. and Communication Engineering, Indian Institute
(LATINCOM), Nov. 2018, pp. 1–9.
of Information Technology, Design and Manufacturing, Jabalpur (IIITDM
[201] (2019). Zero Touch Provisioning. CISCO. Accessed:
Jabalpur). He had ten years of rich academic, industrial, and research expe-
Jun. 14, 2021. [Online]. Available: https://www.cisco.com/c/en/us/td/
rience in various institutes/university, including NIT-Surat, Capgemini India
docs/switches/lan/catalyst3850/software/release/16-5/configuration_
guide/prog/b_165_prog_3850_cg/zero_touch_provisioning.pdf
Private Ltd., and SMVD University. He has published more than 101 journal
articles out of which more than 61 SCI journal articles, including IEEE
TRANSACTIONS, IEEE journal, Elsevier, Springer, Taylor & Francis, and Hin-
dawi. He has published more than 25 interference including ITU-T, IEEE
ANTS, INDICON, and APAN. His one concept related to the router of
wireless communication has been accepted by the International Telecommu-
nication Union (ITU), in 2010. He has filed eight patents out of which four
are published. He has organized many workshops and has also been invited
as a resource person in many workshops organized by prestigious research
institutes. He has guided five Ph.D. students, one submitted the thesis, three
defended pre-Ph.D. synopsis, and three students are presently pursuing.
He has guided more than 15 M.Tech. and more than 41 B.Tech. students
for various projects. He has more than 4001 citations in his credit in the
area of wireless communication. His research interests include wireless com-
munication, power optimizations, wireless security issues, and optical fiber
communication. He is a Senior Member of GISFI, SIAM, the International
Association of Engineers (IAENG), Advanced Computing and Communica-
tion Society (ACCS), and CSI. He has also served as an organizing member
and a TPC member for several national and international conferences. He has
received the Young Scientist Author Award by ITU, in December 2010.
He has received an APAN Fellowship, in 2011, Srilanka (2012), 2016, and
China (2017), Singapore (2018), New Zealand (2018), South Korea (2019),
and a Student Travel Grant from COMSNET 2012.

SANJEEV JAIN (Senior Member, IEEE) was born

in Vidisha, Madhya Pradesh, in 1967. He received
the post graduate degree in computer science and
engineering from the Indian Institute of Tech-
nology, Delhi, in 1992, and the Ph.D. degree in
computer science and engineering. He has over
24 years’ experience in teaching and research.
MANTISHA GUPTA (Student Member, IEEE) He has served as the Director for the Madhav Insti-
received the B.E. degree in electronics and com- tute of Technology and Science (MITS), Gwalior.
munication engineering from Jammu University, He has worked as a Vice Chancellor at Shri Mata
Jammu and Kashmir, India, in 2017, and the Vaishno Devi University, Katra. He is currently a Professor with the Com-
M.Tech. degree in electronics and communication puter Science Department, Central University of Jammu, Jammu and Kash-
engineering from Shri Mata Vaishno Devi Univer- mir. Besides teaching at post graduate level, he has the credit of making
sity, Katra, Jammu and Kashmir, in 2019. She significant contribution to research and development in the area of image
is pursuing the Ph.D. degree in IoT-configured processing and mobile ad-hoc networks. He has guided Ph.D. scholars
networks in B5G/6G wireless communication sys- and has undertaken a number of major research and development projects
tems with the School of Electronics and Communi- sponsored by the Government and Private Agencies. His work on digital
cation Engineering(SoECE), Shri Mata Vaishno Devi University (SMVDU). watermarking for image authentication is highly valued in the research field.
She is a Student Member of Institute of Electrical and Electronics Engineers He is a member of Association for Computing Machinery (ACM).
(IEEE).

VOLUME 11, 2023 30689