ACCEPTED FROM OPEN CALL

A Speculative Study on 6G
Faisal Tariq, Muhammad R. A. Khandaker, Kai-Kit Wong, Muhammad A. Imran, Mehdi Bennis, and Mérouane Debbah

Abstract revolutionary elements in 6G, and describe their

features beyond the capability of 5G.
While 5G is being tested worldwide and antici- The rest of this article is organized as follows.
pated to be rolled out gradually in 2019, research- In the following section, we present our 6G vision
ers around the world are beginning to turn their and speculate on the requirements. We then
attention to what 6G might be in 10+ years time, extend some 5G use cases to more ambitious sce-
and there are already initiatives in various coun- narios that are expected to appear in 6G. We will
tries focusing on the research of possible 6G tech- then discuss some key challenges, and present a
nologies. This article aims to extend the vision of few visionary technologies or research directions
5G to more ambitious scenarios in a more distant that may form key parts of 6G. Finally, we con-
future and speculates on the visionary technolo- clude this article.
gies that could provide the step changes needed
for enabling 6G. Vision
Building upon the 5G vision, 6G will continue to
Introduction empower our cities to be super smart and fully
The last decade has experienced a never-ending connected with a plethora of autonomous ser-
growth in global mobile data traffic. The Interna- vices. Human life will experience widespread
tional Telecommunication Union (ITU) predicted proliferation of humanoid robotic and intelligent
that the overall mobile data traffic will reach 5 devices capable of decision making with mini-
zettabytes (ZB) per month (see Fig. 1). The fifth mum human intervention. This will result in the
generation (5G) is the latest attempt that brings demise or significant reduction of traditional
mobile communications technology up to speed, mobile/handheld devices. Everything around us
but it is expected that 5G will reach its limits by will be very intelligent, giving rise to the concept
2030 and the chase continues. of Internet of Intelligent Things (IoIT) underpinned
Coupled with the rise of Internet-of-Things by the artificial intelligence (AI) capability embed-
(IoT), massive machine-type communications ded in almost all levels, from network orchestration
(MTC) etc., 5G is so much more than traditional and management to coding and signal processing,
cellular networks [1]. The European Telecommu- manipulation of smart structures, and data min-
nications Standards Institute (ETSI) published a ing for service-based context-aware communica-
document on 5G scenarios where the target peak tions. Few will disagree that AI will be an integral
rate is 10Gb/s in the uplink and 20Gb/s for the part of 6G due to the availability of big data and
downlink (3GPP TR 38.913). great advances in computation capability. Recent
The myriad of emerging massively data-inten- interest has also shifted toward virtualization and
sive use cases and applications such as multi- softwarization of physical resources and infrastruc-
way virtual meeting with holographic projection ture which will ensure intelligent organization of
[2] , virtual and augmented reality (VAR) [3], content and will make them available at the edge
brain communication interface [4], teleporta- of the network resulting in orders of magnitude
tion [5], remote surgery, and so on, to name improvements in terms of latency and energy effi-
just a few, indicates that the latency, reliabili- ciency.
ty and data rate requirements of these future In 6G, we expect to see AI in operation with
applications is clearly beyond the capacity of distributed training at the network edges including
existing 5G systems. Thus, the next step change small-cell base stations (SBS) and UEs, which is
in 6G will need to be called upon to fully sup- still an open problem. In contrast to the use of
port the services. Presumably, 6G will continue conventional AI algorithms on mobile network
to benefit from many 5G technologies, but new data in 5G, 6G will realize the notion of collective
technologies will certainly be needed to make AI, a step-up from the current AI techniques that
the next step change. will address the coexistence of multiple distribut-
The objective of this article is to provide ed mobile radio learning agents for individual as
an expert view on the most trending and novel well as global benefits.
research directions that have the potential to shape In order to accommodate the above mentioned
6G mobile technologies. Although the develop- changes, devices will experience a paradigm shift
ment of 6G is in an early stage and it is expected in terms of design, operation, size and chip archi-
that some ideas will only emerge in later years, this tecture. The next generation of neuromorphic
visionary article takes brave approach to specu- chips capable of emulating complex brain func-
late on the possible enabling technologies and the tion will emerge, which will exploit advances in

Digital Object Identifier: Faisal Tariq is with University of Glasgow; Muhammad Khandaker is with Heriot-Watt University; Kai-Kit Wong University College London;
10.1109/MWC.001.1900488 Muhammad Imran is with University of Glasgow; Mehdi Bennis is with University of Oulu; M. Debbah is with Huawei France R&D Center.

118 1536-1284/20/$25.00 © 2020 IEEE IEEE Wireless Communications • August 2020

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR. Downloaded on October 04,2023 at 11:51:07 UTC from IEEE Xplore. Restrictions apply.
 nano-photonics and nanoscale laser technologies
to deliver billion times faster computation than bio-
logical neurons. At the device level, another reality
is that (nano) radar technologies will be integrated
with mobile communication technologies in order
to provide all-round contextual information. 6G will
likely see physical-layer security finally thrive to pro-
vide a layer of defense, in addition to cryptograph-
ic techniques, for a variety of mobile devices with
different capabilities. Devices will also be much
smarter, empowered by AI trained by the behavior-
al data of the environment from radars.
On the other hand, after 30 years of effort, 6G
will finally see Mitola radio, a.k.a. cognitive radio,
reach its full potential. Despite the enormous inter-
est after Mitola et al. first introduced the concept
back in 1999 [6], little has been achieved so far
and there is still a huge gap between 5G new radio
(NR) and the cognitive radio Mitola envisioned. FIGURE 1. Global mobile data traffic forecast by ITU. Overall mobile data traffic
The 6G Mitola radio will see self-regulating societ- is estimated to grow at an annual rate of around 55 percent in 2020–2030 to
ies of mobile radios for fair as well as efficient coex- reach 607 exabytes (EB) in 2025 and 5,016 EB in 2030. (Source: Cisco).
istence and facilitate seamless mobile convergence
across all wireless systems and networks.
Characteristics 5G 6G
Apart from these, 6G will experience a para-
digm shift by the introduction of programmable Individual data rate 1 Gb/s 100 Gb/s
wireless environment as opposed to traditional
thinking of uncontrollable wireless channels, thanks DL data rate 20 Gb/s >1 Tb/s
to the smart metamaterial based structures that U-plane latency 0.5 ms <0.1 ms
provide an additional degree of freedom (DoF).
Smart reflecting surfaces will be installed in build- C-plane latency 10 ms <1 ms
ings to increase the antenna aperture to collect as
Mobility up to 500 km/h up to 1000 km/hr
much radio signals as possible that had not been
possible before. In the smaller scale, 6G will also DL spectral efficiency 30 b/s/Hz 100 b/s/Hz
see flexible antenna structure possible at UEs. Early
results on fluid antennas in [7] revealed a whole Operating frequency 3–300 GHz up to 1 THz
new possibility for designing wireless communi- TABLE 1. KPIs for 5G versus 6G.
cations systems. Also, metamaterial-based anten-
nas may also be implemented to make even more age. Volume spectral efficiency (in bps/Hz/m3),
compact wideband antennas. Such intelligent struc- as opposed to the often used area spectral effi-
tures seek to engineer the environment to cater to ciency (bps/Hz/m2), will be more suitable in 6G
different applications, to improve link quality, block to properly measure the system capacity in a three
interference, enhance privacy and security, avoid dimensional operating space. Ultra-reliable low-la-
adversarial attacks and many more [3, 5]. tency communication (URLLC), one key feature in
There is also a glimpse of successes in other 5G NR, will be a key driver in 6G again requiring
emerging areas which are not yet making much latency of less than 1ms in C plane and 0.1ms in U
of an impact in 5G but could become reality in plane. Applications such as remote surgery cannot
6G, including wireless power transfer (WPT) and be realized if the latency is not below these values
RF energy harvesting for massively distributed bat- and will pose significant risk to the patient. Energy
tery-less nanosensors, optical wireless communica- efficiency will be extremely important to prolong
tions or Li-Fi. Furthermore, there is a possibility that the battery life of UEs. The key performance indi-
6G will be more than wireless, and need to handle cators (KPIs) for 6G in comparison with 5G are
the coexistence of traditional mobile communica- shown in Table 1.
tions and interconnects inside PCs.
In terms of the requirements in 6G, the con- Use Cases
sensus seems to be that the data rate will race to Most of the use cases in 6G will evolve from
1Tb/s to enable autonomous management of var- those of 5G in terms of functionalities and quality
ious activities in the future smart cities. This target of experience. When the enablers of 6G become
data rate primarily suits the spectrum availability available, new use cases will emerge. While 5G
which is also anticipated to move toward terahartz use cases are being designed to overcome the
band capable of delivering Tb/s data rate. For communication distance barriers, 6G use cases
individual users, data rate is expected to increase will aim to overcome the limitations of physical
from 1Gb/s in 5G to at least 10Gb/s per user, and distance barriers. Here, we use examples scenar-
up to 100Gb/s in some use cases (such as holo- ios to describe what 6G may bring beyond the
graphic projection of super high vision 10K videos capability of 5G. A comparison of their use cases
with 60fps and 16 bit colour depth needs around is provided in Table 2.
160Gb/s data rate), in the emerging 6G systems.
For backhaul data rate, it would increase to 10Tb/s Immersive eXtended Reality (IXR)
to ensure support for high data rate applications A significant focus of 6G technologies will aim
such as VAR [2]. 6G is also expected to integrate toward overcoming the operational barriers cre-
with satellites for providing global mobile cover- ated by physical distance, IXR powered by haptic

IEEE Wireless Communications • August 2020 119

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR. Downloaded on October 04,2023 at 11:51:07 UTC from IEEE Xplore. Restrictions apply.
 Use case 5G 6G a holistic approach in an integrated fashion for a
truly smart city. Some example use cases of the
Centre of gravity user-centric service-centric future city-life are elaborated below.
Super Smart Home Environment: Most ele-
Ultra-sensitive applications not feasible feasible
ments around us will have a brain (thanks to the
True AI absent present expected rapid development in neuromorphic chip
design as well as progress in distributed and per-
Reliability not extreme extreme vasive AI with the capability of mimicking human
VAR partial massive scale cognition system) which will decide based on data
fusion from a myriad of (nano) sensors embedded
Time buffer not real-time real-time in all the elements. The IoIT infrastructure will be
predominantly operated by voice, gestures and
Capacity 1–D (b/s/Hz) or 2-D (b/s/Hz/m2) 3–D (b/s/Hz/m3)
other types of sensory communications. Therefore,
VLC no yes complete rethinking in system design approach will
be necessary.
Satellite integration no ye Ultra Smart Transport Infrastructure: The
WPT no yes automotive and transportation industries are expe-
riencing a generational change, partly due to the
Smart city components separate integrated connectivity and networking capability offered
by 5G and beyond systems. With the emergence
Autonomous V2X partially fully
of 6G, the overall transport system will be influ-
TABLE 2. Comparison between 5G and 6G use cases. enced by three things. First, the in-vehicle sensors
and actuators will be intelligent, powered by brain-
communication, advanced VAR, holographic pro- like AI capabilities leading to fully autonomous
jection, and so on, will be a dominant use case vehicles. Second, as anticipated in the previous
in the future. The hangout and meetings that are section, overcoming physical barriers will reduce
happening via video calls will be shifted toward a lots of travelling, and where people need to trav-
holographic presence giving a realistic feel; sur- el will be taken by driverless and fully automated
geons will carry out sophisticated surgery in dis- vehicles. Third, with the aid of advanced AI and
tant hospitals and so on; and engineers will be extreme data rate capability, overall transport infra-
able to solve mechanical problems from remote structure will be fully autonomous where safety
locations with the aid of extended reality environ- and security of the system will be ensured by mas-
ments. Concepts of learning and education will sively embedded intelligent sensors and actuators.
experience a revolution as people will learn from Note that a massive amount of data will need to
using VAR and therefore the learning rate will be be shared among vehicles to update live traffic and
much faster. Biological research and experiments real-time hazard information on the roads and to
which took decades will be done in years or even provide high-definition 3D maps. As vehicles gen-
months by simulating the scenarios created in IXR erally move at very high speeds, the network needs
coupled with accurately predicting possible out- extremely low round-trip time for communication.
comes using advanced AI capabilities as well as It is likely that vehicle-to-everything (V2X) technolo-
collaborating from various parts of the world in gies will not mature in 5G and its full potential will
IXR environments. only be realized in 6G with VLC, OAM and emerg-
Most of these use cases are data intensive and ing terahertz technologies.
will require 100s of Gb/s data rate. Clearly, 5G is Smart Ubiquitous Healthcare: The ageing pop-
not capable of supporting holographic projection ulation is putting a huge burden on the healthcare
from multiple locations in bidirectional communica- system due to the continuation of traditional phys-
tion that requires terabit data transfer. Also, multi- ical/manual management and devices with very
view video coding needs to advance significantly limited communication and networking capability
which will help to ensure extreme data compres- as well as limited agility. The rapid and anticipated
sion. Apart from these, the bidirectional and closed advances in bendable electronics and nano-bio
loop latency requirement of at least 10 times quick- sensors will revolutionize health monitoring and
er (0.1ms) than current capabilities is required for management. In 5G, some of the technologies
applications such as remote surgery. While 5G can are expected to be available in a localized fashion,
provide some basic mixed reality support in very whereas ubiquitous high quality coverage of 6G
controlled environments, it is simply not capable of will enable remote healthcare management aimed
supporting these applications in uncontrolled envi- at providing round the clock care regardless of the
ronments such as outdoor scenarios. location of the patients. However, extreme latency
and reliability need to be guaranteed to ensure
IoIT Integrated Ultra Smart City-Life emergency procedures and interventions are done
The jargon “smart city” describes the concept in a timely and uninterrupted manner. Also, prog-
where a city greatly improves the quality of life ress in soft and medical robotics coupled with IXR
(QoL) for the people in it by optimizing its oper- will provide remote surgery and intervention from
ations and functions using the available infra- a distance. Therefore, surgeons with certain exper-
structures to sense, detect, analyze and act by tise can assist and supervise robots to carry out
integrating the core components that run the city. procedures from anywhere in the world provided
Nevertheless, in 5G, a city can only be fragmen- that the critical latency and safety requirements are
tally smart, meaning that the major components met. Privacy is another concern which will primarily
such as utilities (i.e., electricity, water, waste, and be addressed by blockchain or variants of distrib-
so on), healthcare and monitoring, and transporta- uted ledger technology, but much more progress
tion networks are individually smart. 6G will take is required.

120 IEEE Wireless Communications • August 2020

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR. Downloaded on October 04,2023 at 11:51:07 UTC from IEEE Xplore. Restrictions apply.
 Paper
Attribute Yang et al., [8] Zhang et al., [2] Saad et al., [4] Giordani et al., [5] Strinati et al., [3] This article

Peak data
1 Tb/s 1 Tb/s 1 Tb/s 5 Tb/s (for VAR) 1 Tb/s 1 Tb/s
rate

Backhaul
û 10 Tb/s û û û 10 Tb/s
Data Rate
Volumetric
û û ü û 1–10 Gb/s/m3 ü
Capacity

Operating 300GHz + Visible Light 100GHz to 10THz

0.06 to 10 THz Up to 10 THZ Sub THz band + VLF Up to 1 THz + VLF
Frequency Frequency (VLF) + VLF

Mobility û ≥ 1000km/h û 1000 km/h û 1000 km/h

Cplane: ≤ 1ms
Latency ≤ 1ms 0.01–0.1ms ≤ 1ms ≤1ms 1ms
Uplane: ≤ 0.1ms

Holographic Teleportation, Remote Surgery, Haptic

High Precision
Fine Medicine, Projection, Tactile and XR (VR/AR), Brain eHealth, VAR, Comm, Massive IoT
Manufacturing,
Major Use Intelligent Disaster Haptic, Autonomous Communication, Industry 4.0, Enabled Smart City,
Smart Environment,
Cases Prediction, Surreal VR, Driving, Internet of Connected Robotics & Robotics, VAR, Autonomous
Holographic
3D Videos Nano-Bio Things, Autonomous Systems Autonomous Driving, Automation and
Communication
Space Travel Transportation Manufacturing

THz Novel Network THz Communication,

THz Communication, Tinycell, Ubiquitous
Ultra Massive MIMO Communication, Architecture, VLC, Pervasive AI, EH, VLC,
Holographic Network Energy
OAM-MDM, Super VLC, ML, 3D AI at Network Blockchain, Cell Free
Key Enabling Beamforming, Transfer and Harvesting,
Flexible Integrated Networks, Cell- Edge, Battery-less Network, Quantum
Technologies Quantum Comm, Transceiver with
Network, Multi-Domain less Architecture, Devices, THz Comm, Metasurface,
AI/ML, LIS, VLC Integrated Frequency
Index Modulation Energy Harvesting Communication, OAM, WPT, Context.
Blockchain Band, Smart Surface
NFV, Backhaul Distributed Security Comm.
TABLE 3. A comparative study among variuos 6G approaches.

Industry X.0 ment, localized network domain level and overall

network level. This will transform the 6G network
Massive incorporation of robots into automa- from a self organizing regime to a self sustaining
tion and warehouse transportation is vital for regime [4].
industry growth. The emerging concept of Deep learning, the most powerful AI techniques
Industry X.0 aims to enhance Industry 4.0 by at present, is however based on deep neural net-
exploiting social, mobile, analytics and cloud work (DNN), which relies on training in a cen-
(SMAC). The radio environment with a very tralized fashion. Yet 6G is moving toward a more
complex network comprised of hundreds or distributed architecture like fog-RAN handling mil-
thousands of robots, sensors and hardware lions and billions of end-to-end communications
elements is a challenge. 6G will fully support anywhere around the world. The distributed cloud
the Industry X.0 revolution by offering extreme structure necessitates training to be done at the
latency and reliability as well as massive IoT and network edges and handicaps the operation of
built-in AI capability. deep learning. Although the recently developed
federated learning partly addresses this problem by
Key Enabling Technologies allowing training to take place at distributed loca-
6G will look for another major leap from 5G, which tions, this is more a distributed implementation for
would only be possible if breakthroughs could be centralized learning, and communications between
achieved. Here we share some emerging ideas that the distributed clouds and a central network man-
could contribute significantly to enable 6G in the ager is required. Also, federated learning for opti-
next 10 years or so. Table 3 provides a comprehen- mization is much less powerful because updates in
sive study of various approaches proposed on key the user level are averaged before sending back to
enablers, use cases and associated KPIs. the central manager.
For 6G to succeed, AI needs to be integrat-
Pervasive AI ed with game theory to enable a truly distribut-
It is obvious from the preceding discussion that AI ed learning mechanism where many AI agents
and its variants will be at the core of 6G and will can teach and learn from each other by interac-
thus act as the most important enabling technolo- tions. Collective AI is a related concept that has
gy for 6G (Fig. 2). Due to the advances in AI tech- emerged recently to deal with the situation where
niques, especially deep learning and the availability multiple AI agents aim to achieve the same goal
of massive training data, in recent years there has based on local training with no direct communi-
been an overwhelming interest in using AI for the cation among the agents [9]. Also, over the next
design and optimization of wireless networks. decade AI and hardware will be co-developed to
AI is expected to play a key role in all areas. ensure better integration. Simultaneous progress
Those can be broadly classified in three different in formal methods will make the devices capable
levels, namely in the device or end user equip- of reprogramming themselves and thereby allow

IEEE Wireless Communications • August 2020 121

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR. Downloaded on October 04,2023 at 11:51:07 UTC from IEEE Xplore. Restrictions apply.
 To take full advantage of the fluid cells formed by
UAVs, the optimization for resource allocation,
trajectory, content caching and user association
will be achieved jointly. Also, in 6G, UAVs will
not only serve as flying base stations to provide
radio coverage but can also be content provid-
ers and computing servers. There will be a lot of
synergy with other emerging technologies. For
example, AI will take the network usage data to
learn and dynamically find the best paths for the
UAVs and optimize their other parameters. This
will inevitably lead to dynamic reconfigurations
of the network topology. In addition, UAVs will
benefit greatly from WPT and solar power tech-
nologies that can keep them moving all the time,
while UAVs will also help support service-based
network slicing.
Smart Metamaterials based
Programmable Radio Environments
Metamaterials-based antennas have been
researched for two decades, but are not yet mak-
ing an impact in mobile communications. 6G will
see metamaterials-based antennas become the
norm for UEs, permitting massive MIMO technol-
ogy to be adopted even at mobile phones as well.
The maturity of metamaterials-based antennas will
also make small-sized highly efficient wideband
FIGURE 2. The vision of 6G. antennas possible, which gives the hardware flexi-
bility that the 6G Mitola radio needs.
Another new form of antenna technology,
the devices to reconfigure their functionalities as which will come to light in 6G, is fluid antenna
required for the purpose. We estimate that future [7], made of conductive fluid, metal fluid or ion-
software-hardware co-development will trans- ized liquid that can be shaped to any desirable
form collective and pervasive AI into brain-like form to suit the propagation environment. The
functioning elements and will bring the true brain fluidic structure breaks the boundary between
power to 6G. the pre-defined antenna hardware and signal
processing, and makes it possible to optimize
Radar-Enabled Contextual Communications its position and shape for extraordinary diversity
Radar technologies enrich environmental aware- and multiplexing gains, while having the ability
ness for mobile UEs and IoT devices and enable to reduce the electromagnetic fields (EMF) expo-
context-aware communications to a level that sure by adapting to human gestures in the case
has not been possible before. In particular, of mobile phones, based on the environment and
recent development in mmWave radar allows the needs at any given time. Presumably, a single
it to be embedded in mobile devices, and since software-defined fluid antenna can provide the
the miniaturization will continue, more of them rich diversity that only massive MIMO antennas
can be embedded in a single device. This will could achieve, while enjoying the flexibility to
give 6G radios the environmental awareness alter its shape, size and position to fully utilize the
to empower AI at the device level. Combining surface of a UE.
the rich observations from radars with AI, UEs Software-defined materials (SDM) can actually
will be able to identify and localize potential be used to design large intelligent surfaces (LIS)
adversaries by observations from radar and to enable programmable wireless environments as
adapt their communications for enhanced pro- well as for enhancing the coverage area of ultra-
tection using physical layer security approaches. small cells [10]. By doing so, it is possible to control
6G Mitola radios will also store behavioral data the propagation environment by altering its elec-
of the environment (Fig. 2), and predict suspi- tromagnetic properties. For example, SDM can be
cious activities. In addition, methods like phys- laid on walls to provide insulation for unintended
ical layer authentication which relies on users’ radio signals as shown in Fig. 3 [10]. LISs on build-
behavioral data will be possible. The general ings or in indoor environments are predicted by
contextual data gathered by UEs will also assist many to make their mark in 6G.
the network to serve better by predicting UEs’ Programmable metasurfaces indeed can do
next moves. much more than simply alter propagation envi-
ronments. They could replace the design architec-
Cell-Free Networks ture of wireless transceivers entirely. Recent results
In 6G, the full potential of UAV wireless networks reveal that a metasurface can be programmed to
or drone cells will be realized, and their appli- vary the phase, amplitude, frequency and even
cation will be widely extended to mobilize the orbital angular momentum (OAM) of an electro-
network resources and integrating with cell free magnetic (EM) wave, effectively performing the
massive MIMO to achieve truly cell-free networks modulation of a radio signal without a mixer, and
where arbitrarily small latency may be obtained. RF chain [11]. This technology will be quite disrup-

122 IEEE Wireless Communications • August 2020

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR. Downloaded on October 04,2023 at 11:51:07 UTC from IEEE Xplore. Restrictions apply.
 tive and depending on its progress, we may see its
operation as an enhancement of 6G, if clear bene-
fits can be materialized.

Visible Light Communication (VLC)

VLC is a special form of optical wireless communi-
cations and uses white-LED to encode data in the
optical frequencies. Some works have suggested
that each link can achieve up to 0.5Gb/s data
rate [12], hence it is a good candidate to meet
the data rate requirement of 6G. We speculate
that VLC will be especially useful in vehicle-to-ve-
hicle (V2V) communications where the car’s head
lights and tail lights can be used as antennas for
‘control and coordination’ data communication.
VLC will also be useful in scenarios in which tradi- FIGURE 3. A programmable radio environment using SDM.
tional RF communication is less effective, such as
in-cabin Internet service in airplanes, underwater
communication, healthcare zones and so on. devices have already been realized which current-
ly work at a few degrees above the absolute zero
WPT and Energy Harvesting Based temperature. Much work is still needed to make
Battery-less Device Regime them operate in normal temperatures. It may be
WPT does not play a key role in 5G, but in 6G a long shot to see much of an impact of quantum
it will finally shine for a number of reasons. First, communication in 6G.
communication distance will be much shorter,
making WPT meaningful. Second, to enable IoIT, Blockchain Technology
there will be massive deployment of nano-bio Blockchain technology will likely play a major role
sensors which will have battery-less architecture in securing and authenticating future communica-
and will require to be excited by external ener- tion systems thanks to the inherent advantage of
gy sources. Energy scavenging from ambient RF distributed ledger technology. It offers a number
signals will become a viable power source for of benefits including decentralization, transpar-
low-power applications, as energy harvesting ency with adequate privacy and alteration-proof
technologies continue to advance. authentication. Both decentralization and trans-
parency ensures faster processing which is crucial
OAM Communication for 6G systems. As mentioned before, network
Exploiting polarization diversity and OAM mode resources will be virtualized and therefore block-
multiplexing, it has been demonstrated that very chain will play a key role in simultaneous resource
high capacity wireless communication systems allocation and authentication. Also, in the future,
can be built to work over a distance of a few security needs to be highly adaptive depending
meters. The performance is particularly prom- on the location, device capability and applica-
ising at a relatively short distance which can be tion. Healthcare data, for example, needs to be
useful for industrial automation. A millimeter-wave extremely secure and private while emergency
(mmWave) OAM system in [13] was reported responses may require quick access to those data
to have achieved more than 2.5Tb/s rate with via 6G systems, thus it needs to be ensured that
spectral efficiency of 95.7bps/Hz. This can be a the right person has the right level of access to
lucrative technology for Industry X.0 as well as for the data on the fly. Blockchain will also be very
end connectivity for holographic projection in IXR useful as an enabler in other cases such as cloud
environments which are envisaged as key 6G use storage, mobile edge access and so on. However,
cases. significant research needs to be done to avoid
delay caused by blockchain processing to ensure
Quantum Communication and Networks that constraints of latency-critical applications are
Quantum communication is another promising met.
technology which is likely to contribute consider-
ably toward two essential criteria of 6G, namely Challenges
extremely high data rate and security. The inher- As seen in Table 1, 6G looks for several orders of
ent security feature of quantum entanglement magnitude improvements over 5G in all aspects.
that cannot be cloned or accessed without tam- Here, we highlight a few hurdles, where major
pering it, makes it a rightful technology for 6G efforts will be needed.
and beyond systems. A number of works already
demonstrated initial practical implementation Providing Terabit Data Rate Ubiquitously
of quantum key distribution (QKD) and asso- Clearly, the biggest challenge is to provide a mas-
ciated protocols. Another attractive feature of sively high data rate in the various parts of the
quantum communication is that it is suitable for network. The recently formed ITU Focus Group
long distance communication. Nevertheless, the Technologies for Networks 2030 (FG NET-2030)
current repeater concept is not applicable for raised concern that fixed access network capa-
quantum communication as entanglement can- bilities are already lagging behind emerging 5G
not be cloned. Satellites, high altitude platforms systems. It is anticipated that access networks
and UAVs may be adopted as trusted nodes for for backhaul traffic will struggle to cope with
key regeneration and redistribution. In terms of unprecedented data growth and other quali-
designing quantum devices, single photon emitter ty requirements unless necessary steps to boost

IEEE Wireless Communications • August 2020 123

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR. Downloaded on October 04,2023 at 11:51:07 UTC from IEEE Xplore. Restrictions apply.
 In 6G, new types the research effort are initiated. Free space opti- mapping a high-level service abstraction to hard
of resources will cal communications and quantum communica- network components and mapping of network
emerge which did tions are the hopefuls for 6G backhaul to meet functions and infrastructure to vendor implemen-
not exist before. For the requirements. However, those technologies tations as well as end-to-end management and
have a number of challenges to meet before real- orchestration mechanisms are very difficult tasks,
example, program- istic deployment including appropriate circuitry and significant progress will need to be made if
mable metasurfaces design, mismatch between RF and optical domain RaaS is to be implemented in 6G.
and software-defined and so on.
materials will likely be For various applications, higher frequency High Capacity Low Power Consuming Device Regime
considered as part of bands are anticipated along with technologies like Every generation of mobile communications has
OAM and VLC. However, these technologies need been defined by the UE capability. This will be
the network resources to overcome significant challenges such as trans- more so for 6G as 6G will be AI-led and require
and therefore the new mission distance, obstacles, absorption and com- high computational power to run the AI algo-
mechanism needs to patibility/interfacing with other techniques. rithms. Therefore, UE will be more power hungry
be designed to incor- than ever. Energy efficiency at the device level will
porate them as a man- Modelling Sub-mmWave and THz Frequency Channels once again be a KPI in 6G.
An obvious way of supporting a massive increase Conventional transceiver components are most-
ageable resource which in the data rate requirement is to increase the ly based on semiconductor materials like Silicon
will make resource bandwidth and go up the frequencies. Initial dis- (Si) and Gallium Arsenide (GaAs). These devices
management much cussion indicates that frequencies in the range of are not energy efficient and produce excessive
more computationally THz and above will be considered for 6G. These heat. Since the maximum operating frequency of
complex and resource spectrums will be utilized in very short range CMOS transistors has not improved from around
communication or “whisper radio” [14]. Howev- 300GHz in the manufacturing process after 65 nm
intensive. er, the susceptibility of the THz band to block- in line with miniaturization, 300GHz-band ampli-
age, molecular absorption, sampling and circuits fiers with a CMOS integrated circuit are extreme-
for A/D & D/A conversion and communication ly difficult to realize. This means that they are not
range is among the major challenges that need to capable of supporting computationally intensive
be addressed in the coming years. Another issue and ultra fast applications for 6G. Thus, new devic-
is that at higher frequencies, the antenna size and es need to be designed based on new materials
associated circuitry become miniaturized and are which have the characteristics to support the need
difficult to fabricate on chip while ensuring noise for emerging systems. Thus, new devices need to
and inter-component interference suppression. be designed based on new materials which have
On the other hand, the exact propagation char- the characteristics to support the need for emerg-
acteristics in these bands is not well understood, ing systems. There are some promising approaches
although a few recent attempts to address these which aim to revolutionize the domain. One such
bottlenecks have reported encouraging results. concept is superconducting spintronics which will
THz technologies are still in their infancy. Major enable ultra low energy computation by combining
challenges with THz source/detector, adequate superconductivity with magnetism. However, signif-
power generation, THz modulator/demodulator icant research is needed to operate it at room tem-
as well as THz antennas are still yet to be tack- perature. There should be significant work focused
led. Furthermore, appropriate channel sounding on designing super heat sinking capacity which is
and measurement devices need to be designed crucial for 6G system
which is quite challenging due to the limitation
of properties of the materials used in the existing Security and Authentication
devices. With over 50 billion UEs and IoT devices connect-
ed everywhere with different levels of capability,
Resource as a Service (RaaS) 6G will need a holistic approach to secure the
The emergence of software defined networking sheer volume of mobile data across a diverse set
(SDN) and network function virtualization (NFV) of platforms and comply with the strict privacy
has sparked a shift to service oriented and inte- and security requirements.
grated resource distribution which is known as With increasing autonomy and integration of
RaaS. The concept of traditional physical resourc- more aspects of our life in the network, the infor-
es from the management point of view will grad- mation security concern reaches to a level that
ually diminish and most of the network elements never appeared before. With more than 50 bil-
will be fully virtualized where Fog networking, lion devices connected via 6G and IoT systems,
mobile edge computing and network slicing will working in a diverse set of platforms, one of the
play a crucial role. In 6G, new types of resources biggest challenges would be to develop integrated
will emerge which did not exist before. For exam- and holistic security approaches taking privacy and
ple, programmable metasurfaces and software-de- integrity into consideration which will be virtually
fined materials will likely be considered as part unbreakable.
of the network resources and therefore the new
mechanism needs to be designed to incorporate Conclusions
them as a manageable resource which will make As 5G is in its final testing phase getting ready for
resource management much more computation- its launch in 2019, discussion has already begun
ally complex and resource intensive. Thus, one to shape what 6G may be. There are already
trend of development for NFV in the 6G cycle high-profile initiatives around the world aiming to
will include network slicing with software-defined develop technologies for 6G, such as 6Genesis of
materials and programmable metasurfaces, from Finland and TOWS for 6G LiFi in the UK. While
machine learning-enabled cloud random access it is too early to define 6G and there are inevita-
network (C-RAN) to fog-RAN. Nonetheless, bly omissions in any such discussion, this article

124 IEEE Wireless Communications • August 2020

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR. Downloaded on October 04,2023 at 11:51:07 UTC from IEEE Xplore. Restrictions apply.
 has taken a brave approach to identify possible Additional Reading As 5G is in its final
enabling technologies for 6G and describe the [1] E. Bastug et al., “Toward Interconnected Virtual Reality: testing phase getting
features they bring beyond the capability of 5G. Opportunities, Challenges, and Enablers,” IEEE Commun.
Mag., vol. 55, June 2017, pp. 110–17. ready for its launch in
Our 6G vision presents a genuine realization of 2019, discussion has
Mitola radio, which has exceptional awareness
of the environment (by radar technologies) to Biographies already begun to shape
Faisal Tariq [M’13, SM’19] received his Ph.D. degree from the
make decisions using superb intelligence (by col- Open University, UK. He is currently a senior lecturer in the what 6G may be. There
lective AI), with a rich action space to adapt itself James Watt School of Engineering at the University of Glasgow. are already high-profile
in many forms (by intelligent structures, Li-Fi, WPT He has also worked at Queen Mary University of London. His
main research interests include resource management, secure initiatives around the
and energy harvesting, and so on). and programmable wireless communication systems. He is a world aiming to devel-
Additionally, OAM and quantum communica- recipient of the best paper award at the Wireless Personal Mul-
tions may appear in 6G if sufficient advances are timedia Conference (WPMC) in 2013. He is currently serving as op technologies for 6G,
made. This article has also discussed the limitations an editor for Elsevier Journal on Network and Computer Applica- such as 6Genesis of
tions and an associate editor for IEEE Wireless Communications
of 5G that form the basis of our 6G vision. We Letters and IEEE Access. Finland and TOWS for
have attempted to present the 6G use cases but 6G LiFi in the UK.
they are largely natural extensions of the 5G use Muhammad R. A. Khandaker [S’10, M’13, SM’18] received his
cases and scenarios. We anticipate that as 5G is Ph.D. degree from Curtin University, Australia. He is currently an
assistant professor in the School of Engineering and Physical Sci-
being put into service, new use cases will emerge ences at Heriot-Watt University. Before joining Heriot-Watt, he
and the industry will find more ambitious and chal- worked as a postdoctoral research fellow at University College
lenging scenarios. Lastly, we like to add that 6G London, UK, (July 2013 to June 2018). He is an associate editor
will see a shift from the electronic era of 5G to the for IEEE Wireless Communications Letters, IEEE Communications
Letters, IEEE Access and Eurasip Journal on Wireless Communica-
optical and photonics era, but this deserves a sep- tions and Networking.
arate discussion.
Kai-Kit Wong [M’01, SM’08, F’16] received the B.Eng., M.Phil.,
References and Ph.D. degrees, all in electrical and electronic engineering,
[1] J. G. Andrews et al., “What will 5G Be?” IEEE JSAC, vol. 32, from the Hong Kong University of Science and Technology,
Jun. 2014, pp. 1065–82. Hong Kong, in 1996, 1998, and 2001, respectively. He is Chair
[2] Z. Zhang et al., ”6G Wireless Networks: Vision, Require- in Wireless Communications in the Department of Electronic
ments, Architecture, and Key Technologies,” IEEE Vehicular and Electrical Engineering, University College London, UK. His
Technology Mag., vol. 14, no. 3, Sept. 2019, pp. 28–41. current research centers around 5G and beyond mobile com-
[3] E. Calvanese Strinati et al., ”6G: The Next Frontier: From munications. He is a Fellow of IEEE and IET and is also on the
Holographic Messaging to Artificial Intelligence Using Sub- editorial board of several international journals. He is the Editor-
terahertz and Visible Light Communication,” IEEE Vehicular in-Chief for IEEE Wireless Communications Letters.
Technology Mag., vol. 14, no. 3, Sept. 2019, pp. 42–50.
[4] W. Saad, M. Bennis, and M. Chen, ”A Vision of 6G Wire- M uhammad I mran [M’03, SM’12, M’03, SM’12] is a Fellow
less Systems: Applications, Trends, Technologies, and Open of IET, Senior Fellow of HEA, is the Dean of the University of
Research Problems,” IEEE Network, vol. 34, no. 3, May/June Glasgow UESTC and a professor of wireless communication
2020, pp. 134–142. systems with research interests in self organized networks, wire-
[5] M. Giordani et al., ”Towards 6G Networks: Use Cases and less networked control systems and the wireless sensor systems.
Technologies,” arXiv preprint arXiv:1903.12216. 2019, He heads the Communications, Sensing and Imaging research
Mar. 28. group at the University of Glasgow and is the Director of the
[6] J. Mitola et al., “Cognitive Radio: Making Software Radios Glasgow-UESTC Centre for Educational Development and Inno-
More Personal,” IEEE Pers. Commun., vol. 6, Apr. 1999, pp. vation. He has extensive academic and industrial experience
1318–46. with several leading roles in multi-million pounds funded proj-
[7] C. Borda-Fortuny et al., “A Low-Cost Fluid Switch for Fre- ects. He has filed 15 patents, has authored/co-authored over
quency-Reconfigurable Vivaldi Antenna,” IEEE Antennas 400 journal, conference publications and book chapters and has
Wireless Prop. Lett., vol. 16, Nov. 2017, pp. 3151–54. edited seven books.
[8] P. Yang et al., ”6G Wireless Communications: Vision and
Potential Techniques,” IEEE Network, vol. 33, no. 4, July Mehdi Bennis [SM] is an associate professor at the Centre for
2019, pp. 70–75. Wireless Communications, University of Oulu, Finland, and an
[9] L. Zheng et al., “MAgent: A Many-Agent Reinforcement Academy of Finland Research Fellow. He has been the recipi-
Learning Platform for Artificial Collective Intelligence,” Proc. ent of several awards including the Fred W. Ellersick Prize and
The 32nd AAAI Conf. AI (AAAI-18), 2018. Best Tutorial Prize from the IEEE Communications Society and
[10] C. Liaskos et al., “A New Wireless Communication Para- EURASIP Best Paper Award for the Journal on Wireless Com-
digm Through Software-controlled Metasurfaces,” IEEE Com- munications and Networks. He is currently an editor for IEEE
mun. Mag., vol. 56, Sept. 2018, pp. 162–69. Transactions on Communications.
[11] W. Tang et al., “A Programmable Metasurface Based RF Chain-
Free 8PSK Wireless Transmitter,” IET Electronics Lett., 2019. M érouane D ebbah [S’01, M’04, SM’08, F’15] received the
[12] C. Chen et al., “Omnidirectional Transmitter and Receiver M.Sc. and Ph.D. degrees from the Ecole Normale Supérieure
Design for Wireless Infrared Uplink Transmission in LiFi,” Paris-Saclay, France. Since 2014, he has been Vice-President
Proc. IEEE Int’l. Conf. Commun. Workshops, Kansas City, of the Huawei France Research Center. He is jointly the direc-
USA, 2018. tor of the Mathematical and Algorithmic Sciences Lab as well
[13] J. Wang et al., “Terabit Free-Space Data Transmission as the director of the Lagrange Mathematical and Computing
Employing Orbital Angular Momentum Multiplexing,” Research Center. His research interests lie in fundamental math-
Nature Photonics, vol. 6, Jun. 2012, pp. 488–96. ematics, algorithms, statistics, information, and communica-
[14] Y. Xing and T. S. Rappaport, “Propagation Measurement tion sciences research. He was a recipient of the Mario Boella
System and Approach at 140 GHz–Moving to 6G and Award, the Qualcomm Innovation Prize Award and the 2019
above 100 GHz,” Proc. 2018 IEEE Global Commun. Conf. IEEE Radio Communications Committee Technical Recognition
(GLOBECOM), Abu Dhabi, United Arab Emirates, 2018, doi: Award. He is an Associate Editor-in-Chief of the journal Random
10.1109/GLOCOM.2018.8647921, pp. 1–6. Matrix: Theory and Applications.

IEEE Wireless Communications • August 2020 125

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR. Downloaded on October 04,2023 at 11:51:07 UTC from IEEE Xplore. Restrictions apply.