1553-877X (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2019.2897800, IEEE
Communications Surveys & Tutorials
1
Enabling Technologies for Ultra-Reliable and Low
Latency Communications: From PHY and MAC
Layer Perspectives
Gordon J. Sutton, Jie Zeng, Senior Member, IEEE, Ren Ping Liu, Senior Member, IEEE,
Wei Ni, Senior Member, IEEE, Diep N. Nguyen, Beeshanga A. Jayawickrama,
Xiaojing Huang, Senior Member, IEEE, Mehran Abolhasan, Senior Member, IEEE,
Zhang Zhang, Eryk Dutkiewicz, Senior Member, IEEE and Tiejun Lv, Senior Member, IEEE
Abstract—Future 5th generation (5G) networks are expected
to enable three key services - enhanced mobile broadband
(eMBB), massive machine type communications (mMTC) and
ultra-reliable and low latency communications (URLLC). As
per the 3rd generation partnership project (3GPP) URLLC
requirements, it is expected that the reliability of one transmission
of a 32 byte packet will be at least 99.999% and the latency will be
at most 1 ms. This unprecedented level of reliability and latency
will yield various new applications such as smart grids, industrial
automation and intelligent transport systems. In this survey we
present potential future URLLC applications, and summarize the
corresponding reliability and latency requirements. We provide a
comprehensive discussion on physical (PHY) and medium access
control (MAC) layer techniques that enable URLLC, addressing
both licensed and unlicensed bands. The paper evaluates the
relevant PHY and MAC techniques for their ability to improve
the reliability and reduce the latency. We identify that enabling
long-term evolution (LTE) to coexist in the unlicensed spectrum is
also a potential enabler of URLLC in the unlicensed band, and
provide numerical evaluations. Lastly, the paper discusses the
potential future research directions and challenges in achieving
the URLLC requirements.
Index Terms—URLLC, reliability, latency, LTE, unlicensed,
coexistence.
I. INTRODUCTION
5th generation (5G) networks are expected to enable three
key services, concentrating on each service separately so as
to achieve enhanced performance in each. The 5G enhanced
mobile broadband (eMBB) service aims to significantly in-
crease the user data rate; the 5G massive machine type com-
munications (mMTC) service aspires to realize the Internet of
things (IoT) concept by connecting billions of (often low data
G.J. Sutton, R.P. Liu, D.N. Nguyen, B.A. Jayawickrama, X. Huang, M.
Ablohasan, and E. Dutkiewicz are with the Global Big Data Technologies
Centre, University of Technology Sydney, Australia (e-mail: {gordon.sutton,
renping.liu, diep.nguyen, beeshanga.jayawickrama, mehran.abolhasan, xiao-
jing.huang, eryk.dutkiewicz}@uts.edu.au).
J. Zeng, the corresponding author, is with Beijing University of Posts and
Telecommunications, China, Tsinghua University, China, and University of
Technology Sydney, Australia (e-mail: zengjie@tsinghua.edu.cn).
W. Ni is with Data61, Commonwealth Scientific and Industrial Research,
Australia (e-mail: wei.ni@data61.csiro.au).
Z. Zhang is with Huawei Technologies Co. Ltd., China (e-mail:
zhangzhang4@huawei.com).
T. Lv is with the School of Information and Communication Engineer-
ing, Beijing University of Posts and Telecommunications, China (e-mail:
lvtiejun@bupt.edu.cn).
rate) smart devices; and the 5G ultra-reliable and low latency
communications (URLLC) capability is expected to support
unprecedented levels of high reliability and low latency com-
munications. In [1], the 3rd generation partnership project
(3GPP) outlines the general URLLC reliability requirement
for one transmission of a packet as 99.999% (block error rate
(BLER) of 10
−5
) for 32 bytes with a user plane latency of
1 ms. In [2], the authors propose metrics to evaluate reliability
in the time domain, such as mean time to failure, mean time
between failures and mean time to repair. For space-domain
reliability evaluation, [2] proposes metrics such as the mean
covered area and the mean uncovered area, modeled using
Poisson point processes (PPPs) and Voronoi tessellation.
Regardless of the metric, it is certain that the unprecedented
reliability and latency targets of 5G will give rise to various
new and exciting applications, which we discuss in the next
subsection.
This paper provides a comprehensive survey on the state
of art for URLLC from physical (PHY) and medium access
control (MAC) layer perspectives, covering both licensed and
unlicensed spectra below 6 GHz. Utilizing the unlicensed
spectrum as part of URLLC has not been given much attention
previously. To the best of our knowledge, our work is the first
which covers URLLC in such depth and provides a valuable
insight for researchers who are aiming to work in this area.
A. URLLC Potential Applications
There are a number of potential URLLC applications, which
might be operated in either licensed or unlicensed bands, or
both, as depicted in Fig. 1. The major use cases include:
• Smart grid [3]: an electrical grid that consists of several
operational and energy modules, such as smart meters and
devices, as well as renewable energy and energy efficient
resources.
• Professional audio [3]: an audio system which is set up
by professional live event supporting audio engineers,
adopting audio mixers or sound reinforcement systems to
perform sound recording, studio music production, sound
reinforcement system setup and mixing.
• Self-driving car [4]–[6]: a car which can detect the envi-
ronment and automatically drive without being operated
by a person.
1553-877X (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2019.2897800, IEEE
Communications Surveys & Tutorials
2
• Industrial automation [3]: the new unmanned industri-
alization beyond mechanization, which processes with
the help of control systems, e.g., robots, computers, and
information technologies.
• Process automation [3]: an automatic monitoring and de-
cision system for industrial components and procedures,
such as heating, mixing, and pumping.
• E-health [7]–[10]: a new healthcare approach with the
support of information and communication technology.
• Augmented reality [11]–[13], [15]: a technique to aug-
ment the vision of real-world environment by computer-
generated information, such as audio, video, and geo-
graphic information.
• Intelligent transport system (ITS) [3], [16]: a traffic
management system with information and communica-
tion technologies, supporting communication interfaces
between the elements of road transport, such as vehicles,
users and infrastructures, as well as interfaces between
different modes of transport.
• Vehicle-to-vehicle (V2V) [17]–[19]: a wireless ad-hoc
network which supports communications between vehi-
cles.
• Tactile Internet [9], [20]–[22]: an Internet network, which
ensures the tactile sensing with the support of short
transit, low latency, high reliability, high availability and
high security communications.
Requirements for these use cases are compared in Table I.
B. Recent Advances and Standardization
Challenges, solutions and applications of URLLC can be
found in recent literature. At the same time, industry specifica-
tions on URLLC are modified and released by standardization
organizations.
5G is being standardized in the form of two radio tech-
nology components: a novel radio interface denoted as new
radio (NR), and long-term evolution (LTE). Achievable latency
bounds are evaluated and the expected spectral efficiency is
demonstrated in [23]. It is shown that both NR and LTE
can fulfil the requirements of international telecommunication
union (ITU) 5G. In order to enable low-latency communi-
cations, new short slot structures enable faster uplink (UL)
and downlink (DL) transmission for URLLC, called mini-
slot for NR and short transmission time interval (TTI) for
the LTE radio interfaces. In addition, mechanisms to increase
the reliability of URLLC services, such as robust coding
and modulation, and various diversity schemes, are being
developed in accordance with the LTE and NR designs.
The effective bandwidth and effective capacity theories are
used in [24] as an analytical framework for calculating the
maximum achievable rate for given latency and reliability
constraints. The authors point out that the use of a shorter
subframe duration for a reduced hybrid automatic repeat
request (HARQ) transmission delay could reduce the latency.
A fundamental mechanism is proposed in [25] to revise
the methods for encoding control information and data. By
combining the header and data of short packets, the combined
packet can be efficiently coded, so that the data is delivered
faster and with less error. All users have to decode the
combined packet, so energy efficiency is traded for very high
reliability. [25] also catergorizes ultra-reliable communications
(URC) over two dimensions. The first dimension is the time
frame used to measure the reliability of the packet transmission
(long-term URC and short-term URC). The second dimension
is the type of impairment that can affect the communication
reliability in a given scenario. Five reliability impairments are
summarized, namely, decreased power of the useful signal,
uncontrollable interference, resource depletion, protocol relia-
bility mismatch, and equipment failure.
A number of technology components are identified by the
mobile and wireless communications enablers for the twenty-
twenty information society (METIS) project to address the
URLLC requirements, such as reliable service composition
framework and operational device-to-device (D2D) links, radio
resource management (RRM), MAC, and PHY layer chal-
lenges [26]. The trade-offs between bandwith, coding schemes,
diversity order, signal-to-noise ratio (SNR) and error rates,
when transmitting a 100 bit packet with end-to-end delay of
100 µs, are explored in [27]. The exploration demonstrates
that it is feasible to achieve low latency with high reliability by
using short transmission intervals without retransmission and
equipping base stations (BSs) with a sufficiently large number
of antennas to guarantee reliability via a spatial diversity gain.
There are several 3GPP technical reports related to URLLC
[1], [28]–[32]. The results of these studies are yet to be
standardized but are expected to be included in Release 16
NR technical specifications.
3GPP requirements of URLLC are described in detail in [1],
where user plane latency is defined as, “the time it takes to
successfully deliver an application layer packet/message from
the radio protocol layer 2/3 service data unit (SDU) ingress
point to the radio protocol layer 2/3 SDU egress point via the
radio interface in both the UL and DL directions, where neither
device nor base station reception is restricted by discontinuous
reception (DRX).” Reliability can be evaluated as the success
probability of transmitting a specified number of bytes within
a certain user-plane latency, given a certain channel quality
(e.g., a few meters, or coverage-edge). For URLLC, the limit
for user plane latency is 0.5 ms for UL and DL separately,
and a general requirement for reliability is 99.999% for the
transmission of a 32-byte packet with 1 ms user-plane latency.
Put more mathematically, reliability is the probability of
transmitting X bytes within a certain end-to-end delay, T ,
where the end points are the protocol layer 2/3 SDU ingress
and egress points. The end-to-end delay may be over one link
(e.g., a sidelink), or over two links (e.g., between two user
equipment (UE) via the BS. When the transmission is over a
single link, the reliability effectively equals 1 - BLER.
The bit error rate (BER) required to meet a given reliability
depends on how correlated the decoded bit errors are, which
depends on the error correction scheme. To correct for long
error bursts that can occur in deep fading or due to bursty
interferences, an interleaver which spreads out bursts of errors
over time is typically combined with AWGN channel codes
[33]. In addition to transmitting the data bits, low-density
parity-check (LDPC) codes and turbo codes also transmit
1553-877X (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2019.2897800, IEEE
Communications Surveys & Tutorials
3
TABLE I
USE CASE REQUIREMENTS
Use case Latency (ms) Reliability (%) Data Size (bytes) Communication Range (m)
Smart grid 3 ~20 [3] 99.999 [3] 80 ~1000 [3] 10 ~1000 [3]
Professional audio 2 [3] 99.99999 [3] 3 ~1000 [3] 100 [3]
Self-driving car 1 [4] 99 [5] 144 [6] 400 [6]
Industrial automation 0.25 ~10 [3] 99.9999999 [3] 10 ~300 [3] 50 ~100 [3]
Process automation 50 ~100 [3] 99.99 [3] 40 ~100 [3] 100 ~500 [3]
E-health 30 [7] 99.999 [7] 28 ~1400 [8] 300 ~500 [9]
Augmented reality 0.4 ~2 [12] 99.999 [13] 12k ~16k [14] 100 ~400 [15]
ITS 10 ~100 [3] 99.999 [3] 50 ~200 [16] 300 ~1000 [16]
V2V 5 [18] 99.999 [18] 1600 [18] 300 [19]
Tactile internet 1 [9] 99.99999 [21] 250 [20] 100000 [22]
Fig. 1. Potential URLLC use cases
parity bits that equal the parity over a spread of data bits,
thereby spreading out bursts of errors and allowing them to be
corrected. In the asymptotic case, the Shannon capacity can be
approached, with very long code blocks. However, in URLLC
scenarios, the block size is typically too short for the codes
to effectively ensure reliable communications. The obtainable
capacity when transmitting shorter blocks is an emerging field
of research, and the design of short codes for small block
size has been increasingly attracting attention. We survey the
current research of finite blocklength information theory in
Section II.A.3) and the performance of short codes in Section
II.B.2).
Over a single link (e.g., just UL), if the decoded bit errors
are uncorrelated, the required decoded BER, referred to as
the information BER (IBER), is related to the BLER and
reliability as: reliability = 1 - BLER = (1 - IBER)
8X
.
When transmitting X bytes in time t
i
and bandwidth B
i
,
the required coding rate is R
i
= 8X/B
i
t
i
(in bits per channel
use). Let IBER
i
(t
i
, X) and BLER
i
(t
i
, X) respectively be
the IBER and BLER on link i in a quasi-static fading chan-
nel. From finite blocklength information theory [34]–[36], in
typical URLLC transmissions with finite blocklength M, we
have [35, eq.3]
BLER
i
(t
i
, X) = E [
i
(γ
i
)]
≈ E
"
Q
s
M
V (γ
i
)
C(γ
i
) − R
i
!#
,
(1)
where γ
i
is a random variable and denotes the received signal-
to-interference-plus-noise ratio (SINR) on link i;
i
(γ
i
) is
the BLER under a given received SINR γ
i
on link i; E[.]
is the expectation function; C(γ) = log
2
(1 + γ); Q(w) =
R
∞
w
1
√
2π
e
−t
2
/2
dt, and V (γ) =
γ(γ+2)
(1+γ)
2
log
2
2
e. It is reasonable
to assume that the received SINR γ
i
is static during one
URLLC transmission, since the transmission time can be
shorter than the coherence time of the channel [34].
1553-877X (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2019.2897800, IEEE
Communications Surveys & Tutorials
4
With the decoded bit errors assumed uncorrelated, we have
IBER
i
(t
i
, X) = 1 − [1 − BLER
i
(t
i
, X)]
1/(8X)
. (2)
The reliability of transmitting X bytes, in time T , over L
sequential links, denoted R(X, T, L), can be written as
R(X, T, L) =
L
Y
i=1
[1 − BLER
i
(t
i
, X)] , (3)
where
P
L
i=1
t
i
= T .
Numerologies and frame structure related contents are pro-
vided in [31], and a study item is approved that contains
scenarios, requirements and technology components for the
NR access technology and the channel model for frequency
spectrum above 6 GHz. To aid URLLC, mini-slots of length
2-6 symbols are supported for subcarrier spacings of up to
60 kHz [32]. To meet strict 5G URLLC requirements, new
study items and work items on URLLC will be carried out
and reflected in Release 16 and beyond.
Early predictions for 5G were made in [37] from an IEEE
technology perspective, addressing each layer of the protocol
stack, but predominantly discussing higher-layer aspects. The
aurthor of [37] predicted that devices will need to be able to
operate on different wireless networks, and the 5G is expected
to have a flat network architecture, with much functionality
performed at the base stations, to achieve scalability.
Network and radio interface technologies that enable 5G
communications are discussed in [38] and [39]. In [39],
an emphasis is placed on utilizing new mm-Wave bands
(60 GHz), with directional beamforming, massive multi-input
multi-output (MIMO), and the associated spatial division mul-
tiple access (SDMA) MAC protocol. Research directions for
cellular URLLC are explored in [40], and they instead empha-
size non-orthogonal multiple access (NOMA) and coding for
latency reduction.
5G communications is surveyed in [41], including network
architecture and radio interface technologies. They discuss 5G
enabling techniques such as NOMA, sparse coding multiple
access (SCMA), massive MIMO, relaying and in-band full-
duplex, D2D, and mm-Wave, although there is little discussion
on latency or reliability. They also discuss adaptive functional-
ity, such as self-organising networks, cognitive radio and green
communications.
The IEEE time sensitive networking (TSN) standard covers
link-layer operation, and the deterministic networking (Det-
Net) standard covers the network layer. A survey of ultra-low
latency networks is provided in [42] that focuses on TSN and
DetNet. The exploration is broken into flow synchronization,
management, control and integrity, and covers ultra-low la-
tency techniques across the wireless access, fronthaul (ether-
net/optical), backhaul (optical) and core networks. The latency
of ethernet networks is modelled in [43] for a ring topology
in industrial settings, and the role of the ethernet in providing
low latency vehicle-to-infrastructure (V2I) communications is
explored in [44].
In [45], a survey with focus on mm-Wave communications
is conducted. The high frequencies of mm-Wave communi-
cations create new challenges due to high propagation loss,
sensitivity to blockages (e.g., 20-30 dB loss from a human),
and the need for directed transmissions. To combat blockages,
proposals include utilizing wall reflections, static reflectors,
two access points (APs), relays, and spatial diversity. Tech-
nological challenges include achieving MIMO and in-band
full-duplex at mm-Wave frequencies. The 60 GHz mm-Wave
band can only operate over short distances, which aids spectral
re-use, but also creates a need for coexistence with systems
operating at other frequencies that can transmit further.
Another of the three main services to be supported by 5G
is mMTC. The IoT falls into the 5G mMTC category and
has been surveyed in [46]. The IoT includes autonomous
communication of collected data and control messages to/from
smart devices that have sensors and possibly actuators. In
the IoT context, ultra-low latency is not an objective, and
reliability is instead associated with the success rate of packet
delivery without particular focus on latency, for which proba-
bilistic checking of data can be used to identify anomalies
and act as a safeguard. With the large number of devices
expected in IoT networks, scalability is very important. Cloud
and fog computing (i.e., cloudlets or edge computing) have
been proposed to achieve scalability. Millions of smart IoT
devices connect to thousands of Fog gateways, which connect
to hundreds of cloud data centers. A significant portion of the
data storage and computing services are performed through fog
computing, which, by being closer to the end user, reduces
latency. URLLC can also be assisted by upper-layer mech-
anisms, such as C-RAN, mobile edge computing [47]–[50],
network slicing, software-defined networking (SDN) [51]–[56]
and caching [57].
The current paper surveys URLLC from PHY and MAC
layer perspectives, covering both licensed and unlicensed
spectra below 6 GHz. Utilizing the unlicensed spectrum as
part of URLLC has not been given much attention previously.
C. Influence of Public Safety Networks
The need for public safety networks, with high reliability
and high priority, has driven mission critical communications.
Desirable features of public safety LTE (PS LTE) are outlined
in [58], including both manual and automatic prioritization
adaptation, such as geo-fencing, where user priority changes
within a geographical area. Public safety networks can be
pre-planned for quick activation, short-lived (triggered by an
incident), long-lived (such as a festival) or permanent (in a
high-crime area). A particular user can change its priority
due to circumstances (e.g., police officer: normal → tactical
assault role → normal). Pre-emption provides a clear path for
high priority users by knocking other users off the system, if
needed.
The European telecommunications standards institute
(ETSI) standard for terrestrial trunked radio (TETRA) [59] has
been adopted in many countries and uses a dedicated narrow
spectrum. Rather than governments having reserved frequen-
cies for emergencies, 3GPP has been developing LTE mis-
sion critical communication standards to enable public safety
networks since Release 11. As overviewed in [60], 3GPP
Release 11 introduces public safety broadband on Band 14;
1553-877X (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/COMST.2019.2897800, IEEE
Communications Surveys & Tutorials
5
STRUCTURE
A. URLLC Potential
Applications
B. Recent Advances and
Standardization
C. Influence of Public
Safety Networks
D. Survey Outline
I. INTRODUCTION
II. LTE PHY
MECHANISMS
FOR URLLC
D. Accurate CSI
estimation
A. Structure-based
Techniques
B. Diversity-based
Techniques
C. Resource-Reuse-
based Techniques
Frame Structure
Waveform Design
Finite Blocklength
Information Theory
Frequency/Time/Space
Diversity
Modulation and Coding
Scheme
Frequency Hopping
Spectrum Sensing
In-band Full-duplex
Grant-Free NOMA
III. CROSS-LAYER
MECHANISMS
FOR URLLC
A. ARQ/HARQ
B. Radio Resource
Management
C. Multi-connectivity
D. Harmonization
IV. LTE MAC
MECHANISMS
FOR URLLC
V. EVALUATION
OF URLLC
ENABLING
TECHNOLOGIES
A. LTE Mechanisms
for URLLC in
Licensed Spectrum
B. URLLC MAC
Mechanisms for
Incumbent
Technology in
Unlicensed Spectrum
C. Study Case:
Intelligent
Transportation System
D. Coexistence in
Unlicensed Spectrum
A. PHY-layer
Evaluations
B. MAC-layer
Evaluations
C. LTE/Wi-Fi
Coexistence
Simulations
D. Unlicensed Multi-
carrier Access
Admission Control,
ARP, Pre-emption
Congestion Control -
Access Class Barring
Semi-persistent
Scheduling
Device-to-device
Wi-Fi Quality of
Service Mechanisms
Wi-Fi
Dedicated Short-range
Communications
LTE V2X
LTE Access to the
Unlicensed Spectrum
LTE/Wi-Fi
Coexistence Standards
Multefire
TM
EDCA Modeling and
Simulations
LTE/Wi-Fi
Coexistence Modeling
VI. FUTURE
RESEARCH
AREAS
Resource Block Slicing
Advanced Signal
Processing
Location-Aware
Communications
Energy Efficiency
Multichannel Diversity
Unlicensed Channel
Profiling
Carrier Aggregation
Scheduling
Scheduling
Contention-based
Access Markov-Chain
Modeling
PHY-layer Lessons
Cross-layer Lessons
MAC-layer Lessons
A. General Lessons and
Hurdles
B. Specific Research
Areas
Fig. 2. Paper organization and structure
Release 12 introduces proximity services (ProSe), including
D2D, direct discovery, and support for broadcast/multicast;
Release 13 introduces mission-critical push-to-talk (MCPTT)
[61], enhanced ProSe, and support for single-to-many trans-
missions; and Release 14 introduces mission-critical data
(MCData) and mission-critical video (MCVideo) [62].
Public safety networks are based on group communica-
tions, with emergency transmissions received by all members
of a group. The current 3GPP functional requirements for
group communications are outlined in [63]. A number of the
functions have the potential to be used for URLLC, such
as admission control, transmission priority and interruption
mechanisms. Single-cell point-to-multipoint (SC-PTM) is also
possible, which uses a common group radio network tempo-
rary identifier (RNTI) and is transmitted on the physical DL
shared channel (PDSCH), allowing scalability, without using
the multicast channel (MCH).
D. Survey Outline
The organization and structure of the survey is depicted
in Fig. 2. This survey aims to explore the PHY-layer, MAC-
layer, and cross-layer mechanisms that have the potential to
enable URLLC. In Section II, PHY layer mechanisms with
the potential to enable URLLC are considered, predominantly
from an LTE perspective, covering numerology, diversity and
resource reuse. Promising mechanisms include shortening the
TTI to reduce the round trip time (RTT), altering the waveform
to enable faster decoding, and using finite block-length infor-
mation theory to reduce the bit error rate. Section III considers
cross-layer mechanisms, covering automatic repeat request
(ARQ)/HARQ, RRM, multi-connectivity and harmonization.
LTE mechanisms with the potential to enable URLLC
for the licensed spectrum are considered in Section IV-A,
covering prioritizing bearers during random access (RA) pri-
oritization, admission and when scheduling resources, min-
imizing control signaling for periodic resource allocations,
and using D2D communications to reduce the number of
links. In Section IV-B, MAC layer mechanisms used by the
incumbent technology in the unlicensed spectrum, i.e., Wi-Fi,
are explored. The vehicular network use case is considered in
Section IV-C, covering dedicated short-range communications
(DSRC) protocols in the unlicensed bands and vehicle-to-
anything (V2X) communications in the licensed bands, which
rely on D2D communications with semi-permanent schedul-
ing (SPS). In Section IV-D, mechanisms to enable LTE to
coexist in the unlicensed spectrum are covered, including
current protocols. The challenge for LTE coexistence in the
unlicensed spectrum is to maintain the advantages provided
by the centrally scheduled LTE protocols, while assimilating
with the contention-based Wi-Fi protocols.
The impact of the PHY-layer, cross-layer and MAC-layer
URLLC enabling technologies is evaluated in Section V.
Potential areas of future research are given in Section VI. The
survey is concluded in Section VII.
II. LTE PHY MECHANISMS FOR URLLC
There exists a fundamental correlation among three key
performance indicators, reliability, latency and throughput, in