scispace - formally typeset
Search or ask a question
Proceedings ArticleDOI

Periodic Radio Resource Allocation to Meet Latency and Reliability Requirements in 5G Networks

TL;DR: This paper discusses a mechanism of deterministic resource allocation to meet the URLLC requirement in terms of reliability and latency, including initial transmissions and controlled retransmissions, and shows that when applying the proposed resource allocation technique it is possible to achieve very low error rates.
Abstract: Ultra-Reliable and Low Latency Communications (URLLC) is a challenging class of services to be supported by the fifth generation of mobile networks (5G). Among the URLLC services, many use cases, especially those related to factory automation, involve communications with relatively static radio conditions and a periodic generation of control or data packets. The transmission of these packets requires extremely low latency and ultra-reliable communication to enable realtime control of automation processes. In this paper, we discuss a mechanism of deterministic resource allocation to meet the URLLC requirement in terms of reliability and latency, including initial transmissions and controlled retransmissions. A joint resource allocation and modulation and coding schemes selection is performed so that the resource consumption is minimized, subject to latency and reliability constraints. We show that when applying the proposed resource allocation technique it is possible to achieve very low error rates.

Summary (3 min read)

Introduction

  • Among the URLLC services, many use cases, especially those related to factory automation, involve communications with relatively static radio conditions and a periodic generation of control or data packets.
  • This large range of use cases presents different requirements in terms of latency and reliability.

II. SYSTEM MODEL

  • Wired technologies, such as Ethernet-based systems, have been used for a long time for factory automation and they are still the dominant technologies in this field because of their reliability and real-time guarantee.
  • The evolution in factory automation from wired to wireless communications should take place gradually and it is commonly agreed that the wires cannot be removed everywhere.
  • The authors study focuses on the wireless communication scheduling.
  • On the wireless side, UEs are characterized by different radio conditions so that their resource requirements are different and depend on their long-term Signal to Noise Ratios (SNRs) and on their chosen MCSs.
  • The whole system has to be designed so that the final packet loss rate is very low i.e., 10−5), with a minimal number of retransmissions in order to respect the low latency requirements (i.e., few milliseconds).

III. RADIO RESOURCE ALLOCATION SCHEME

  • The authors work targets deterministic radio resource allocation mechanism over a wireless link.
  • In particular, in the time domain, they are allocated every Transmission Time Interval (TTI).
  • The authors will see, in the following section, how resource allocation can be done to achieve the URLLC requirements.
  • Indeed, a more robust MCS ensures fewer retransmissions, and then less reservation of resources for HARQ, but has a lower spectral efficiency leading to more reserved resources for the first transmission.
  • The optimal tradeoff is studied in Section IV-C. 14G subchannels are of 180 kHz, each composed of 12 consecutive and equally spaced Orthogonal Frequency-Division Multiplexing (OFDM) subcarriers.

A. Resource Allocation for first transmissions

  • The first step is to estimate the number of RBs for transmitting one applicative packet of URLLC service.
  • This number is determined by the MCS and the structure of one RB.
  • Η refers to the information rate that can be transmitted over a given bandwidth and is given by: η = ηcηslog2M (1) where M denotes modulation order, ηc denotes code efficiency and ηs denotes efficiency of signaling.

B. Resource Allocation for retransmissions

  • Since the MCS stays the same in retransmission due to HARQ, the required number of RBs is exactly the same as the previous transmission for each UE.
  • Therefore, the authors have as objectives are to determine the loss probability knowing the users MCS selections, #»µ , and the resources reserved for retransmissions H and to derive optimal resource allocation, as presented in the next section.

IV. OPTIMAL JOINT MCS SELECTION AND RESOURCE ALLOCATION

  • In order to obtain the optimal scheme, the authors start by computing the performance under a given users MCS selection and resource reservation.
  • The authors then derive the optimal scheme under performance constraints.

A. Distribution of the number of losses

  • To calculate the probability mass function of lost RBs, every user’s transmission process should be considered.
  • The authors can now compute the distribution of the overall number of retransmissions.
  • The lost number of RBs is a sum of independent weighted Bernoulli trials that are not identically distributed.
  • The distribution function is necessary to decide how many RBs should be reserved for HARQ for a given target reliability.
  • The probability of event #»α could be expressed with error rate of each user: q ( #»α) =.

B. Reliability computation

  • To evaluate the reliability of their resource allocation mechanism, the authors have to consider two possible events for loss as follows.
  • Second, even if there is enough space for a retransmissions, the retransmission may fail again.
  • The authors here adopt the policy that accommodates the highest number of retransmissions.
  • The MCS-dependent error rates for first and second transmissions δi and δ (1) i are calculated using link level simulations as will be explained next.

C. Optimal scheme

  • Above the authors have developed a framework for evaluating the performance (in terms of packet losses) for URLLC services using periodic scheduling.
  • The performance was derived for a set of selected MCSs and for a given reservation of resources for retransmissions.
  • A two stage optimization problem is proposed.
  • This search is very efficient and the complexity is logarithmic in the maximum number of RBs possible.
  • Once the minimal resource allocation for a given MCS selection is computed, the authors can derive the global optimal MCS for minimizing overall resource consumption: #»µ∗ = argmin #»µ.

V. NUMERICAL RESULTS

  • The authors consider a system where the machine type equipments are randomly and uniformly distributed throughout the factory.
  • The environment is assumed to be controlled, including interferences, so that the average SNRs are known (or estimated on a long term).
  • The authors consider a traffic with small applicative packets generated periodically, with one packet generated every 20 ms.
  • The simulation parameters are listed in Table I.

B. Performance evaluation

  • For each MCS combination, defined in Table III, the authors compute the amount of resources needed for the first transmission and for retransmissions under constraints (8).
  • Following Figure 3, the optimal combination of MCS is No.7, which corresponds to using MIMO 4×4 OFDM 4QAM for UEs with SNR 5 dB or 10 dB and Polar Alamouti OFDM 16QAM for UEs with 15 dB (see Table III).
  • The trade-off between robustness of MCS and error probability can also be observed from Figure 3.
  • The system is constrained with users having the lowest SNR, and an error rate lower than 10−5 is not achievable.

VI. CONCLUSION

  • The authors have developed a joint MCS selection and resource reservation scheme for reaching stringent latency and reliability targets in factory-like environments, while ensuring minimal resource consumption.
  • The authors considered a system with quasi-static users and reserve resources for first transmissions and a limited set of retransmissions and formulated a twostage optimization problem for selecting the optimal MCS combination.
  • The authors simulation results illustrate the optimal scheme and the impact of different reliability and latency targets on the resource reservation.

Did you find this useful? Give us your feedback

Content maybe subject to copyright    Report

HAL Id: hal-01812631
https://hal.archives-ouvertes.fr/hal-01812631
Submitted on 11 Jun 2018
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-
entic research documents, whether they are pub-
lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diusion de documents
scientiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Periodic radio resource allocation to meet latency and
reliability requirements in 5G networks
Yishu Han, Salah Eddine Elayoubi, Ana Galindo-Serrano, Vineeth Varma,
Malek Messai
To cite this version:
Yishu Han, Salah Eddine Elayoubi, Ana Galindo-Serrano, Vineeth Varma, Malek Messai. Periodic
radio resource allocation to meet latency and reliability requirements in 5G networks. 87th Vehicular
Technology Conference, VTC2018-Spring, Jun 2018, Porto, Portugal. �hal-01812631�

Periodic Radio Resource Allocation to Meet
Latency and Reliability Requirements in 5G
Networks
Yishu Han
, Salah Eddine Elayoubi
, Ana Galindo-Serrano
, Vineeth S. Varma
, Malek Messai
Orange Labs, Chatillon, France
Universite de Lorraine, CRAN, UMR 7039, France
Abstract—Ultra-Reliable and Low Latency Communications
(URLLC) is a challenging class of services to be supported
by the fifth generation of mobile networks (5G). Among the
URLLC services, many use cases, especially those related to
factory automation, involve communications with relatively static
radio conditions and a periodic generation of control or data
packets. The transmission of these packets requires extremely
low latency and ultra-reliable communication to enable real-
time control of automation processes. In this paper, we discuss
a mechanism of deterministic resource allocation to meet the
URLLC requirement in terms of reliability and latency, including
initial transmissions and controlled retransmissions. A joint
resource allocation and modulation and coding schemes selection
is performed so that the resource consumption is minimized,
subject to latency and reliability constraints. We show that when
applying the proposed resource allocation technique it is possible
to achieve very low error rates.
I. INTRODUCTION
The 5th generation of mobile networks (5G) will not only
support an evolution of traditional mobile communication
services such as personal mobile broadband services; 5G will
in addition support new use cases and enable a everything
connected society [1]. These use cases are particularly for new
Machine-Type Communications (MTC) services, including
massive MTC and mission-critical MTC. The latter category
requires communication with very high reliability, as well
as very low latencies. One area in 5G that is considered
increasingly important is, therefore, the capability to provide
Ultra-Reliable and Low Latency Communications (URLLC),
enabling new mission-critical MTC use cases [2]. Many
verticals present interesting URLLC use cases and such as
industry automation, tactile internet, virtual and augmented
reality, smart grids, autonomous vehicles and drones. This
large range of use cases presents different requirements in
terms of latency and reliability. Namely, depending on the
application the end to end latency can vary from 0.5 to 10
ms and the reliability from 10
3
to 10
5
or even fewer [10].
The URLLC framework enables real-time control and au-
tomation of dynamic processes in various fields, such as indus-
trial process automation and manufacturing, which is the focus
of this paper. Connecting machines in an industrial site enable
many new opportunities for discrete manufacturing and help
industrials to achieve a more efficient production. Today fac-
tory automation is largely based on wired connectivity. Among
other standards, IEEE TSN (Time-Sensitive Networking) is
increasingly used for ensuring deterministic communications
protocol in industrial control applications. It is a set of Ethernet
sub-standards that describe several mechanisms for guaranteed
real-time delivery of Ethernet traffic. The core concept of
TSN is deterministically scheduling traffic in queues through
switched networks.
In order to provide more flexibility for the factories of
the future, replacing wired communication links with wireless
ones is foreseen in short to mid-term. Wireless communica-
tions in industrial venues are particularly due to their potential
isolation and typical indoor environments. That is why the
use of unlicensed spectrum has been proposed by numerous
equipment vendors. This would allow the use of a proprietary
infrastructure dedicated only to the industrial site wireless
communications. Example of such systems is WirelessHART
[3]. Nevertheless, the use of unlicensed spectrum will not
guarantee communications without interference or spectrum
crowding, which is of paramount importance when it comes
to URLLC type of communications. We propose in this paper
the usage of licensed 5G spectrum for URLLC, ensuring
interference protection and full control of the operator.
We propose in this paper a joint Modulation and Coding
Scheme (MCS) selection and resource allocation scheme for
URLLC in factory automation use cases. Indeed, due to con-
straints in delay and reliability, sufficient radio resource should
be provided, not only for the first transmission but also for
retransmissions (within the delay budget). The choice of the
MCS of the different users will impact the number of resources
reserved for the first transmissions, but also the number of
retransmissions. Indeed, a robust MCS will have a lower spec-
tral efficiency, leading to a larger number of Resource Blocks
(RBs) for sending the same applicative packet. However, this
robust MCS will result in a lower packet loss probability
and will necessitate a smaller amount of reserved RBs for
retransmissions. We formulate an optimization problem where
the MCS for each user is chosen so that the overall radio
resource consumption is minimized, subject to latency and
reliability constraints. The model takes as output detailed
link level simulations that provide, for each modulation and
channel coding scheme, the packet loss probability.
The remainder of this paper is structured as follows. In
Section II we first introduce the system model along with
our design considerations for an ultra-reliable and low-latency

Fig. 1. System model
scheduling mechanism. Also, main simulation assumptions
that are going to be used in the performance evaluations
are presented. Section III provides the formalization of the
network model to calculate radio resource consumption. In
Section IV, the performance evaluations are presented with
the detailed trade-off analysis in order to find the optimal
solution. In Section V we discuss an extension of performance
evaluation to meet per User Equipment (UE) requirement.
Finally, Section VI concludes the article.
II. SYSTEM MODEL
Wired technologies, such as Ethernet-based systems, have
been used for a long time for factory automation and they
are still the dominant technologies in this field because of
their reliability and real-time guarantee. However, the interest
of using wireless communication for factory automation has
grown recently thanks to the advantages in terms of flexible
deployment and easy maintenance. The evolution in factory
automation from wired to wireless communications should
take place gradually and it is commonly agreed that the wires
cannot be removed everywhere. In our study, we consider a
mixed wired/wireless system within the factory, as illustrated
in Fig. 1. In this figure, the main factory switches are fixed and
connected via wires, while the last hop, i.e., the link between
the UEs, integrated to the machines, and the switch is wireless.
The scheduling problem could also be divided into two
parts: wired and wireless. We suppose that the TSN protocol,
or equivalent, is applied in the wired network so that low
and bounded jitter and deterministic end-to-end latency are
ensured. Our study focuses on the wireless communication
scheduling.
On the wireless side, UEs are characterized by different
radio conditions so that their resource requirements are dif-
ferent and depend on their long-term Signal to Noise Ratios
(SNRs) and on their chosen MCSs. Even if we suppose that
the long-term SNRs are known, as machines are either fixed
or characterized by a low mobility, their instantaneous radio
conditions change due to fast fading, leading to some losses of
packets. Retransmissions are then needed for recovering these
lost packets, when possible. In our work, we adopt the Hybrid
Automatic Repeat reQuest (HARQ), with Chase combining
between packets. The whole system has to be designed so
that the final packet loss rate is very low i.e., 10
5
), with a
minimal number of retransmissions in order to respect the low
latency requirements (i.e., few milliseconds). A joint optimal
resource allocation for first transmissions and retransmissions
and MCS selections are then to be designed, as detailed in the
following section.
III. RADIO RESOURCE ALLOCATION SCHEME
Our work targets (quasi) deterministic radio resource allo-
cation mechanism over a wireless link. Radio resources are
allocated into the time/frequency domain. In particular, in
the time domain, they are allocated every Transmission Time
Interval (TTI). In 4G, a TTI lasts for 1 ms, while different
TTI sizes are being defined for 5G. In the frequency domain,
instead, the total bandwidth is divided in sub-channels
1
. A
combination of a TTI and a subchannel is called RB and
corresponds to the smallest radio resource unit that can be
assigned to an UE for data transmission. In addition, a portion
of the spectrum and certain time interval should be given to
payloads of signaling. We will see, in the following section,
how resource allocation can be done to achieve the URLLC
requirements.
To guarantee deterministic scheduling, we propose that
a periodic resource reservation is performed. Based on the
average SNR, different MCSs are used for each user. The
number of RBs that will be reserved can, therefore, be
calculated. However, some packets will be lost with a user-
specific packet error rate that depends on the chosen MCS and
the SNR. An additional amount of RBs should thus be reserved
for retransmissions and, if a retransmission occurs, HARQ
with soft combining is used to augment the performance of
decoding.
We consider a system with N UEs, indexed by i. Fig. 2
shows an example of the proposed periodic resource reserva-
tion for 4 users. As we consider that the environment is slowly
changing inside the factory, the average SNRs are constant for
a relatively long time. Therefore, the reservation is performed
identically in each cycle. As the services are delay-constrained,
we only allow one retransmission. Our model can be easily
extended to a larger (but limited) number of retransmissions.
There is clearly a tradeoff between the number of resources
reserved for the first transmission and for retransmissions.
Indeed, a more robust MCS ensures fewer retransmissions, and
then less reservation of resources for HARQ, but has a lower
spectral efficiency leading to more reserved resources for the
first transmission. There is an optimal tradeoff to seek. In this
section, we start by considering that the MCS is chosen for
each UE and proposing the corresponding resource allocation
scheme. The optimal tradeoff is studied in Section IV-C.
1
4G subchannels are of 180 kHz, each composed of 12 consecutive
and equally spaced Orthogonal Frequency-Division Multiplexing (OFDM)
subcarriers. Different subcarrier spacings are defined for 5G, but our model
is sufficiently generic to cover the different cases.

Fig. 2. Periodic Resource Block reservation
A. Resource Allocation for first transmissions
The first step is to estimate the number of RBs for transmit-
ting one applicative packet of URLLC service. This number
is determined by the MCS and the structure of one RB.
First of all, we should calculate the spectral efficiency,
η, measured in bit/s/Hz, of a specific MCS. η refers to
the information rate that can be transmitted over a given
bandwidth and is given by:
η = η
c
η
s
log
2
M (1)
where M denotes modulation order, η
c
denotes code efficiency
and η
s
denotes efficiency of signaling.
Suppose that the size of an applicative packet is b bits. The
bandwidth for a RB is represented by ω. And the TTI is τ.
The number of physical RBs, R, for transmitting an applicative
packet is:
R = d
b
ητω
e (2)
In the following, we will define µ
i
as the chosen MCS for
UE i. The required RBs for the first transmission of a packet
of UE i is then:
R
i
= g(µ
i
) = d
b
η(µ
i
)τω
e (3)
We also define the users MCS selections vector
#»
µ =
[µ
1
, µ
2
, . . . , µ
N
].
B. Resource Allocation for retransmissions
Since the MCS stays the same in retransmission due to
HARQ, the required number of RBs is exactly the same as
the previous transmission for each UE. However, it is not
optimal to reserve resources of retransmissions of all UEs
as the loss probability is usually low, leading to a waste of
resources for full reservation. We suppose that the amount
of resources reserved for retransmissions is H. Therefore,
we have as objectives are to determine the loss probability
knowing the users MCS selections,
#»
µ, and the resources
reserved for retransmissions H and to derive optimal resource
allocation, as presented in the next section.
IV. OPTIMAL JOINT MCS SELECTION AND RESOURCE
ALLOCATION
In order to obtain the optimal scheme, we start by comput-
ing the performance under a given users MCS selection and
resource reservation. We then derive the optimal scheme under
performance constraints.
A. Distribution of the number of losses
To calculate the probability mass function of lost RBs,
every user’s transmission process should be considered. The
relationship between loss rate of UE
i
and MCSs could be
described by an error rate function f. The number of needed
RBs for retransmission can be modeled as a Bernoulli random
variable
i
:
i
=
(
R
i
δ
i
0 1 δ
i
(4)
with δ
i
= f (µ
i
)
where δ
i
is the error probability of the first transmission for
user i and it is related with its MCS choice. This relationship
is defined by function f.
We can now compute the distribution of the overall number
of retransmissions. Let X =
P
N
i=1
i
be the number of needed
RBs for retransmissions, we need to compute:
P (x,
#»
µ) = P (X = x|
#»
µ)
The lost number of RBs is a sum of independent weighted
Bernoulli trials that are not identically distributed. In other
words, it is the probability distribution of the total number of
lost RBs in a sequence of N independent experiments with
error probabilities δ
1
, δ
2
, ... , δ
N
. The distribution function
is necessary to decide how many RBs should be reserved for
HARQ for a given target reliability. To simplify notations, let
α
i
=
i
/R
i
to be the first transmission result for UE
i
(α
i
= 1
means a failed transmission), and
#»
α = (α
1
, α
1
, . . . , α
N
) .
The state of all possible events is represented by:
A = {(α
1
, α
1
, . . . , α
N
) |α
i
{1, 0}}
The probability of event
#»
α could be expressed with error
rate of each user:
q (
#»
α) =
N
Y
i
[δ
i
1
α
i
=1
+ (1 δ
i
)
1
α
i
=0
]
The indicator function
1
E
is equal to 1 if condition E is
verified and to 0 otherwise. The probability mass function of
lost RBs is shown as below:
P (x,
#»
µ) =
X
#»
α ∈A
1
{R(
#»
α )=x}
q (
#»
α) (5)
with R(
#»
α) =
N
X
i=1
α
i
R
i
Once we have the probability mass function of lost RBs,
the cumulative distribution function could be easily derived
(equation(6) ) and the number of RBs to be reserved will be
calculated with final loss constraint.
F
X
(x) =
X
x
i
x
P (x
i
,
#»
µ) (6)

B. Reliability computation
To evaluate the reliability of our resource allocation mech-
anism, we have to consider two possible events for loss as
follows. First, if the number of needed resources for retrans-
missions is larger than H, some of the lost packets cannot
be retransmitted, leading to a definite loss. Second, even if
there is enough space for a retransmissions, the retransmission
may fail again. Note also that, for the first event, i.e., when
there is no enough space to accommodate all retransmissions,
several policies are possible for selecting the packets to be
dropped. We here adopt the policy that accommodates the
highest number of retransmissions. For instance, if the UEs
are ordered so that R
i
R
j
if i < j and the vector of losses is
#»
α, the packets of UE 1 to I
(α, H) are served with I
(α, H)
the largest index so that
P
I
i=1
R
i
α
i
< H.
The following equation calculates the final error rate for UE
i:
e
i
(
#»
µ, H) =
X
#»
α ∈A
1
{α
i
=1}
q (
#»
α) {
1
iI
(
#»
α ,H)
δ
(1)
i
+
1
i>I
(
#»
α ,H)
}
(7)
where δ
(1)
i
denotes the error rate during the retransmission for
user i after retransmission and decoding. The MCS-dependent
error rates for first and second transmissions δ
i
and δ
(1)
i
are
calculated using link level simulations as will be explained
next.
C. Optimal scheme
Above we have developed a framework for evaluating the
performance (in terms of packet losses) for URLLC services
using periodic scheduling. The performance was derived for a
set of selected MCSs and for a given reservation of resources
for retransmissions. In this section we present an optimization
framework that minimizes the overall reserved resources while
achieving the target requirements. A two stage optimization
problem is proposed. First, for given MCSs, the minimal
reservation for HARQ is derived. Based on this result, a second
optimization problem is formulated for the selection of MCSs.
1) Optimal resource allocation knowing the users MCS
selection: For a predetermined selection of MCSs, the op-
timal reservation of resources for retransmissions (denoted
by H
(
#»
µ)) is the smallest H so that the following set of
constraints is verified:
e
i
(
#»
µ, H) Θ, i [1, N ] (8)
where Θ denotes the reliability target and e
i
(
#»
µ, H) the per
user packet loss rate of as expresed in equation (7). Since H
is an integer and e
i
(
#»
µ, H) e
i
(
#»
µ, H+1), we can simply do
a binary search with minimum values 0 and the maximum
resource blocks and find the optimal H
(
#»
µ). This search
is very efficient and the complexity is logarithmic in the
maximum number of RBs possible.
2) Global optimization: Once the minimal resource al-
location for a given MCS selection is computed, we can
derive the global optimal MCS for minimizing overall resource
consumption:
#»
µ
= argmin
#»
µ
N
X
i=1
R
i
(µ
i
) + H
(
#»
µ) (9)
This can be done by exhaustively searching over all
possible MCS combinations.
V. NUMERICAL RESULTS
We consider a system where the machine type equipments
are randomly and uniformly distributed throughout the factory.
The environment is assumed to be controlled, including inter-
ferences, so that the average SNRs are known (or estimated on
a long term). We consider for illustration three classes of SNR
(5 dB, 10 dB and 15 dB). We consider a traffic with small
applicative packets generated periodically, with one packet
generated every 20 ms. The simulation parameters are listed
in Table I.
A. Link level simulations
Optimal resource allocation analysis is performed using
a link-level simulator where different modulations (64QAM,
16QAM, 4QAM), channel coding schemes (turbo, polar) and
MIMO schemes are implemented, along with HARQ with
Chase combining. Table II provides the results of the link level
simulations, in terms of packet loss after the first transmission
and after retransmission, for each MCS. As the reliability
depends on the MCS combination
#»
µ (equation (7)), we define
in Table III the different possible combinations, eliminating
combinations for which it is obvious that the target reliability
cannot be achieved.
TABLE I
SIMULATION ASSUMPTIONS
Applicative packet size 64 bytes
Number of UE 15
Number of subcarrier 12
Subcarrier spacing 15 kHz
TTI 0.5 ms
Bandwidth of RB 180 kHz
Efficiency of signaling 11/14
Reliability target 10
5
B. Performance evaluation
For each MCS combination, defined in Table III, we com-
pute the amount of resources needed for the first transmission
and for retransmissions under constraints (8). Figure 3 shows
these allocations for the different combinations of Table III,
and in the same time illustrates the global optimal solution of
problem (9).

Citations
More filters
Journal ArticleDOI
TL;DR: The slicing concept in the 5G RAN with the related challenges and research problems is investigated to identify the plausible options for implementing the slicing concept at the RAN level by the mobile network operator to respond to the needs of verticals.
Abstract: This article investigates the slicing concept in the 5G RAN with the related challenges and research problems. The objective is to identify the plausible options for implementing the slicing concept at the RAN level by the mobile network operator to respond to the needs of verticals. We start by identifying the different slice granularity options, that is, how to define slices by combining customer and service needs. We then present how the 5G NR features can be used for facilitating slice implementation and provide typical configurations for different slice types from technology and RAN architecture perspectives. The main challenges for RAN slicing are then discussed, with special attention to the resource allocation problem between slices sharing the same spectrum band. We also investigate the multi-tenant slicing implementation in terms of the openness of the network to third parties, which is regarded as a key issue that may encourage vertical players to use operators' networks rather than deploying their own infrastructure.

149 citations


Cites background from "Periodic Radio Resource Allocation ..."

  • ...Industry automation (controlled indoor environment) Periodic generation of packets by a limited number of machines Persistent (cyclic) resource reservation for each UE [12]....

    [...]

Journal ArticleDOI
TL;DR: This paper design and implement a cloud-based novel architecture for the formal verification of IoT jobs and provide a simulation environment for a typical RT-IoT application where the feasibility of real-time remote tasks is perceived, and is the first of its kind effort to support not only the feasibility analysis ofreal-time tasks but also to provide a real environment in which it formally monitors and evaluates different IoT tasks from anywhere.
Abstract: Real-Time Internet of Things (RT-IoT) is a newer technology paradigm envisioned as a global inter-networking of devices and physical things enabling real-time communication over the Internet. The research in Edge Computing and 5G technology is making way for the realisation of future IoT applications. In RT-IoT tasks will be performed in real-time for the remotely controlling and automating of various jobs and therefore, missing their deadline may lead to hazardous situations in many cases. For instance, in the case of safety-critical and mission-critical IoT systems, a missed task could lead to a human loss. Consequently, these systems must be simulated, as a result, and tasks should only be deployed in a real scenario if the deadline is guaranteed to be met. Numerous simulation tools are proposed for traditional real-time systems using desktop technologies, but these relatively older tools do not adapt to the new constraints imposed by the IoT paradigm. In this paper, we design and implement a cloud-based novel architecture for the formal verification of IoT jobs and provide a simulation environment for a typical RT-IoT application where the feasibility of real-time remote tasks is perceived. The proposed tool, to the best of our knowledge, is the first of its kind effort to support not only the feasibility analysis of real-time tasks but also to provide a real environment in which it formally monitors and evaluates different IoT tasks from anywhere. Furthermore, it will also act as a centralised server for evaluating and tracking the real-time scheduled jobs in a smart space. The novelty of the platform is purported by a comparative analysis with the state-of-art solutions against attributes which is vital for any open-source tools in general and IoT in specifics.

35 citations


Cites background from "Periodic Radio Resource Allocation ..."

  • ...Till recent past, the realisation of RT-IoT was not possible due to the communication delays but since the introduction of mobile edge computing and 5G technologies the end-to-end delays and latencies can be ultra-reliable which is necessary for RT-IoT systems [28,29,35]....

    [...]

Proceedings ArticleDOI
26 Jun 2018
TL;DR: The user-plane latency performance of Ultra-Reliable and Low-Latency Communications (URLLC) for 15, 30, 60 and 120 kHz numerologies is studied and the service availability that is considered as essential metric for measuring the performance of the 5G networks is studied.
Abstract: The 5-th Generation (5G) of wireless networks is expected to support a wide set of use cases with ultra-reliable and very-low latency requirements New Radio (NR) system design and different numerologies (subcarrier spacings) are adopted in the 3rd Generation Partnership Project (3GPP) standards in order to satisfy these stringent requirements In this paper we study i) the user-plane latency performance of Ultra-Reliable and Low-Latency Communications (URLLC) for 15, 30, 60 and 120 kHz numerologies; ii) the service availability that is considered as essential metric for measuring the performance of the 5G networks; iii) the impact of allocating a dedicated bandwidth to URLLC services on the latency targets of URLLC packets and on the throughput of the coexisting evolved Mobile Broadband (eMBB) services

21 citations


Cites background from "Periodic Radio Resource Allocation ..."

  • ...A joint resource allocation and Modulation and Coding Scheme (MCS) that considers reliability and latency constraints is proposed for factory automation in [11]....

    [...]

Proceedings ArticleDOI
01 Oct 2019
TL;DR: This paper exploits the radio frame features in the New Radio (NR) and evaluates the users’ performances in Downlink (DL) scheme and studies the impact of different radio configurations and the dedication of a reserved bandwidth to the critical traffic on the URLLC achieved user plane latency and packet loss probability.
Abstract: The future Radio Access Network (RAN) is being engineered to accommodate the emergence of Ultra Reliable Low Latency Communications (URLLC) and to handle the possible coexistence between URLLC and enhanced Mobile Broad Band (eMBB) services. To achieve this, new features are being implemented in the future RAN to fulfill the strict requirements of the URLLC traffic in terms of latency and reliability. In this paper, we exploit the radio frame features in the New Radio (NR) and evaluate the users’ performances in Downlink (DL) scheme. We conduct extensive system level simulations in realistic network deployment considering the Industry 4.0 use case. We study the impact of 1) different radio configurations and 2) the dedication of a reserved bandwidth to the critical traffic, on the URLLC achieved user plane latency and packet loss probability. The coexistence of URLLC and eMBB traffic is also evaluated by measuring the eMBB users’ throughput.

16 citations


Cites methods from "Periodic Radio Resource Allocation ..."

  • ...In [13], a joint resource allocation and Modulation Coding Scheme (MCS) was analyzed while considering reliability and latency constraints....

    [...]

Journal ArticleDOI
TL;DR: This study presents a user-initiated probability elastic resource (UPER) approach by dynamically adjusting the probability of using the shared spectrum for eMBB and URLLC traffic based on the current success and failure status of packet transmission status and shows that UPER can improve the reliability performance up to 54% compared with other methods.
Abstract: Ultra Reliable and Low Latency Communications (URLLC) play a key role in 5G vertical markets, but pose many technical challenges especially when sharing the spectrum with Enhanced Mobile Broadband (eMBB) customers. This study aims to overcome the spectrum inefficiency issue of fully separate (FS) approach and the contention issue of the fully overlap (FO) approach. We present a user-initiated probability elastic resource (UPER) approach by dynamically adjusting the probability of using the shared spectrum for eMBB and URLLC traffic based on the current success and failure status of packet transmission status. The probabilities of successful transmission are derived for UPER, FS, and FO and partially overlap (PO) sharing spectrum approaches. We find that the successful transmission probability of UPER approach is 28% and 46% higher than FS and FO approaches, respectively. We further evaluate the reliability and throughput performance of URLLC and eMBB. When the URLLC packet load is low, the UPER method can almost achieve the best performance of the FS method. When the URLLC packet load is high, we show that UPER can improve the reliability performance up to 54% compared with other methods.

11 citations


Cites background or methods from "Periodic Radio Resource Allocation ..."

  • ...In [18], the dedicated spectrum was allocated for periodic URLLC packets to improve the reliability performance....

    [...]

  • ...• Dedicated spectrum allocation [18]–[20]: Basically, retaining some exclusive spectrum for URLLC users can improve the URLLC performance....

    [...]

  • ...In the literature, power control [13], [17] and spectrum management [18]–[23] are the two main technical directions to solve the packet collision issue in a spectrum sharing environment for URLLC and eMBB....

    [...]

  • ...In the literature, some spectrum allocation methods were proposed to reduce the collision of data packets in URLLC and eMBB multiplexing systems [18]–[23]....

    [...]

References
More filters
Proceedings ArticleDOI
08 Jun 2015
TL;DR: It is shown that it is possible to achieve very low error rates and latencies over a radio channel, also when considering fast fading signal and interference, channel estimation errors, and antenna correlation.
Abstract: Fifth generation wireless networks are currently being developed to handle a wide range of new use cases. One important emerging area is ultra-reliable communication with guaranteed low latencies well beyond what current wireless technologies can provide. In this paper, we explore the viability of using wireless communication for low-latency, high-reliability communication in an example scenario of factory automation, and outline important design choices for such a system. We show that it is possible to achieve very low error rates and latencies over a radio channel, also when considering fast fading signal and interference, channel estimation errors, and antenna correlation. The most important tool to ensure high reliability is diversity, and low latency is achieved by using short transmission intervals without retransmissions, which, however, introduces a natural restriction on coverage area.

210 citations

Proceedings ArticleDOI
01 Dec 2015
TL;DR: The coverage and capacity aspects of the 5G mission-critical MTC solution designed to meet the needs of factory automation applications are analyzed based on a series of system-level evaluations considering both noise-limited and interference- limited operations.
Abstract: Factory automation is one of the challenging use cases that the fifth generation, 5G, networks are expected to support. It involves mission-critical machine-type communications, MTC, with requirements of extreme low-latency and ultra-reliable communication to enable real-time control of automation processes in manufacturing facilities. In this paper, we discuss the deployment strategies for the 5G mission-critical MTC solution designed to meet the needs of factory automation applications. The paper analyzes the coverage and capacity aspects based on a series of system-level evaluations considering both noise-limited and interference- limited operations. It further analyzes the related trade-offs to provide insights on the network deployment strategies for a realistic factory scenario.

61 citations


"Periodic Radio Resource Allocation ..." refers background in this paper

  • ...where Θ denotes the reliability target and ei( #» μ,H) the per user packet loss rate of as expressed in equation (7)....

    [...]

  • ...As the reliability depends on the MCS combination #» μ (equation (7)), we define in Table III the different possible combinations, eliminating combinations for which it is obvious that the target reliability...

    [...]

DOI
27 Feb 2017
TL;DR: In this paper, the authors proposed a 5G service model for the next generation mobile network that enables innovation and supports progressive change across all vertical industries and across our society. And they proposed a new value chain linking stakeholders from the telecommunications world and the vertical industries in win-win situations.
Abstract: 5G is the next generation mobile network that enables innovation and supports progressive change across all vertical industries and across our society. Through its Radio Access Network (RAN) design and its orchestrated end-to-end architecture, it has the potential to boost innovation and generate economic growth in the European economy. The 5G service models support agility and dynamicity, thereby impacting the granularity, duration and trustworthiness of business relationships. The ability to combine private and public networks and data centres across multiple domains in a secure and controlled way facilitates collaborative business processes. It reshapes the digital business ecosystem with new value chains linking stakeholders from the telecommunications world and the vertical industries in win-win situations. New stakeholders emerge in this evolved ecosystem, for example cloud companies and software houses that profit from the cloudification and virtualization of the infrastructure, and brokers that facilitate sharing of spectrum and trading of connectivity and processing resources. Small and medium-sized enterprises and start-ups are able to embed 5G in their innovative products and services for existing and new customers and markets, leveraging on the Anything as a Service (XaaS) model.

43 citations

Proceedings ArticleDOI
01 Sep 2014
TL;DR: This work looks closely at industrial deployments in which production as well as consumption of messages is carried out within software tasks running on distributed embedded nodes, and presents an approach to minimize the end-to-end latency of such tasks, respecting their precedence constraints as the scheduled communication in an underlying switched TTEthernet network.
Abstract: Mixed-criticality and high availability distributed systems, like those on large industrial deployments, strongly rely on deterministic communication in order to guarantee the realtime behavior. The time-triggered paradigm-as in TTEthernet-guarantees the deterministic delivery of messages with fixed latency and limited jitter. We look closely at industrial deployments in which production as well as consumption of messages is carried out within software tasks running on distributed embedded nodes (i.e. end-systems). We present an approach to minimize the end-to-end latency of such tasks, respecting their precedence constraints as well as the scheduled communication in an underlying switched TTEthernet network. The approach is based on and validated by a large industrial use-case for which we analyze a test bed implementing our solution.

25 citations


"Periodic Radio Resource Allocation ..." refers methods in this paper

  • ...For each MCS combination, defined in Table III, we compute the amount of resources needed for the first transmission and for retransmissions under constraints (8)....

    [...]

Proceedings ArticleDOI
01 Sep 2011
TL;DR: This work addresses the time- and channel-optimal link scheduling problem for convergecast in multi-line networks, operating according to the recent WirelessHART standard.
Abstract: Fast convergecast is a critical functionality for wireless networks deployed for industrial monitoring and control. We address the time- and channel-optimal link scheduling problem for convergecast in multi-line networks, operating according to the recent WirelessHART standard. We first establish the lower bound on convergecast time and the lower bound on the number of channels for time-optimal convergecast. Then we present optimal scheduling policy for time-optimal convergecast and near-optimal policy for jointly time- and channel-optimal convergecast. Numerical and simulation results show that our scheme can provide fast convergecast inWirelessHART networks.

10 citations

Frequently Asked Questions (2)
Q1. What are the contributions in "Periodic radio resource allocation to meet latency and reliability requirements in 5g networks" ?

In this paper, the authors discuss a mechanism of deterministic resource allocation to meet the URLLC requirement in terms of reliability and latency, including initial transmissions and controlled retransmissions. The authors show that when applying the proposed resource allocation technique it is possible to achieve very low error rates. 

As a future work, the authors aim at extending their model to multiple retransmissions ( within the latency budget ) and to more distributed schemes where collisions between retransmissions may occur.