Performance Analysis of Energy Optimized LTE-V2X Networks for Delay Sensitive Real-Time Services

International Journal of Computer Networks & Communications (IJCNC) Vol.14, No.1, January 2022
DOI: 10.5121/ijcnc.2022.14103 41
PERFORMANCE ANALYSIS OF ENERGY
OPTIMIZED LTE-V2X NETWORKS FOR DELAY
SENSITIVE REAL-TIME SERVICES
Palash Kundu, Atunu Sarkar and Bhaskar Sardar
Department of Information Technology, Jadavpur University, Kolkata, India
ABSTRACT
Energy-efficient relaying technology in multi-hop data transmission can help the challenges faced in
cellular Vehicle-to-Everything (cellular-V2X) communication. However, due to high demand of emergency
service requirements of the systems such as Public Protection and Disaster Relief (PPDR), National
Security and Public Safety (NSPS), Intelligent Transport System (ITS) etc., least energy consumed user
equipment (UEs)/Vehicular-UEs are required which can either run real-time applications or relay the
application data. To support these scenarios, we present a high way based system model in rural area and
enhance its scope for applying single-hop direct, relay assisted multi-hop cellular-V2X and Store-Carry-
Forward (SCF) modes of uplink data transmission. We compare the performance of three modes of
transmissions in terms of overall energy consumption and overall transmission delay with specific delay
constraints of VoIP and video applications. With the varying cell radius and irrespective type of
applications, our numerical results, validated with ns-3 show that, least energy is always consumed in SCF
mode due to its inherent property but applications suffer a lot due to high delay incurred whereas single-
hop direct mode shows the reverse. When compared with cellular-V2X mode, overall transmission delay
for single-hop direct mode is acceptable within cell radius 600m but beyond that, relay assisted multi-hop
cellular-V2X mode always outperforms (with low latency and moderate energy consumption).
KEYWORDS
V2X communication, Energy efficiency, Direct communication, Store-carry-forward, Delay sensitive.
1. INTRODUCTION
With the exponential growth of high data rate demand, the next-generation wireless networks
have been put under pressure to offload mobile data [1][4] and simultaneously include cellular-
V2X (as Vehicle-to-Everything) communication. Cellular-V2X is a promising technology
providing the requirements of public safety services such as Public Protection and Disaster Relief
(PPDR), National Security and Public Safety (NSPS), Intelligent Transport System (ITS) etc.
This emerging technology aims to fulfil the requirements of new set of applications with
optimized energy efficiency to enhancements of road safety and traffic efficiency by enabling
reliable and low latency services. Thus, cellular-V2X is highly suitable for providing emergency
services which require delay sensitive networks, supporting real time and interactive applications
such as audio and video calls, Internet browsing etc. Moreover, energy consumption of the user
equipment (UE) is surprisingly increasing with the rapid growth of bandwidth-hungry
applications. Due to limited battery power constraint of UE/vehicular-UE (VUE), performance of
the system affects the quality-of-experience (QoE) of the users [5]. Hence, energy efficiency for
handheld equipments (with limited battery power) has become a key performance indicator (KPI)
of the cellular system which draws attention from research community to industry.

42
As known, components of V2X communication include not only the communication between
vehicles as V2V but also communication between vehicle and infrastructure as V2I, vehicle and
pedestrian as V2P, and vehicle and network as V2N (shown in Figure 1). All components of
V2X communication enable VUEs to communicate with a higher data rate for transmitting its
cached data, roadside information such as accident, speed limit, etc. quickly [6]. Thus, quality of
the effective communication provides high traffic safety and high quality-of-service (QoS) by
minimizing latency and maximizing the reliability of the network.
Figure 1. Use case with various components of V2X communication used in Public Safety Networks
In-coverage UE/VUE can directly communicate to eNB (in a single hop transmission) resulting
in high energy consumption, high end-to-end delay for long-distance communication.
Furthermore, direct mode of transmission increases network load when transceivers are at the cell
edge. In contrast, cellular-V2X mode can alleviate the issues of single-hop direct communication.
In this mode, VUEs can play a dual role i.e., 1. they can be a source/destination at the end of the
entire communication or 2. they can act as relays for receiving, processing, and forwarding data.
Relaying data reduces power consumption of the devices in one side and in other side, it
decreases latency which has potential impact on applications running at the ends. Significant
performance of out-of-coverage V2X mode is also seen when compared with in-coverage V2X
mode. The later mode cannot be intervened by eNB which results autonomous resource selection
based on sensing the environment. Thus, this mode may suffer from significant collisions and
interferences in a dense environment but still it is widely accepted for multi-hop short range
communications. Due to unavailability of the VUE in the proximity, store-carry-forward (SCF)
mode of transmission is also used as another competent scheme. Though SCF mode, as a derived
mode of cellular-V2X, is well-known least energy consumption scheme but it incurs high delay
[7].
In general, VUE (as mobile relay compared to UE as the stationary relay) broadens the scope of
seamless communication due to availability and mobility in high ways. Consequently, VUE can
help in providing emergency requirements of Public Safety Network (PSN) services depending
on efﬁcient mission-critical applications such as voice, and video communications between ﬁrst
responders/receivers and victims. Areas belonging to natural disasters such as earthquakes,
floods, tsunamis, hurricanes etc. demand emergency service requirements with immediate effect.
Due to communication infrastructure being often damaged by natural calamities to large extent,
made services unavailable or at least heavily congested, the stationary UEs in newly formed

43
adhoc network may use cellular-V2X/SCF mode of transmission for sending information to
traditional network via relay-assisted VUEs (as shown in the use case of Figure 1).
The main objective of this article is to find out the best suitable mode of transmissions among
single-hop direct mode, SCF mode, and relay-assisted multi-hop cellular-V2X mode by
optimizing the overall energy consumption of the UE and/or VUEs and average transmission
delay (from UE to eNB). The goal also includes to minimizing the power consumption in uplink
(UL) data transmission with the delay constraints of various delay sensitive applications such as
VoIP and video.
To optimize energy consumption with specific delay constraints, we first present a system model
and enhance its scope in a high way platform through rural areas. Second, by exploiting VUE as
mobile relay, we apply traditional single-hop direct (to eNB), multi-hop cellular-V2X, and SCF
mode of UL transmission on the enhanced system model. Third, based on numerical results,
validated with ns-3, we measure overall energy consumption and delay metric for three modes of
transmissions with the respective delay constraints of VoIP, and video applications.
The rest of this paper is organized as follows. In section 2, some background research works
related to the topic of the article with the overview of their contributions are discussed. We
present a system model and enhance it over the proposed scenario and formulate the problem in
section 3. Simulation and numerical results with the analysis and comparison of the performances
of three modes of data transmissions are discussed in section 4. Section 5 concludes the article
and finally, section 6 gives some future research directions.
2. RELATED WORKS
To use the potential benefit of V2X communications, many works have been done on PSN
(especially in ITS platform) in [5] and [8]-[12]. Authors in [5] formulate and propose an efficient
solution methodology of optimization problem for minimizing the total power consumption by
associating VUE. In [8], authors propose a hybrid communication approach using LTE and
DSRC (i.e., 802.11p) for supporting high bandwidth and reliable applications of road safety and
traffic efficiency such as video streaming. A new protocol architecture based on V2X
communications has been suggested in [9]-[10] to avoid traffic jam at intersections. To improve
the packet reception probability by reducing access latency and increasing high reliability, non-
orthogonal multiple access has been exploited in [11]. For safety applications, authors of [12]
improve reliability and reduce the latency of V2X communication by optimizing the network
performance against signal power and delay constraints. In communications between users and
vehicles on a highway scenario, they consider orthogonal resource block allocation and show
significant performance gain.
Relay based multi-hop cellular networks (MCNs) concept is exploited in [13]-[15] to increase the
efficiency of energy consumption in great extent for delay-tolerant applications. Authors in [13]
propose a context-based opportunistic forwarding scheme (i.e., SCF) using location based
optimum mobile relaying in two hop scenario for MCNs with D2D communications and show
significant energy benefits for delay tolerant services by comparing with other forwarding
schemes and traditional single-hop cellular communication. In [14], authors use the MCN
concept to dramatically reduce the energy consumption of eNB and UE both, by exploiting
vehicles as mobile relay in cellular network. The key concept in [15] is to search the next mobile
relay within its short transmission zone in MCN. If the next mobile relay is found, the message is
immediately forwarded otherwise it is carried. Based on the approach, authors reduce the energy
consumption compared to traditional single hop communication. To extend cellular coverage,

44
multi-hop D2D architecture on ISM band (using 802.11) is proposed in [16]. In this, authors also
propose a modified Optimized Link State Routing Protocol (OLSR) based proactive routing, i.e.,
cell-OSLR to improve QoS of 4G/5G based applications.
In [7] and its extension [17], energy-efficient V2X communication has been proposed and
analysed over LTE networks. In their articles, authors compare their proposed optimal resource
allocation based V2X enable transmission and show their proposed scheme is more energy
conserving than traditional direct transmission, decoding-and-forwarding (DF) based
transmission and opportunistic SCF transmission with varying cell radius.
All existing works show an immense potentiality of V2X communications in PSNs (including
ITS). However, no such extensive investigations made in the recent past on optimal energy
efficiency with delay constraints for delay sensitive applications which is crucial not only in the
performance of road safety and traffic efficiency applications but also helps in providing the
requirements of emergency services fast.
3. OVERVIEW OF CELLULAR-V2X AND SCF MODES OF TRANSMISSIONS
3.1. Cellular-V2X mode
For being an advance low latency communication technology, V2X mode is exhaustively used in
road transportation system to exchange information with other vehicles, pedestrian,
infrastructure, network etc. Due to several limitations of Dedicated Short Range Communication
(DSRC), such as unreliable broadcasts, short range communication etc., cellular-V2X has been
evolved in LTE and 5G-NR architectures (in 3GPP released 14 and 16 respectively) to meet the
traffic safety requirements such as high capacity, low latency, and long-range communications.
From the resource scheduling perspective, LTE-V2X supports two modes named as LTE-
cellular-V2X and LTE-direct-V2X (as Transmission Mode 3 and Mode 4 respectively). In both
the approaches, Cooperative Awareness Message (CAM) is periodically broadcasted by a VUE to
its neighbours to inform about its identification, state, velocity, location, public safety
information etc. In LTE-cellular-V2X, centralized scheduling is used over Uu interface with 1.8-
2.0 GHz band which is fully controlled by eNB. This mode of communication is operated (in UL
and downlink (DL)), when a VUE is within the coverage of eNB. On the other hand, LTE-direct-
V2X uses distributed scheduling approach over PC5 interface with 5.9 GHz band. As no
involvement of eNB is required, LTE-direct-V2X mode can be operated in-coverage as well as
out-of-coverage of eNB (using SL communication in PSSCH and PSCCH channels). LTE-direct-
V2X (as cellular-V2X Mode 4) uses SC-FDM to cover larger transmission range with limited
power. A VUE (using both dedicated and shared band in LTE-direct and LTE-cellular
respectively) must need two individual transceivers otherwise single transceiver is sufficient for
both the modes with shared band. Thus, the cellular-V2X mode of transmission not only
improves application reliability but also conserves energy compared to direct transmission.
3.2. SCF mode
Another well-accepted energy efficient mode of transmission in the domain of V2X
communications is SCF. Unlike cellular-V2X mode, SCF mode of transmission takes place in a
chunk of data rather than on a per-packet basis. Applications where delay is not a crucial
constraint such as file transfer, email, audio/video on demand etc. can run smoothly using SCF
mode of transmission. This mode is not only a delay feeding but also a highly energy conserving
mechanism. During prolonged carry phase of SCF mode, instead of transmission, data are
transported which results significant energy saving. Thus, for being highly energy optimized, this

45
mode of transmission is considered in this article. In contrary, all real time applications such as
IP telephony, video conferencing and interactive applications such as web-access, remote login
etc. suffer a lot in this mode. Moreover, SCF mode of transmission internally uses V2X mode in
terms of V2P and V2I while receiving text/data from UE and uploading to eNB respectively. The
store and followed by carry operations of SCF mode also reduces the energy consumption by
minimizing the required internal nodes-processing and protocol overhead. For more, readers may
study [13]-[15].
4. SYSTEM MODEL AND PROBLEM FORMULATION
In this section, we present a highway based system model and formulate the problem to measure
energy consumption and overall transmission delay due to the transmission of VoIP/video data
from an UE (which is almost at the cell edge) to eNB in single-hop direct mode, multi-hop
cellular-V2X mode, and SCF mode of transmissions. Various notations used in the problem
formulation and their meanings are given in Table I.
Table I. Nomenclature
Notations Meanings Notations Meanings
ω Antenna characteristics heNB Antenna height of eNB
v Velocity of the vehicle hUE Antenna height of UE
r Radius of the cell pmax Max. transmit power of UE
α Path loss constant pd Transmit power of UE
Ψ Noise power pc Circuit power of UE
M VoIP/video data size B Channel bandwidth
w Width of the road Dmax Max. transmission delay
rUE-eNB Distance between UE and eNB n Effective no. of vehicles in V2V
communication
rV-V Distance between vehicle and vehicle pUE-V Transmit power of UE to vehicle
sUE-V Slant height between UE and vehicle pV-V Transmit power of vehicle to vehicle
sV-eNB Slant height between vehicle and eNB pV-eNB Transmit power of vehicle to eNB
dmax Max. transmission range of V2V
communication
TUE-eNB Transmission delay from UE to eNB
TV-eNB Transmission delay from vehicle to
eNB
TUE-V Transmission delay from UE to vehicle
Ei Energy consumed in ith
phase, iϵ{1, 2,
3}
TV-V Transmission delay from vehicle to vehicle
Figure 2. High way based system model supporting various modes data transmissions

46
In the system model (shown in Figure 2), we assume a two-way highway road (in rural areas)
where a stationary UE is at the LTE-cell edge with radius r (i.e., rUE_eNB) and the vehicles
(functioning as mobile relays) move along the road towards/away from UE at a constant velocity
v. We further assume that the UE (as an intelligent relay) is either a part of an ad-hoc network
where VoIP/video data with size M, is generated requiring for emergency services or itself a
source of data. By using relay assisted vehicles (i.e., VUE), data can be transmitted to
infrastructure based network using any three modes of transmissions.
4.1. Energy consumption in single-hop direct mode:
For calculating achievable data rate in conventional UL transmission, basic Shannon capacity
formula is stated as:
(1)
Following [17], SINR (Signal-to-Interference-Plus-Noise-Ratio) is calculated as:
(2)
where, (BER as bit error rate).
Substituting Eq. (2) in Eq. (1), the achievable data rate corresponds to transmit power is obtained
as:
(3)
where, K is a function of pd.
Hence, using Eq. (3), direct UL transmit power of UE for data rate K can be obtained as:
(4)
So, pd is a strictly monotonically increasing function of K (provided all other factors remain
constant).
Thus, overall energy consumption (Ed) in UL single-hop direct mode of transmission of UE is
formulated as:
(5)
The objective function in Eq. (5) satisfies the constraints:
(5.1)
and,

47
(5.2)
In eq. (5.1) and (5.2) ensure the power transmission and transmission delay within the range of
respective thresholds of cellular network.
As Eq. (4) is a strictly monotonic function and also , we get . Moreover,
at pmax the Ineq. (5.2) also satisfies the inequality: .
Thus, modifying Eq. (5), the objective function is reconstructed as:
(6)
subject to,
(6.1)
4.2. Energy consumption in multi-hop cellular-V2X mode
In cellular-V2X mode, data are first transmitted by UE to a nearby vehicle (i.e., relay assisted
VUE) moving towards eNB, and then receiving VUE decodes, processes, and forwards the data
to the next VUE. Thus, the transmission repeats in multi-hop until the data are forwarded by the
VUE to eNB in close proximity. Hence, overall transmission in this mode can be divided in three
following phases as shown in Figure 2.
 1st
phase: As VUE is within the coverage, UE directly transmits data to VUE (which is
nothing but a V2P transmission and spanned from A1 to A2).
 2nd
phase: VUEs move gradually towards/away from the eNB and transmission in this phase
is due to relaying data in a multi-hop fashion (which is nothing but a multi-hop V2V
transmission and spanned from A2 to B1).
 3rd
phase: As and when eNB is in the close proximity of the VUE, the received data is directly
sent to eNB by UL transmission (which is nothing but a V2I transmission and spanned from
B1 to B2).
4.2.1. Energy consumption in 1st
phase of cellular-V2X mode
Since, UE directly uploads the data to its nearest VUE, the energy consumption (E1) in this phase
can be obtained as:
(7)
It is obvious that E1 << Ed. However, for complete transmission of data, the following constraint
should be satisfied by TUE_V.
(8)
where, sUT-V is a function of t and defined by: sUE-V = [hUE
2
+ (v.TUE-V – v.t)2
]1/2
.

48
4.2.2. Energy consumption in 2nd
phase of the cellular-V2X mode
In this phase, each VUE transmits data to nearby VUE at each hop of entire multi-hop V2V
transmissions. To send data destined to eNB, each VUE functions as a relay for receiving and
transmitting the data unit. Hence, energy consumption (E2) in this phase is the summation of
energy consumed by each VUE of an individual hop. For simplicity, it is assumed that VUEs are
equidistant (as they are moving with uniform velocity) and data transmission time is constant for
VUE. Thus, E2 can be calculated as:
(9)
For guaranteed transmission of the message in TV-V interval at each hop of V2V transmission, the
following constraint is satisfied by TV-V, which is stated as:
(10)
4.2.3. Energy consumption in 3rd
phase of cellular-V2X mode
Like E1 in Eq. (7), the energy consumption (E3) in 3rd
phase for uploading data unit to eNB by the
VUE (in close proximity of eNB) of the entire multi-hop V2V transmission can be obtained as:
(11)
For complete transmission of the message to eNB, the transmission time (i.e., TV_eNB) of the
vehicle satisfies the following condition which is stated as:
(12)
where, sV-eNB = [heNB
2
+ w2
+ (v.TV-eNB – v.t)2
]1/2
is a function of t.
Combining Eq. (9), Eq. (11), and Eq. (13), total energy consumption in multi-hop cellular-V2X
mode can be formulated as:
(13)
In eq. (13) also satisfies the constraints, sUE_V ≤ dmax and rV-V ≤ dmax.
To obtain optimized energy consumption in cellular-V2X mode, the following inequalities are
also satisfied by In eq. (13), which are stated as:
(13.1)
(13.2)
(13.3)
4.3. Energy consumption in SCF mode
In SCF mode, UE transmits data unit to nearby VUE which is similar to 1st
phase of cellular-V2X
mode. Unlike its 2nd
phase, the receiving VUE transmits data neither to adjacent VUE nor to eNB
directly (in UL transmission). Instead, the receiving VUE stores the entire unit of data and carries
it towards eNB. Thus, energy consumption in this phase of SCF mode is zero. Like 3rd
phase of

49
cellular-V2X mode, the transceiver-VUE simply uploads the data unit to eNB in its close
proximity and thus completes the entire transmission. Overall energy consumption in this mode
depends on 1st
and 3rd
phases only. Therefore, using Eq. (7) and Eq. (11), total energy
consumption in SCF mode is obtained as:
(14)
where, In eq. (14) also satisfies the constraints (8) and (12).
4.4. Average transmission delay (UE-to-eNB)
Average respective transmission delays (D1 and D3) of 1st
phase and 3rd
phase of cellular-V2X
and SCF modes are equal and can be computed as:
and, (15)
Average respective transmission and transportation delay in 2nd
phase of cellular-V2X and SCF
modes are calculated as:
(16.1)
(16.2)
Thus, average transmission delay from UE to eNB of direct, cellular-V2X and SCF modes can be
formulated as:
(17.1)
(17.2)
(17.3)
5. RESULT ANALYSIS
The goal of this section is to evaluate the performance of single-hop direct mode, multi-hop
cellular-V2X mode, and SCF mode of UL data transmission in an LTE cell in terms of energy
consumption and average transmission delay (from UE to eNB). To optimize the metrics (i.e.,
energy and transmission delay), we use the derived equations (of section 4) and parameter values
of Table II (which are adopted from [7] and [17]) and carry out the (numerical and simulation)
results. Based on optimized metrics value, we find out the best suitable mode of transmission for
delay-sensitive applications.
Table II. Parameter Values
Parameter Value Parameter Value
r 500-1000 m pc 100 mW
M (Mbit) VoIP data: 0.5 - 3.0
Video data: 10 - 60
v 72 Km/h
Dmax (sec) 15 (VoIP), 20 (Video)
B 15 MHz dmax 200 m
Ψ -114 dBm/MHz pmax 250 mW
α 3.75 heNB 30 m

50
ω 10-2
hUE 1.5 m
BER 10-6
w 40 m
5.1. Simulation setup
Entire simulation is carried out in ns-31
version 3.0 over LTE module. Inside the LTE module, we
use LTE-Lena and LTE-D2D2
modules. To simulate the system model in LTE-V2X
environment, we explore the programs of [18]-[19] as the base model of LTE-D2D
communication, and its extension in [20] to support UE-to-Network relay operation. This
includes D2D communication over sidelink (SL) application which only supports IPv6-based
relaying feature in UE-to-Network. Data rate varies with the variation of transmission power,
controlled by LteUePhy::TxPower of UE. For simplicity, we consider six moving nodes (as
VUEs at an average speed 72 km/hr) in between UE and eNB where five nodes move in the
direction of eNB and one node moves in the reverse direction which is a typical case, seen in a
highway. The classes LteUeRrc and LteSlUeRrc2
are responsible for creating and maintaining
one-to-one direct communication via SL between every pair of UEs/VUEs. All intermediary
nodes (i.e., VUEs) are connected through the SL for providing relay services using the object
ProseHelper of the class LteSideLinkHelper2
.
To employ UDP applications, we set up UDP client-server model by adding UdpClientHelper
and UdpServerHelper classes at UE and Remote UE respectively. We also provide required set of
attributes values in the constructor such as server port number at UE etc. To send UDP-
datagrams, the object of ApplicationContainer class always remains active at UE. The
SetAttribute method of the class UdpClientHelper, sets maximum number of packets to be
transmitted, packet intervals, and maximum packet size etc. To simulate the entire system model,
we consider datagram size as 1024 bytes and packet interval as 1 msec. To support relay-assisted
cellular-V2X mode, we run the simulator for 5 sec and 22 sec for sending 185 VoIP data packets
and 3691 video data packets respectively, one after another from UE/VUEs to remote-UE.
5.2. Simulation and numerical results
Both simulation and numerical results are shown in Figure 3, Figure 4, and Figure 5. In all
figures, the simulation results are slightly higher than numerical results. This bit of deviation is
due to: 1. there is an impact of signal strength in simulation results leading higher delay and more
energy consumption. 2. an inter packet delay is set to 1 msec in simulation which causes higher
delay compared to numerical results. For SCF mode of all the figures, energy consumption and
delay in the simulation and numerical results are almost same which noticeably differ in case of
single-hop direct and multi-hop cellular-V2X modes. Table III, IV, and V explicitly show the
detailed empirical results of three modes of transmissions for VoIP and video applications.
Table III. Overall Energy Consumption W.R.T. Messaze Size
Overall energy consumption
for VoIP application
for Video application
Message size
(Mbit)
Direct mode
(mJ)
SCF mode
(mJ)
V2X mode
(mJ)
Message size
(Mbit)
Direct mode
(mJ)
SCF mode
(mJ)
V2X mode
(mJ)
0.5 32.00 2.50 19.04 10.00 600.00 50.00 385.00
1.0 66.89 5.00 38.49 20.00 1200.00 100.00 769.00
1.5 100.34 7.50 57.93 30.00 1800.00 150.00 1158.00
1
https://www.nsnam.org/
2
https://github.com/usnistgov/psc-ns3/blob/psc-4.0/src/lte/examples/d2d-examples/lte-sl-in-covrg-relay-building.cc

51
2.0 134.89 10.00 76.98 40.00 2400.00 200.00 1543.00
2.5 169.00 12.50 96.42 50.00 3000.00 250.00 1928.00
3.0 197.00 15.10 115.87 60.00 3600.00 300.00 2313.00
Table IV. Overall Energy Consumption W.R.T. Cell Radius
Cell radius
(m)
Direct mode
(mJ)
SCF mode
(mJ)
V2X mode
(mJ)
Cell radius
(m)
Direct mode
(mJ)
SCF mode
(mJ)
V2X mode
(mJ)
500 90.00 7.50 57.93 500 1800.00 150.00 1158.00
600 180.00 7.50 59.27 600 3600.00 150.00 1189.00
700 321.00 7.50 61.95 700 6300.00 150.00 1243.00
800 --- 7.50 64.00 800 --- 150.00 1287.00
900 --- 7.50 70.90 900 --- 150.00 1423.00
1000 --- 7.50 74.69 1000 --- 150.00 1498.00
Table V. Overall Transmission Delay W.R.T. Cell Radius
Overall transmission delay
Overall transmission delay
Cell radius
(m)
Direct mode
(sec)
SCF mode
(sec)
V2X mode
(sec)
Cell radius
(m)
Direct mode
(sec)
SCF mode
(sec)
V2X mode
(sec)
500 0.30 25.075 0.189 500 6.00 26.50 11.50
600 0.60 30.075 0.382 600 13.00 31.50 11.78
700 1.07 35.075 0.575 700 23.42 36.05 12.26
800 --- 40.075 0.764 800 --- 41.05 12.50
900 --- 45.075 0.957 900 --- 46.05 13.30
1000 --- 50.075 1.150 1000 --- 51.05 13.62
Figure 3 demonstrates the total energy consumptions in three modes with the fixed cell radius,
500 m and 6 vehicles. It is clearly seen that energy consumption increases gradually with the
increase of message size. Being at the cell edge, UE transmits data to eNB and covers a long
distance and thus energy consumption is achieved the maximum in single-hop direct mode of
transmission. In relay-assisted multi-hop cellular-V2X mode, energy consumption depends not
only on UE but also relay-assisted VUEs moving towards eNB. As a result, cellular-V2X mode is
always comparable than other two modes. On the other hand, SCF mode internally uses V2X
transmission in first and last hops of its entire path of communication with transmission &
transportation. Since energy consumption in the intermediary hops of SCF mode is zero as
transmission is entirely stopped in this phase for carrying data, the overall energy consumption
reaches to minimum. Though SCF mode is most energy efficient, but delay sensitive applications
suffer a lot in this mode for its transportation phase.

52
Figure 3. Energy consumption w.r.t. (a) VoIP data size, and (b) video data size for r = 500 m and n = 6
Performance of three modes transmissions in terms of energy consumption w.r.t. varying cell
radius are compared in Figure 4. It is clearly seen that, overall energy consumption increases
exponentially in direct mode, almost linearly in cellular-V2X mode, and remains constant in SCF
mode with the increase of cell radius for respective average data size of VoIP and video
applications. Being at the cell edge and covering long distance, the channel gain between UE and
eNB reduces significantly, which results more transmission power and longer transmission time
in direct mode. In this mode, UE transmits data with its full battery power (i.e., 250 mW) at the
cell radius 500 m, and entire transmission stops at cell radius 700 m which is due to fully
exhausted battery power. In contrary, least power and least transmission delay are incurred at
each hop of V2V mode which results overall significant performance. On the other hand, total
energy consumption of SCF mode depends only on 1st
and 3rd
phases and thus, it is independent
of cell radius. As a consequence, SCF mode is most advantageous only in term of energy
consumption for large cell radius.

53
Figure 4. Energy consumption w.r.t. cell radius, where (a) average VoIP data size = 1.5 Mb and (b) average
video data size = 30 Mb with n = 6
Figure 5, shows the effect of cell radius on average transmission delay (from UE to eNB) of three
modes for average respective data sizes (of VoIP and video applications). Transmission delay has
a significant impact on delay-sensitive applications. Due to huge delay incurred by SCF mode,
none of the real-time applications run smoothly. On the other hand, average transmission delays
of single-hop direct and multi-hop cellular-V2X mode are always considerable (with delay
thresholds 15 sec for VoIP, and 20 sec for video applications with their respective acceptable data
rates). Irrespective type of applications, transmission delay of direct mode is much better than
cellular-V2X mode within cell radius 600 m, which is more prominent for video application (in
Fig. 4(b)). In direct mode, the entire transmission stops at the cell radius 700 m onwards as the
battery power is full exhausted, which cripples the applications running at the ends. This is due to
the maximum power allocation of UE for achieving high data rate. However, average
transmission delay of cellular-V2X mode always outperforms than direct mode beyond 600 m
cell radius. This is because of total power allocation of 6 VUEs (in our experiment) is greater
than maximum power of UE in direct mode. Thus, cellular-V2X mode is the best suitable

54
transmission mode, optimizing both overall energy and overall transmission delay for all delay
sensitive real time applications than other two because this mode exploits the relaying and
mobility features of VUEs.
From Figure 3, Figure 4, and Figure 5, the following points are derived:
 Delay sensitive applications seamlessly run in direct mode satisfying conditions: cell radius <
700 m and transmission delay ≤ Dmax (which is application specific constraint).
 SCF mode of transmission should be avoided for the applications with the requirement of
(soft) guaranteed latency.
 Relying-based multi-hop cellular V2X mode is proved to be provided services requiring
accessibility, latency and throughput with sustainable energy consumption.
Figure 5. Average transmission delay (from UE to eNB) w.r.t. cell radius where Mavg.(VoIP) = 1.5 Mb, avg.
Mavg.(Video) = 30 Mb, Dmax(VoIP) = 15 sec, Dmax(Video) = 20 sec with n = 6

55
6. CONCLUSIONS
In this article, we consider relaying-based cellular-V2X enabled delay-sensitive model to achieve
energy-saving and low latency communication while UE transmits data from the cell edge.
Satisfying the system model, we formulate the problem for direct, cellular-V2X, and SCF mode
of transmissions with the guaranteed delay constraints of various delay sensitive applications.
Numerical results validated by ns-3, demonstrate that direct mode always consumes maximum
energy with the considerable delay, whereas least energy with maximum delay (beyond the
threshold) is always incurred in SCF mode. On contrary, cellular-V2X mode requires moderate
energy and minimum delay. Though there is always a trade-of between energy consumption and
delay, the multi-hop cellular-V2X mode is most preferable for Public Safety Networks (PSNs)
due to its moderate power with satisfactory energy-saving-low-delay requirement. Thus, the
relay-assisted cellular-V2X mode is exhaustively used in UL data transmission for real-time
applications of PSN.
7. FUTURE WORKS
The system model (used in this article) may be further enhanced to support highway scenarios in
rural as well as urban areas for analyzing the impact of relay-based LTE-V2X/5G-NR-V2X
networks for various services provided by transport protocols. Some optimal and/or opportunistic
transmission modes may also be included for providing better services.
CONFLICT OF INTEREST
Authors of the paper declare no conflict of interest in all respects.
ACKNOWLEDGEMENTS
This work is partially supported by the RUSA 2.0 project, Ref. no.: R-11/662/19, and the DST
SERB project, File no.: EEQ/2017/000563, at Jadavpur University, Kolkata, India.
REFERENCES
[1] K. Nasim and T. J. Hall, "Techniques for Offloading LTE Evolved Packet Core Traffic Using
Openflow : A Comparative Survey & Design Reference," International Journal of Computer
Networks & Communications (IJCNC), vol. 11, no. 1, pp. 109-125, Jan. 2019.
[2] N. N. Sirhan, G. L. Heileman, C. C. Lamb, "TRAFFIC OFFLOADING IMPACT ON THE
PERFORMANCE OF CHANNEL-AWARE/QOS-AWARE SCHEDULING ALGORITHMS FOR
VIDEO-APPLICATIONS OVER LTE-A HETNETS USING CARRIER AGGREGATION,"
International Journal of Computer Networks & Communications (IJCNC), vol. 7, no. 3, pp. 75-90,
May 2015.
[3] P. Kundu, M. K. Rana, B. Sardar, and D. Saha, "Finding Practicable Nesting Level in Multi-level
Nested Mobile Networks," in proc. of 2019 IEEE International Conference on Advanced Networks
and Telecommunications Systems (ANTS), pp. 1-6, Dec. 2019.
[4] P. Kundu, M. Mahata, M. K. Rana, and B. Sardar, "Improving Spectral Efficiency for Device-to-
Device Data Offloading in Underlay Cellular Networks," in proc. of 3rd
IEEE Conference on
Information and Communication Technology, pp. 1-6, Dec. 2019.
[5] M.Saimler, and S.C. Ergen, "Power Efficient Communication Interface Selection in Cellular Vehicle
to Everything Networks," in proc. of 2019 IEEE Wireless Communications and Networking
Conference (WCNC), pp. 1-6, Oct. 2019.
[6] U. Demir, C. Toker and Ö. Ekici, "Energy-Efficient Deployment of UAV in V2X Network
Considering Latency and Backhaul Issues," in proc. of 2020 IEEE International Black Sea
Conference on Communications and Networking (BlackSeaCom), 2020, pp. 1-6, May 2020.

56
[7] C. Zheng, D. Feng, S. Zhang, X. Xia, G. Qian, and G. Y. Li, "V2X-Enabled Energy-Efficient
Transmission in Cellular Networks," 2018 10th
International Conference on Wireless
Communications and Signal Processing (WCSP), pp. 1-6, Oct. 2018.
[8] M. B. Brahim, Z. H. Mir, W. Znaidi, F. Filali, and N. Hamdi, “QoS aware video transmission over
hybrid wireless network for connected vehicles,” IEEE Access, vol. 5, pp. 8313–8323, June 2017.
[9] S. E. Li, S. Xu, X. Huang, B. Cheng, and H. Peng, “Eco-departure of connected vehicles with V2X
communication at signalized intersections,” IEEE Trans. Veh. Technology, vol. 64, no. 12, pp. 5439–
5449, Dec. 2015.
[10] L. Le, A. Festag, R. Baldessari, and W. Zhang, “Vehicular wireless short range communication for
improving intersection safety,” IEEE Communication Mag., vol. 47, no. 11, pp. 104–110, Nov. 2009.
[11] B. Di, L. Song, Y. Li, and G. Y. Li, “Non-orthogonal multiple access for high-reliable and low-
latency V2X communications in 5G systems,” IEEE J. Select. Areas Communication, vol. 35, no. 10,
pp. 2383–2397, Oct. 2017.
[12] A. Masmoudi, S. Feki, K. Mnif, and F. Zarai, “A mixed traffic sharing and resource allocation for
V2X communication,” in proc. of Int. Conf. EBus. Telecommunication New York, NY, USA:
Springer, pp. 219–233, Nov. 2019.
[13] B. Coll-Perales, J. Gozalvez, and V. Friderikos, “Energy-efficient opportunistic forwarding in multi-
hop cellular networks using device-to-device communication” in Emerging Telecommunication
Technologies,” in Transactions on Emerging Telecommunications Technologies, vol. 27 , no. 2 , pp.
249-265, Feb. 2016.
[14] P. Kolios, V. Friderikos, and K. Papadaki, "Ultra Low Energy Store-Carry and Forward Relaying
within the Cell," 2009 IEEE 70th Vehicular Technology Conference Fall, pp. 1-5, Sep. 2009.
[15] B. Coll-Perales, J. Gozalvez, and V. Friderikos, "Opportunistic networking for improving the energy
efficiency of multi-hop cellular networks," in proc. of 2014 IEEE 11th
Consumer Communications
and Networking Conference (CCNC), pp. 569-574, Jan. 2014.
[16] Y. Lmoumen, Y. Ruichek, and R. Touahni,"A Cooperative Approach to Extend Cellular Coverage
via D2D Architecture based on OLSR Protocol," International Journal of Computer Networks &
Communications (IJCNC), vol. 12, no. 5, pp. 81-96, Sept. 2020.
[17] C. Zheng, D. Feng, S. Zhang, X. Xia, G. Qian, and G. Y. Li, "Energy Efficient V2X-Enabled
Communications in Cellular Networks," in IEEE Transactions on Vehicular Technology, vol. 68, no.
1, pp. 554-564, Jan. 2019.
[18] R. Rouil, F.J. Cintrón, A. Ben Mosbah, and S. Gamboa, “An LTE Device-to-Device module for ns-
3”, in proc. of the Workshop on ns-3, 15-16 June 2016, Seattle (Washington).
[19] R. Rouil, F.J. Cintrón, A. Ben Mosbah, and S. Gamboa, “Implementation and Validation of an LTE-
D2D Model for ns-3”, in proc. of the Workshop on ns-3, 13-14 June 2017, Porto (Portugal).
[20] Gamboa, S., Thanigaivel, R. and Rouil, R., “System Level Evaluation of UE-to-Network Relays in
D2D enabled LTE Networks”, in proc. 2019 IEEE 24th
International Workshop on Computer Aided
Modeling and Design of Communication Links and Networks, September 2019, Limassol (Cyprus).

57
AUTHORS
Palash Kundu was born in West Bengal, India. He received his MCA degree from the
Department of Computer Science and Engineering, and ME degree in Software
Engineering from the Department of Information Technology, Jadavpur University,
Kolkata, India in 2009 and 2014 respectively. He is currently working as an Assistant
Professor in the Department of Information Technology, Jadavpur University. His
research interest includes Mobile and Wireless Communications. He published various
papers in reputed IEEE conferences.
Atunu Sarkar was born in West Bengal, India. He received his ME degree in Software
Engineering from the Department of Information Technology, Jadavpur University,
Kolkata, India in 2021. His research interest includes Mobile and Wireless
Communications. He has a keen interest in ns-3.
Bhaskar Sardar was born in West Bengal, India. He received his Ph.D., M.E, and B.E.
in 2010, 2003, and 2001 respectively from the Department of Computer Science and
Engineering, Jadavpur University, Kolkata, India. He is currently working as a Professor
in the Department of Information Technology, Jadavpur University. His research
interest includes Mobile and Wireless Communications. He published various papers in
reputed International Journals and Conferences including IEEE, Elsevier, Springer, etc.

Performance Analysis of Energy Optimized LTE-V2X Networks for Delay Sensitive Real-Time Services

More Related Content

What's hot (19)

Similar to Performance Analysis of Energy Optimized LTE-V2X Networks for Delay Sensitive Real-Time Services (20)

More from IJCNCJournal (20)

Recently uploaded (20)

Performance Analysis of Energy Optimized LTE-V2X Networks for Delay Sensitive Real-Time Services