Wireless Personal Communications

https://doi.org/10.1007/s11277-020-07261-x

Performance Evaluation of Next‑Generation Wireless (5G)

UAV Relay

Shah Khalid Khan1 · Madiha Farasat2 · Usman Naseem2 · Farasat Ali3

© Springer Science+Business Media, LLC, part of Springer Nature 2020

Abstract
Future wireless communication, especially the densified 5G network using millimeter-
Wave (mmWave) will bring numerous innovations to the current telecommunication indus-
try. In such scenario, the use of Unmanned Aerial Vehicle (UAV) as Base Station (BS)
becomes one of the viable options for providing 5G services. The focus of this study is to
investigate, analyze and describe the distinctive rich characteristics of mmWave propaga-
tion in Access and backhaul network simultaneously using UAV. The mathematical frame-
work is formulated for calculating UE (User Equipment) received power for the relay path
(BS–UAV–UE) based on Friis Transmission Equation. We conduct simulations using the
ray-tracing simulator in different scenarios while comparing and verifying the simulation
results vs mathematical equations. Using ray racing simulator, the effectiveness of dif-
fracted, reflected, and scattered paths versus direct paths is described. Furthermore, using
extensive simulations, we highlight the impact of UAV location to maximize the perfor-
mance of an Amplify-and-Forward UAV based relay for providing enhanced coverage to
the users.

Keywords  Unnamed aerial vehicle (UAV) · Key point indicator (KPI) · MmWave ·
Unmanned aerial systems (UAS) · Aerial network (AN) · 5G

* Shah Khalid Khan

s3680269@student.rmit.edu.au
Madiha Farasat
madiha.farasat@student.uts.edu.au
Usman Naseem
usman.naseem@student.uts.edu.au
Farasat Ali
engineer_farasatali@yahoo.com
1
School of Engineering, RMIT University, Melbourne, VIC 3000, Australia
2
University of Technology, Sydney, Australia
3
Leadlease Group, Sydney, Australia

13
Vol.:(0123456789)
 S. Khan et al.

1 Introduction

Radio communication evolved a lot in the last 40 years moving through different eras.
In this perspective, access cellular part will be very dense to support 10 Gbps data rate
and 1-ms latency of the future wireless networks. In this context, mmWave can fulfill
the scarcity of bandwidth, and it has been trialed in many cases for access or back-
haul communication, but its capability in using it simultaneously for both access and
backhaul link is an open research area, and still need to be explored. Moreover, the
other challenges which are not yet answered, include the practical implementations of
such scenarios for fulfilling the cellular mobile industry norms and standards. Use of
UAVs in cellular communication is getting a profound interest, both from academia and
mobile network organizations [1–3]. Modernization and novelties in UAV research and
technology has resulted in unwrapping of new opportunities, especially its unparalleled
productivity in deployment of cellular communications by providing coverage, scalabil-
ity due to line-of-sight (LoS) link, and tractable movement in three-dimensional (3D)
space in two most needed scenarios: i) re-establishing communications shortly after a
disaster, to ensure lives are not lost and communities move towards an earliest recovery,
and ii) enhancing network capacity in sports events, concerts, and high-level inaugura-
tions [4, 5].
In addition to the aviation regulations about the UAVs, the key areas which require,
in-depth study are UAV channel modeling, UAV deployment, cellular network plan-
ning with UAVs, resource management, energy efficiency and trajectory optimization
[6]. The supper connected era 5G, possibly having mmWave is anticipated to endeavor
multi-gigabit bandwidth for evolving applications i.e. augmented reality/virtual reality
traffic, generation of million minutes of video content per second in 2021, and twice
growth in global mobile data traffic as fixed IP traffic from 2016 to 2021. The main
transformations in cellular mobile communication compared to 4G, as 5G is developed
and implemented, will be the use of wider spectrum allocations at tiny Wave frequency
bands, higher data rates in larger portions of the coverage area, little outage probability,
pencil beamforming antennas at both TX and RX, longer battery life, and lower OPEX/
CAPEX costs. The two most promising benefits of mmWave are: (1) huge bandwidth
availability between 30 and 300 GHz; around 200 times more spectrum as compare to
the current cellular network [7, 8], (2) generate beamforming with very high gain due
to multiple antenna systems which can offer a transmission range that exceeds around
130 m in one direction [5], and in some specific scenarios coverage can be reached up
to 200 m [9]. Nevertheless, the critical challenges of using mmWave in the wireless cel-
lular networks are propagation loss and sensitivity to blockage, depicted in Fig. 1.
This study comprises of six sections. Section  1 is based on Introduction. Section  2
highlights the work done and background studies as literature review. Section  3 dis-
cusses the system architecture. Section 4 is focused on link-level performance modeling.
Section 5 describes propagation modeling in ray-tracing simulation. Section 6 concludes
the study with discussion and future directions.

13
Performance Evaluation of Next-Generation Wireless (5G) UAV…

Fig. 1  Benefits and challenges in mmWave

2 Background and Literature Review

Use of UAV in cellular network is one of the key areas of study, both in academia and
industry in terms of its future attractiveness in wireless communication. In [10], the
authors describe the opportunities and challenges of UAVs in wireless communica-
tion technologies. Firstly, an overview of UAV supported wireless communications
is presented as summarizing the basic networking architecture for three practical sce-
narios: (1) UAV-aided ubiquitous coverage under the overloaded base station (BS) or
malfunctioning BS, (2) UAV–BS relaying, and (3) UAV–BS data and information col-
lection. The paper also provides an overview of the UAVs main channel characteris-
tics and design considerations considering both UAV-ground and UAV–UAV channels
models. Finally, enhancing techniques that utilize the UAVs mobility is also discussed.
Results are presented based on simulation under specific system parameter for path loss
with static and mobile relaying. Results demonstrate that UAV flies between the TX
and RX aiming to reduce the link distances during both UAV information reception
and relaying phases (data ferrying or load- carry-and-delivery). The authors conclude
that: (1) on-demand UAV systems are more cost-effective and can be much more swiftly
deployed; suitable for unexpected or limited-duration missions, (2) short Range Light
of Sing (LoS) leads to noticeable performance improvement over direct communication
between TX and RX, and (3) UAV controllable movement opens new opportunities for
improving performance. Finally, the authors pointed out that effective resource manage-
ment for UAV communication systems is needed. Additionally, impact of Doppler shift
due high relative velocity of UAV–BS is an open area of study. Even if the paper has a
good cover of the challenges that UAVs communication faced, there is some missing
points of this paper, for example, using the specific frequency bands of mmWave for
UAV–UAV links to be used is not discussed.
In [11], the author describes bandwidth requirement for future Unmanned Aerial
System (UAS). There is also an overview about new predictive models, and evolving
spectrum sharing technologies for UAS. Initially, the authors highlighted that a UAS

13
 S. Khan et al.

communications cell in terms of 3D shape. As per author, cell varies in relation to different
flight parameters (speed, height etc.) and services. Mathematical equations were used to
predict numbers of UAVs-in order to determine RF spectrum requirements. Results indi-
cate significant increase in commercial UAVs will outnumber public agency UAVs espe-
cially for dense urban environments. The authors conclude that: (1) UAV growth and RF
spectrum are intertwined i.e. huge spectrum is required for growing number of UAVs, (2)
Suitability of OFDM sub-carrier frequency spacing in lessening inter-carrier interference
(ICI), (3) Priority based Spectrum allocations. Finally, the authors emphasize the need for
careful infrastructure planning. However, the missing point in the paper which required in
depth study is algorithm for spectrum management with UAVs route planning.
In [12], the authors give an overview on front-haul and back-haul wireless transport over
millimeter wave (mmWave) for 5G with discussion on data rate and latency for front-haul
interfaces. In the beginning, functional split between CU (Centralized Unit) and DU (Dis-
tributed Unit) in C-RAN is proposed in three splits: (1) PHY-RF split, (2) Intra-PHY split,
and (3) PDCP-RLC split. The authors make comparison for data rate and latency require-
ments for each of the above split options, based on some assumptions, such as number of
antennas, number of ports. Results demonstrate that higher functional splits required low
data rate and reasonably high latency, and vice versa. Moreover, the authors conclude that
mmWave-based wireless back-haul/front-haul solutions are good alternatives in contrast to
optical fibre with two possible situations: (1) Capped at the frequency 100  GHz includ-
ing V and E bands where prototypes/commercial-grade products have appeared, (2) Bands
above 100 GHz (W and D) are less intense despite their great potentials and needs further
exploration for real deployment. At the end, the authors pin down the need for in depth
study and analysis of using W-Band and D-Band which offer total channel bandwidth of
17.85  GHz and 31.8  GHz respectively. Nonetheless, the missing point in the paper is to
analyze the area of unifying the transport network of existing and new front-haul/back-haul
traffic into a common-haul SDN/NFV (Software-defined networks and network function
virtualization) based packet switching network.
In [13], the author discussed the use of millimeter-wave in back-haul and access link in
the current cellular network. At the beginning, the author proposed a concept of a millim-
eter-wave overlay in heavily populated heterogeneous networks (HetNet), where small cell
base stations using millimeter-wave are incorporated into traditional cellular networks, and
there is logical split between control and user plane. The author’s main conclusions are:

• Splitting user and data plan will result in activating small cells only when needed, on
demand data coverage, which means energy saving too,
• License free 60 GHz band having 9 GHz of unbroken spectrum having high propaga-
tion loss in free space due to oxygen absorption which comforts interference between
neighborhood connections,
• Millimeter-wave outdoor channels RMS delay spread of is in the same range as indoor
and in-cabin propagation.

At the end, the authors highlight the necessity for further study on coordinated central-
ized resource management and adaptive beam forming beam tracking. However, primary
area missing in the paper was to evaluate business continuity (i.e. service continuation) in
case of outage in Small (mm Wave) cells.
In this letter, the focus is to explore, analyze and describe the distinctive rich charac-
teristics of mmWave propagation. End-to-end equations have been derived for calculating
received power by end user while getting coverage through an amplify-and-forward UAV

13
Performance Evaluation of Next-Generation Wireless (5G) UAV…

relay. To our best knowledge, simultaneous use of mmWave in both access and front-haul
network in deployment with UAVs in cellular communication system is not studied exten-
sively in the literature till date.

3 System Architecture Description

The proposed architecture in this article is depicted in Fig. 2, the aerial-terrestrial network
is based on the combination of traditional terrestrial base station (BS) communicating with
UAV equipped as an amplify-and-forward relay (UAV–BS). The back-haul link between
BS and UAV–BS is using mmWave link. The UAV–BS is providing coverage to User
Equipment (UE) through access mmWave link as well.
Motivated by the adoption of terrestrial channel models for air-to-ground channels of
static UAV–BSs; UAV–BS is deployed as a quasi-stationary aerial BS behaving like ortho-
dox cell tower [14, 15], (hence, Doppler effect is not considered). High rate mmWave links
constraints UAVs from using high altitude due to its intrinsically high path loss with grow-
ing distance from transmitter to receiver. Consequently, the altitude of a UAV–BS must
be carefully plan. Additionally, intrinsically high path loss feature of mmWave to distance
necessitate the optimal deployment of mmWave UAV–BS, in consideration to three-
dimensional (3D) distance analysis from transmitter to receiver. Isotropic antennas are con-
sidered, because, standards bodies are typically interested in isotropic or omnidirectional
models, particularly for NLoS links, since, arbitrary antenna patterns and MIMO process-
ing simulation and analysis is easy as a step forward with available isotropic antenna mod-
els. Unlike the terrestrial BS, UAV–BS can provide a dynamic coverage and had become
mature in terms of light weight and longer battery life [16]. UAV–BS is deployed as Layer
1 amplify-and-forward relay which can readily incorporate in a cellular communication
system. System parameter and notations are described in Table 1.

Fig. 2  System architecture for aerial-terrestrial networks using mmWave, UAV is used as an amplify-and-
forward relay

13
 S. Khan et al.

Table 1  Notation and parameters used for modelling

Description Parameter Unit

Unmanned aerial vehicle-BS UAV–BS –

User equipment UE –
BS coordinates (xBS, yBS, zBS) –
UAV–BS coordinates (xUAV, yUAV, zUAV) –
UE coordinates (xUE, yUE, zUE) –
BS height hUE m
UAV–BS height hUAV m
UE height hUE m
3D distance between BS and UAV–BS D2 m
3D distance between UAV–BS and UE D1 m
Horizontal distance between base station and UAV–BS d2 m
Horizontal distance between UAV–BS and UE d1 m
Horizontal distance between BS and UE D m
UE received power via direct path PUE − D dBm
UE received power via relay path PUE−R dBm
UAV received power PUAV−R dBm
BS transmitted power PBS dBm
UAV transmitted power PUAV−T dBm
BS antenna gain GBS dBm
UAV–BS antenna gain GUAV−BS dBi
UE antenna gain GUE dBi
UAV–BS antenna gain GUAV dBi
Amplification factor β dB

4 Link‑Level Performance Modelling

In mmWave band, accurate channel modelling is an important prerequisite for designing of

an efficient wireless communication network, especially for fostering new techniques that
can comply with its propagation characteristics. Broadly, channel model can be divided
into two sub-categories:

• Analytical models: mathematical analysis of channel,

• Physical models: exploring the electromagnetic properties of the signal from transmit-
ter to receiver [17].

4.1 Relay Path (BS–UAV–UE): Friis Transmission Equation

Relaying and Cooperative Diversity Protocols: Relaying is a promising aspect of future

wireless networks. 3GPP Release 10/11 (LTE-A) and IEEE 802.16j visualized the applica-
tion cases scenarios, illustrated in Fig. 3.

• Capacity Enhancement UAVs have a great potential to augment capacity of cellular

networks i.e. to increase data rate, densifying an existing network, and to cope with

13
Performance Evaluation of Next-Generation Wireless (5G) UAV…

Fig. 3  Relaying in cellular
networks

high traffic demands especially in dense urban areas or special events. Besides the
long-range connectivity, multiple UAVs anchorage can facilitate better load balancing
and traffic offload [18].
• Group Mobility The mobility models are application dependent e.g. aggregating traffic
of group users within a high-speed train or bus. Group mobility in mobile networks can
cause degradation in network quality because of dynamic changes of network utiliza-
tion.
• Coverage To provide UEs signals of an acceptable network acquisition, especially in
black or shadowed zones. Cellular connectivity for UAV systems is interesting because
it promises coverage in beyond visual line of sight scenarios.

Different cooperative diversity protocols can be employed in this context. UAVs inte-
gration as relay into Next Generation Wireless Network needs efficient UAVs place-
ment to enhance overall system efficiency. Broadly, cooperative diversity protocols can
be classified into three categories based on the type of processing and requirements by
relay and destination terminals, highlighted in Fig. 4 [19].

• Fixed Relaying Relay works as an amplify-and-forward i.e. amplify the received signal
subject to their power constraint or decode and forward relay, and then re-transmit the
signal [20],
• Selection Relaying Adaptive strategies are followed. Relay select a suitable coopera-
tive/no cooperative action based on Signal to Noise Ratio (SNR) between the transmit-
ted terminal and the relay,
• Incremental Relaying It enhances spectral efficiency of both selection and fixed relaying
by getting limited feedback form the destination terminal and relaying when required.

Fig. 4  Cooperative diversity
protocols

13
 S. Khan et al.

Using Friis Transmission Equation, and referring to Fig.  5, received power by UE is

given by [21] :
( )
𝜆2
PUAV - R = PBS + GUAV + GBS −20 log10 (1)
4ΠD2

where:

• PUAV−R is the UAV received power in dBm,

• PBS is the BS transmitted power in dBm,
• GUAV is the UAV–BS antenna gain,
• GBS is the BS antenna gain,
• λ2 is wavelength of the transmitted signal in m,
• D2 is the distance between BS and UAV–BS and it is calculated by distance formula:
( )
𝜆2
Let k2 =
4ΠD2

Then,
( )
k2
PUAV−R = PBS + GUAV + GBS −20 log10 (2)
D2

Deploying UAV–BS as amplify-and-forward relay;

PUAV−T = 𝛽 + PUAV−R (3)
where:

• PUAV−R is the UAV received power in dBm

• β is an amplification factor in dB,

Similarly, using Friis Transmission Equation, referring to Fig. 5, received power by UE,
is given by:
( )
𝜆1
PUE = PUAV − T + GUE + GUAV + GBS 20 log10
4ΠD1

where:

• PUE is the UE received power via UAV–BS in dBm,

Fig. 5  End-to-end representation of the BS To UE communication using a relay node

13
Performance Evaluation of Next-Generation Wireless (5G) UAV…

• λ1 is wavelength of the transmitted signal in m,

• PUAV−T is the UAV–BS transmitted power in dBm,
• D1 is the distance between BS and UAV–BS and it is calculated by distance formula
( )
𝜆1
Let k1 =
4ΠD1

( )
k1
PUE = PUAV−T + GUE + GUAV + GBS −20 log10 (4)
D1

Using Eq. 2, Eq. 3, and Eq. 4 received power of the UE via UAV–BS is
( ) ( )
𝜆2 k1
PUE PBS + 2GUAV + GBS + GUE −𝛽 + 20 log10 20 log10 (5)
4ΠD2 D1

Form Eq. 5, received bower by UE has dependency on three key parameters, in addi-
tion to the frequency and antenna gains:

• Distance between BS and UAV–BS ­(D2),

• Distance between UAV–BS and UE (­ D1), and
• Scaling factor (β),

5 Propagation Modelling in Ray‑Tracing Simulation

In this section, the setup of mmWave channel propagation simulation is described. Simula-
tions are performed using Wireless In-Site ray-tracing simulator. Ray-tracing simulations
provide an avenue for harvesting novel technologies. Wireless In-Site’s unique collection
of features simplifies the analysis of complex and massive propagation problems. It has the
feature of X3D propagation model, diffuse scattering, MIMO beamforming, spatial multi-
plexing, and full communication systems analysis. Moreover, it offers an empirical propa-
gation models suite, designed for urban and indoor analysis, depicted in Fig. 6.

Fig. 6  Example of wireless
in-site ray-tracing simulator
propagation paths: BS com-
municating with UAV–BS and
UAV–BS providing coverage to
ground UEs

13
 S. Khan et al.

Table 2  Simulation parameter Parameter Value

Frequency 28 GHz
Antenna Isotropic
Channel Model free space
Number of reflection 0
Number of transmissions 0
Number of diffractions 0
Transmit power 43 dBm
Number of UEs 500

Fig. 7  Propagating paths distri-

bution

5.1 Dominant Propagation Paths

For this set of simulations, we conduct a set of simulations to explore channel model of the
UAV–BS to UE. The scenario of our simulation is a semi-dense urban area for UAV–BS
at the typical height of 100 m, and UEs at a height of 1.5 m in 300 m × 300 m area, for
around 3600 receiving points. Furthermore, UAV–BS is considered as stationary to lessen
channel variations to help battery savings and increase flight time. More reflected and dif-
fracted paths validate the characteristics of mmWave as an excellent reflector of building
materials, with the reflection coefficient as large as 0.896 for tinted glass [22]. The rest of
the parameters are used as described in Table 2. Interestingly as observed in Fig. 7, only
2% among the dominant propagation paths (potential communication channels) reaching
to the receiver points are direct paths (LoS). The highest number of paths belongs to both
diffracted and reflected paths, followed by diffracted only paths, and reflected only paths.
Additionally, upon analysis of 25 propagation paths at a certain receiving point, the results
suggest that some paths arrived late, having more path length and of high signal strength.
For example, in Fig.  8, propagation path at point B, has a high Time of Arrival (ToA),
increased path length, and good signal strength, as compared to point A. The same obser-
vations are noticeable for point D, and C as well. In contrast to terrestrial BS, UAV–BS has
more focused LoS paths with the ground users, lesser reflected, scattered, and diffracted
paths, specifically around the UAV–BS, and in the terrain as well.

13
Performance Evaluation of Next-Generation Wireless (5G) UAV…

Fig. 8  Received power vs path length and time of arrival

In [23], the author deduced an analytical result for the optimal height of UAV–BS and
describes the effect of blockers (human body) and its impact on the desired height, cover-
age radius, and the number of served users. However, from our results, we observed that
for ground users, even in the case of UAV–BS, LoS or direct path is not the main source of
best signals.

5.2 Scenario 1: Variable UAV Height

In this set of simulations, we deploy UAV at horizontal distance of 60  m from the BS
with UAV height varied from 10 m to 100 m. System model is shown in Fig. 9a. We cal-
culate the received power by UEs via an Amplify-and- Forward relay (UAV) and results
are depicted in Fig. 9b. We observed that as the UAV height increases, the received power
by the UEs decreases and the greatest number of UEs are receiving good coverage when
the UAV height is at 30 m (which is the same height of the BS). Likewise, we notice that
the edge user’s coverage is optimum at this point. Additionally, we observed that with the
increase in the UAV height, the level of the received power by the UEs decreases.

5.3 Scenario 2: Variable UAV Horizontal Distance

In this set of simulation, we deploy UAV at a fixed height of 30 m (same height as the BS)
and move UAV in a horizontal direction (away) from the BS. We calculate the received
power by the UEs via the Amplify-and-Forward relay (UAV) and results are depicted in
Fig. 10. We observe that as the distance between the UAV and BS increases, the received

Fig. 9  Scenario 1: Performance of UAV relay with variable UAV height

13
 S. Khan et al.

Fig. 10  Scenario 2: Variable
UAV horizontal distance

power the by the UEs decreases. Hence, in this case, the nearest point with the BS will be
the optimum point (if applicable).

5.4 Scenario 3: Variable UAV Height and Horizontal Distance

In this set of simulations, we perform the analysis by varying both the UAV elevation and
the UAV horizontal distance. We deploy UAV at a horizontal distance of 20 m and 60 m
from the BS, respectively. For each horizontal distance, we elevate the UAV from 10 to
100 m. We calculate the received power by UEs via the Amplify-and-Forward relay (UAV)
and results are shown in Fig. 11. We observe that as the UAV height increases, the received
power by the UEs decreases and the greatest number of UEs are receiving good coverage
when the UAV height is at 30 m (which is the same height of the BS). Likewise, we notice
that the edge user’s coverage is optimum at this point.

5.5 Models Comparison and Verification

Here, we present comparison and verification of the simulation results vs mathematical

equations.
For this purpose, we consider the scenario, depicted in Fig. 12a, where terrestrial BS is
communicating with UAV, which is providing coverage to the ground UEs. We use 28 GHz
frequency band for both links (Access and Backhaul). Furthermore, an isotropic antenna
is deployed both at BS and UAV and we feed BS with a transmit power of 43 dBm. We
deploy UAV at a horizontal distance of 60 m from the BS, and UAV was elevated from 10
to 100 m. We calculate received power by UEs (around 1900) using two methods:

• Friis Transmission Equation (Eq. 3), and

Fig. 11  Scenario 3: UAV vertical and horizontal movement

13
Performance Evaluation of Next-Generation Wireless (5G) UAV…

Fig. 12  Model verification and comparison

• Wireless InSite Ray-Tracing Simulator (Free Space

In Fig.  12b, we depict the comparison of both the Friis Trans- mission Equation and
Wireless In Site simulations results.

6 Final Remarks and Future Directions

Ray-tracing simulator facilitates in the planning of coverage and understanding key propa-
gation parameters. The architecture presented in this study provides a promising solution to
the above problems through UAV (allowing for aerial-to-ground LoS) link, and a control-
lable movement (in 3D space). The goal of this study is the evaluation of the performance
of next-generation UAV networks using mmWave in Access and Backhaul links. We con-
duct simulations using the ray-tracing simulator in different scenarios while comparing and
verifying the simulation results vs mathematical equations. Using ray racing simulator, the
effectiveness of diffracted, reflected and scattered paths versus direct paths is described.
Furthermore, using extensive simulations, we highlight the impact of UAV location to
maximize the performance of an Amplify-and-Forward UAV based relay for providing
enhanced coverage to the users. We compare and verify the simulation results against the
presented mathematical framework to validate both the model and the simulation results.
This fundamental study will facilitate further research in exploring UAV-supported 5G net-
work at unparalleled mmWave frequency band
For future work, we plan to use machine learning algorithms for optimizing the deploy-
ment of UAV–BS in 3D space, identifying propagation environment features such as
mountains, ridges, rural areas, cities and determining an appropriate level of details to the
respective environmental model  [24]. Additionally, in cellular communication industry
perspective, using mmWave for cellular coverage is a key concern. We plan to extend our
study using MIMO antennas in the UAV–BS. Large antenna arrays can permit compensa-
tion of severe propagation conditions at high frequency.

References
1. Qualcomm, L. (2017). Unmanned aircraft systems—Trial report. San Diego: Qualcomm Technologies
Inc.
2. Lin, X., et al. (2018). The sky is not the limit: LTE for unmanned aerial vehicles. IEEE Communica-
tions Magazine, 56(4), 204–210.

13
 S. Khan et al.

3. Khan, S. K. (2019). Performance evaluation of next generation wireless UAV relay with millimeter-
wave in access and backhaul. Masters by Research, Engineering, RMIT University.
4. Wu, Q., Xu, J., & Zhang, R. J. I. J. S. A. C. (2018). Capacity characterization of UAV-enabled two-
user broadcast channel. IEEE Journal on Selected Areas in Communications, 36(9), 1955–1971.
5. Kong, L., et  al. (2017). Autonomous relay for millimeter-wave wireless communications. IEEE
Journal on Selected Areas in Communications, 35(9), 2127–2136.
6. Mozaffari, Mohammad, et al. (2019). A tutorial on UAVs for wireless networks: Applications, chal-
lenges, and open problems. IEEE Communications Surveys & Tutorials, 21(3), 2334–2360.
7. Niu, Y., et al. (2015). A survey of millimeter wave communications (mmWave) for 5G: opportuni-
ties and challenges. Wireless Networks, 21(8), 2657–2676.
8. Kutty, S., & Sen, D. J. I. C. S. (2015). Tutorials, Beamforming for millimeter wave communica-
tions: An inclusive survey. IEEE Communications Surveys & Tutorials, 18(2), 949–973.
9. Akdeniz, M. R., et al. (2014). Millimeter wave channel modeling and cellular capacity evaluation.
IEEE Journal on Selected Areas in Communications, 32(6), 1164–1179.
10. Zeng, Y., Zhang, R., & Lim, T. J. J. I. C. M. (2016). Wireless communications with unmanned
aerial vehicles: Opportunities and challenges. IEEE Communications Magazine, 54(5), 36–42.
11. Kakar, J., & Marojevic V. (2017). Waveform and spectrum management for unmanned aerial sys-
tems beyond 2025. In 2017 IEEE 28th Annual international symposium on personal, indoor, and
mobile radio communications (PIMRC). IEEE.
12. Kuo, P.-H., & Mourad, A. (2017). Millimeter wave for 5G mobile fronthaul and backhaul. In 2017
European conference on networks and communications (EuCNC). IEEE.
13. Weiler, R.J., et al. (2014) Enabling 5G backhaul and access with millimeter-waves. In 2014 Euro-
pean conference on networks and communications (EuCNC). IEEE.
14. Bor-Yaliniz, R.I., El-Keyi, A., & Yanikomeroglu, H. (2016) Efficient 3-D placement of an aerial
base station in next generation cellular networks. In 2016 IEEE international conference on com-
munications (ICC). IEEE.
15. Mozaffari, M., et al. (2015). Drone small cells in the clouds: Design, deployment and performance
analysis. In 2015 IEEE global communications conference (GLOBECOM). IEEE.
16. Saha, B., et  al. (2011). Battery health management system for electric UAVs. In 2011 aerospace
conference. IEEE.
17. Almers, P., et al. (2007). Survey of channel and radio propagation models for wireless MIMO sys-
tems. EURASIP Journal on Wireless Communications and Networking, 2007(1), 019070.
18. Sharma, V., Sabatini, R., & Ramasamy, S. J. I. C. L. (2016). UAVs assisted delay optimization in
heterogeneous wireless networks. IEEE Communications Letters, 20(12), 2526–2529.
19. Laneman, J.J.I.T.I.T. (2004). Cooperative diversity in wireless networks: Efficient protocols and
outage behavior. IEEE Transactions on Information Theory, 50(12), 283–289.
20. Iwamura, M., Takahashi, H., & Nagata, S. J. N. D. T. J. (2010). Relay technology in LTE-advanced.
NTT DoCoMo Technical Journal, 12(2), 29–36.
21. Saakian, A. (2011). Radio wave propagation fundamentals. Norwood: Artech House.
22. Zhao, H., et al. (2013). 28 GHz millimeter wave cellular communication measurements for reflec-
tion and penetration loss in and around buildings in New York city. In 2013 IEEE international
conference on communications (ICC). IEEE.
23. Gapeyenko, M., et al. (2018). Effects of blockage in deploying mmWave drone base stations for 5 g
networks and beyond. In 2018 IEEE international conference on communications workshops (icc
workshops). IEEE.
24. Naseem, U., Khan, S. K., Razzak, I., & Hameed, I. A. (2019). Hybrid words representation for airlines
sentiment analysis. In J. Liu & J. Bailey (Eds.), AI 2019: Advances in artificial intelligence. Lecture
notes in computer science (Vol. 11919, pp. 381–392). Cham: Springer.

Publisher’s Note  Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.

13
Performance Evaluation of Next-Generation Wireless (5G) UAV…

Shah Khalid Khan  is a research scholar at RMIT University, Melbourne,

Australia. He is a qualified ICT professional, having worked with multi-
national telecom vendors and operators in multiple domains for more
than 08  years after doing B.E (Electrical Engineering) in 2008. He
completed master by research degree from RMIT University, Mel-
bourne, Australia. His research interests include machine learning, nat-
ural language processing, next generation wireless networks, millime-
ter-wave technology,  Connected and Autonomous Vehicles (CAVs),
cybersecurity, data science and data analytics.

Madiha Farasat  is a Ph.D. student of University Of Technology, Syd-

ney. She received her master’s degree in Telecom from University of
Engineering and Technology, Lahore in 2019. Her current research
interests include Data Science and Analytics, Machine learning natural
language processing.

Usman Naseem  received his bachelor’s degree in Electrical engineer-

ing from COMSATS University, Pakistan, in 2009. He is currently
doing Master in Analytics (Research) from the University of Technol-
ogy Sydney, UTS from Australia. He has worked in leading telecom
companies for more than 8  years before joining Master in Analytics
(Research). His current research interests include Data Science and
Analytics, Machine learning, and natural language processing.

13
 S. Khan et al.

Farasat Ali  is an engineering professional serving in wold’s leading

companies from over a decade to provide technical and managerial
solutions. He completed bachelor of engineering from MUET, Jam-
shoro, Pakistan in 2006. He is always keen to take part in technological
research developments.

13