electronics

Article
Broadband Microstrip Antenna for 5G Wireless Systems
Operating at 28 GHz
Rafal Przesmycki *, Marek Bugaj and Leszek Nowosielski

Department of Electronics, Military University of Technology, 00-908 Warsaw, Poland;

marek.bugaj@wat.edu.pl (M.B.); leszek.nowosielski@wat.edu.pl (L.N.)
* Correspondence: rafal.przesmycki@wat.edu.pl; Tel.: +48-504-059-739

Abstract: Communication systems have been driven towards the fifth generation (5G) due to the
demands of compact, high speed, and large bandwidth systems. These types of radio communication
systems require new and more efficient antenna designs. This article presents a new design solution
of a broadband microstrip antenna intended for use in 5G systems. The proposed antenna has a
central operating frequency of 28 GHz and can be used in the LMDS (local multipoint distribution
service) frequency band. The dimensions of the antenna and its parameters have been calculated,
simulated, and optimized using the FEKO software. The antenna has a compact structure with
dimensions (6.2 × 8.4 × 1.57) mm. Rogers RT Duroid 5880 material was used as a substrate for the
antenna construction, which has a dielectric coefficient of 2.2 and a thickness of 1.57 mm. The antenna
described in the article is characterized by a low reflection coefficient of −22.51 dB, a high energy
gain value of 3.6 dBi, a wide operating band of 5.57 GHz (19.89%), and high energy efficiency.

Keywords: antenna; microstrip antenna; antenna 5G; 28 GHz band; 5G wireless system; broad-
band antenna



1. Introduction


The growing demand for telecommunications services is stimulating the development
Citation: Przesmycki, R.; Bugaj, M.;
Nowosielski, L. Broadband Microstrip
of new call-handling technologies. Each generation of mobile technologies has brought
Antenna for 5G Wireless Systems Op-
with it an increase in the data transmission speed along with improved connection quality
erating at 28 GHz. Electronics 2021, and new functionalities. The fourth generation (4G) technology, which is currently in
10, 1. https://dx.doi.org/10.3390/ use, has been available worldwide since 2009. The fifth generation (5G) network will
electronics10010001 enable a number of new services, including those related to the Internet of Things (IoT)
and the concept of smart cities. The new technology will make use of low, medium,
Received: 20 November 2020 and high frequency bands, all of which have their advantages and limitations. However,
Accepted: 18 December 2020 wide-scale deployment of a 5G network requires preparation of antenna infrastructure
Published: 22 December 2020 and implementation of new technological solutions. A significant number of antennas
(apart from antennas used for mobile devices) will be to be installed inside buildings,
Publisher’s Note: MDPI stays neu- especially public utility buildings, including stadiums, railway stations, and shopping
tral with regard to jurisdictional claims centers. It should be noted, at this point, that antennas installed in locations close to crowds
in published maps and institutional would be smaller than those used in current macrocell transmitters. This is a fundamental
affiliations. difference and a common misunderstanding in public discussion. In a traditional antenna
system, the power is radiated according to the established spatial characteristics. Therefore,
the area in which users can be located, is predefined. In contrast, the power in a 5G antenna
Copyright: © 2020 by the authors. Li-
is radiated directionally, and focused on individual users or groups of users. Antenna
censee MDPI, Basel, Switzerland. This radiation directions can change almost automatically, to focus on mobile users [1–3].
article is an open access article distributed In this article, we describe the design process and a model of a microstrip antenna
under the terms and conditions of the designed in FEKO software by Altair. The antenna’s main assumption is its frequency of
Creative Commons Attribution (CC BY) operation in the 28 GHz band, i.e., in one of the frequency bands designed for operation of
license (https://creativecommons.org/ the 5G system [4]. Other important assumptions for the antenna model include a small
licenses/by/4.0/). size and the widest possible bandwidth. The antenna design process has been optimized

Electronics 2021, 10, 1. https://dx.doi.org/10.3390/electronics10010001 https://www.mdpi.com/journal/electronics

Electronics 2021, 10, 1 2 of 19

to minimize dimensions and weight, thereby enabling the use of the designed antenna in
mobile terminals, as well as facilitating integration with electronic devices.

2. Characteristics of the 5G System

The 5G network makes use of new technological solutions to meet the growing require-
ments of users. As a result, the new system will be able to handle an increasing number
of devices, and to satisfy higher quality thresholds required by modern applications. It is
an evolution of the 4G networks of today, which incorporates technologies capable of
handling the rapidly increasing amount of transmitted data and facilitating data exchange
between an ever-growing number of IoT devices. As is typical for the introduction of any
next-generation network, it is expected that until its coverage and functionality can match
or surpass existing 4G networks, the 5G network will need to coexist with such [1].
In addition to the existing usage scenarios of mobile networks, three additional
scenarios are planned for the emerging 5G network, all of which will be of particular
importance to users and will distinguish the 5G network from previous generations.
The first new usage scenario is an enhanced mobile broadband (eMBB), which enables
high-speed internet access (up to 1 Gbps) and will be the defining feature of this network
as compared with existing networks, especially in the initial phase of its implementation.
This advantage of the 5G system over legacy solutions will increase the efficiency and
quality of communications in society. As an example, this will enable services based on the
provision of high-resolution multimedia, attractive methods of communication (e.g., video,
augmented and virtual reality), as well as smart city services (e.g., transmission of content
from high-resolution cameras) [2,3].
The second use of 5G networks is based on massive machine type communications
(mMTC), where 5G will be able to support a very large number of connections from
low-power devices, referred to as the Internet of Things (IoT), to the mobile network.
These devices asynchronously exchange data, using the mobile network to communicate.
This scenario assumes support for a large number of device types, with the reservation
that these devices will use the mobile network in an occasional manner, exchanging small
volumes of data [5].
The third use is referred to as ultra-reliable low latency communications (URLLC),
which be a technology providing a minimum (1 ms) latency for data exchange over a
mobile network for critical applications (e.g., drone control). In previous generations of
mobile networks, latency values were longer and amounted to about 100 milliseconds for 3G
networks, with about 30 milliseconds in case of 4G (LTE—Long Term Evolution) network [6].
According to the current state of standardization of the 5G network, it is intended to
operate in three frequency bands, i.e., low, medium, and high. The use case of a particular
band depends on its characteristics, which include two factors in particular, i.e., radio signal
propagation and capacity of spectrum resources. The first factor is related to the physical
properties of electromagnetic waves and determines the obtainable radio transmission
range in changing weather conditions and radio signal coverage in hard-to-reach areas
(e.g., interiors of buildings). The second factor is associated with the available amount
of RF bands in a given frequency range, which can be used by 5G networks. It should
be noted that high bandwidths also require a wide radio band, which, being a limited
resource, is subject to rationing, and its use must also take into account RF communication
applications other than 5G networks, such as TV broadcasts, radio communication for
home automation equipment, etc.
In a 5G system, the following three frequency bands are assumed to be used first [1,2]:
• From 694 to 790 MHz (700 MHz band);
• From 3400 to 3800 MHz (3.6 GHz band);
• From 27.50 to 28.35 GHz (28 GHz band).
The 700 MHz band is characterized by good signal propagation and relatively low
attenuation (absorption of signal by various obstacles), which helps cover large areas.
Thanks to these characteristics, this band can be used for mMTC services. However, the
  from 27.50 to 28.35 GHz (28 GHz band).
The 700 MHz band is characterized by good signal propagation and relatively low
attenuation (absorption of signal by various obstacles), which helps cover large areas.
Electronics 2021, 10, 1 Thanks to these characteristics, this band can be used for mMTC services. However, 3 ofthe
19
700 MHz band alone would not be able to provide broadband internet access to mobile
users (eMBB), as it does not allow the use of MIMO, which would increase the capacity of
each cell site. However, it can be used together with the bands listed below, which have
700 MHz
large bandresources.
spectral alone would Thisnot be able
manner of to provideimproves
operation broadband internet
signal access to mobile
transmission quality
users (eMBB), as it does not allow the use of
from the user to the base station (i.e., “upstream”) [7].MIMO, which would increase the capacity of
each The
cell 3.6
site.GHz
However, it can
band does be used
allow the usetogether
of mMIMOwith the bandsMIMO),
(massive listed below,
while,which have
at the same
large spectral resources. This manner of operation improves signal transmission
time, it constitutes a compromise between propagation and capacity in terms of spectral quality
from the user
resources, to the base
especially when station (i.e., “upstream”)
combined with the 700 [7].
MHz band, which would improve up-
ward connectivity. This band would be used to build (massive
The 3.6 GHz band does allow the use of mMIMO a coverageMIMO), while,
layer for at the
eMBB same
services
time, it constitutes a compromise between propagation and capacity in
in several of the largest cities, including communication routes between these locations. terms of spectral
resources,
This bandespecially
can also be when
usedcombined with the
to introduce 700 MHz
services thatband, which
requires wouldtransmission
reliable improve upward and
particularly low latency (URLLC) in applications requiring the transmission ofinparticu-
connectivity. This band would be used to build a coverage layer for eMBB services several
of the largest cities, including communication routes between these locations. This band can
larly large amounts of data, such as high-resolution images for medical or navigation pur-
also be used to introduce services that requires reliable transmission and particularly low
poses [7].
latency (URLLC) in applications requiring the transmission of particularly large amounts of
The 28 GHz band is limited in its use, in particular, because of the requirements ap-
data, such as high-resolution images for medical or navigation purposes [7].
plied to transmission between the user and the base station (the “upstream” link). It can
The 28 GHz band is limited in its use, in particular, because of the requirements applied
be used, for example, for broadband access points and picocell applications
to transmission between the user and the base station (the “upstream” link). It can be used,
(cMTC/URLLC). This band, due to its large capacity and the possibility of allocating large
for example, for broadband access points and picocell applications (cMTC/URLLC). This band,
spectrum resources, may also be considered for the provision of internet access via a fixed
due to its large capacity and the possibility of allocating large spectrum resources, may also
wireless access service [1,7].
be considered for the provision of internet access via a fixed wireless access service [1,7].
3.
3. Frequency
Frequency Range Range forfor Local
Local Multipoint
Multipoint Distribution
Distribution Service
Service
The technological solutions used in the 5G
The technological solutions used in the 5G system eliminate system eliminate thethe disadvantages
disadvantages of of
LTE. These solutions make use of very high frequencies (in a 28
LTE. These solutions make use of very high frequencies (in a 28 GHz millimeter band) andGHz millimeter band)
and beamforming
beamforming techniques.
techniques. In several
In that, that, several
clientsclients
withinwithin the of
the range range of thebase
the same same base
station
station
can usecan a 1 use
Gbps a 1internet
Gbps internet connection.
connection.
The usefulness of the 28 GHz band is limited, in particular, because of requirements
applying
applying to totransmission
transmissionbetween
betweenthe theuser
userand
andthethe
base station
base (the(the
station “upstream”
“upstream”link).link).
The
28 GHz frequency band has been made available by the Federal
The 28 GHz frequency band has been made available by the Federal Communications Communications Com-
mission
Commission (FCC)(FCC)by reallocating the local
by reallocating multipoint
the local blockblock
multipoint distribution service
distribution (LMDS)
service (LMDS)A1,
which
A1, whichis 850is MHz
850 MHz widewide
and and
lies lies
between 27.50
between andand
27.50 28.35 GHzGHz
28.35 bands [8–10].
bands [8–10].
In response
response to tothe
thegrowing
growingdemanddemandfor anan
for additional
additional frequency
frequency spectrum,
spectrum,mostly
mostlyfor
fixed/mobile
for fixed/mobile 5G applications that require
5G applications significant
that require bandwidth
significant to support
bandwidth higher speeds
to support higher
>1.0 Gbps
speeds >1.0at Gbps
a lower at latency, the Federal
a lower latency, the Communications
Federal Communications Commission (FCC) has
Commission ordered
(FCC) has
ordered
the the A1 channel
A1 channel and reallocated
and reallocated it as partitofasthe
partnewof the newMicrowave
Upper Upper MicrowaveFlexibleFlexible
Use Li-
Use License
cense (UMFUS). (UMFUS). The ordered
The ordered part ofpart of the frequency
the frequency spectrum
spectrum has beenhastransformed
been transformed
into a
into alicensing
new new licensing
systemsystem
basedbased
on two on425twoMHz
425 MHz
wide wide
blocksblocks
(blocks(blocks
L1 andL1 L2)
and[10–13].
L2) [10–13].
The
The division
division of theof LMDS
the LMDS spectrum
spectrum in thein the 28 GHz
28 GHz frequency
frequency bandband is presented
is presented in Figure
in Figure 1. 1.

Figure 1.
Figure 1. The
The 28
28 GHz
GHz (27.50–28.35 GHz) local
(27.50–28.35 GHz) local multipoint
multipoint distribution
distribution service
service (LMDS)
(LMDS) spectrum.
spectrum.

The 28 GHz band and other millimeter frequency bands (mmWave), such as 24 GHz
37/39GHz
and 37/39 GHzbands,
bands,will
willplay
playa akey
keyrole
role
inin
5G5G implementations
implementations under
under thethe
newnew Up-
Upper
per Microwave
Microwave Flexible
Flexible Use License
Use License (UMFUS).
(UMFUS). The UMFUS
The UMFUS bandsbands are standardized
are standardized for
for 3GPP
3GPP
in in accordance
accordance with
with 5G New5GRadio
New Radio (NR) guidelines,
(NR) guidelines, in particular
in particular withinwithin Frequency
Frequency Band
2Band 2 (FR2),
(FR2), whichwhich includes
includes millimeter
millimeter frequencies.
frequencies.
The UMFUS bands are currently maintained and used by national mobile phone opera-
tors, innovative mobile phone companies, regional phone operators, and other organizations
wishing to implement mobile networks or provide fixed wireless access (FWA) services [8–10].
The UMFUS bands are subject to rules and regulations, which impose power limits
at fixed and base stations operating in conjunction with mobile systems (EIRP density
Electronics 2021, 10, 1 4 of 19

limit + 75 dBm/100 MHz). The average power of the sum of all antenna components in a
mobile base station must not exceed a maximum EIRP value of +43 dBm. Network deploy-
ments may use any desired duplex scheme, provided that other technical or operational
requirements are met [13].
The 28 GHz band is ideal for a wide range of applications making use of speeds
>1 Gb/s (depending on channel size), which are achievable using the spectrum designated
by UMFUS. High-speed backhaul, “at home” 5G/FWA, and other applications make use
of the ability to implement fixed PTP (point-to-point) and PTMP (point-to-multi-point)
configurations. The 28 GHz band can also be used for mobile applications, which are
currently a key pillar of 5G deployment championed by national carriers and other organi-
zations [11–13].

4. Analysis of Current Antenna Solutions Operating in 5G Systems at a 28 GHz Frequency

Current literature contains many proposed solutions of microstrip antennas operating
in 5G systems at a frequency of 28 GHz [14–23]. These solutions are characterized by
their compactness, small geometric dimensions, and antenna bandwidth of about 2 GHz.
Many of them are based on a RT Duroid 5880 laminate with a permittivity of 2.2. In the
analyzed antennas, the radiator is mostly rectangular, and is powered by a microstrip feed
line. The proposed antenna is different, due to the modifications made to the shape of
the patch, which were made on the basis of solutions originally used in fractal antennas.
These modifications can be used to create a proposed antenna capable of supporting
different bandwidths.
The antenna, proposed in this article, is capable of operating in systems intended for
5G communication, is designed for a resonance frequency equal to 28 GHz, and is based on
a RT Duroid 5880 laminate with a rectangular radiator. The proposed antenna is designed
to achieve a greater bandwidth than the antennas analyzed in the literature mentioned
above. Antennas with a wider operating band can only be created by using complex spatial
radiating structures, which makes them difficult to implement in 5G mobile devices.
The developed antenna, despite the use of the well-known inset feed in its design,
has achieved a much wider bandwidth than similar constructions of this type found
in the literature [14–23]. This effect was achieved through optimization with a large
number of variables, which was aimed at achieving the lowest possible reflectance and
voltage standing wave ratio (VSWR) in the widest possible frequency range. The results
significantly exceed the previously published parameter values for this type of antenna
in terms of bandwidth, especially in cases of strict matching conditions (VSWR ≤ 2,
VSWR ≤ 1.5, and VSWR ≤ 1.25).

5. Broadband Microstrip Antenna Designed for Use in 5G Systems

The design process of a microstrip antenna consists of multiple stages. The main steps
of the project process are presented below. The procedure for designing a single-piece
rectangular microstrip antenna are the following:
• Determine operational frequency;
• Determine operational bandwidth;
• Choose a substrate;
• Choose substrate height;
• Determine the dimensions of the patch;
• Determine the power supply;
• Determine the electrical parameters and characteristics of the antenna;
• Optimize the antenna to obtain the best possible parameters in the given frequency range.
Before the development of a proper numerical model of the designed antenna could
begin, it was necessary to perform preliminary calculations of its geometric parame-
ters. These calculations were done to work out an approximate shape of the antenna,
which would ensure that the structure complied with the assumptions adopted for its
design. The dimensions of the individual edges of the antenna depend primarily on the
Electronics 2021, 10, 1 5 of 19

resonance frequency, fr , and the relative permittivity, εr , of the dielectric layer of the copper
laminate, which are the foundation of the new antenna [24].
The antenna is designed to operate in a 5G system, on frequencies ranging from
27.50 GHz to 28.35 GHz (LMDS band) and the center frequency fr = 28.00 GHz. Choosing
the thickness of the substrate is one of the most important stages in the antenna design
process, as the thickness of the substrate directly affects the efficiency and bandwidth of
the microstrip antenna. One of the assumptions for this antenna is to obtain the widest
bandwidth possible. As the thickness of the substrate increases, the antenna’s bandwidth
increases, while its efficiency decreases. The upper value of the substrate thickness was
determined from the following relationship [14,15]:

0.3c 0.3 × 3 × 108 90 × 107

h≤ √ ≤ √ ≤ ≤ 3.45 mm. (1)
2π f ε r 2π × 28 × 109 2.2 260.8 × 109

Taking into account the result provided above, the RT Duroid 5880 laminate was selected
as a substrate, with a thickness of h = 1.57 mm, permittivity εr = 2.2, and tanδ = 9.0 × 10−4 .
In the next calculation step, the width of the radiating element should be determined from
the following relationship [16,17]:
r
3 × 108
r
c 2 2
W= = = 4.17 mm. (2)
2 fr εr + 1 2 × 28 × 109 2.2 + 1

To determine the length of the radiating element, the effective permittivity εreff , of the
substrate must first be calculated. It is defined by the following relationship [18,19]:

εr + 1 εr − 1 2.2 + 1 2.2 − 1
ε re f f = + q = + q = 1.6 + 0.21 = 1.81. (3)
2 h
2 1 + 12 W 2 2 1 + 12 × 1.57
4.17

Upon calculating εreff , the effective length of the patch Le should be determined from
the following relationship [20,21]:

c 3 × 108
Le = √ = √ = 3.98 mm. (4)
2 f r ε re f f 2 × 28 × 109 1.81

Then, we calculate by how much the patch should be shortened using the following
relationship [22,25]:

    
0.412· h ε re f f + 0.3 W h + 0.264 0.412 × 1.57 ( 1.81 + 0.3 ) 4.17
1.57 + 0.264
∆L =    =   = 0.74 mm.
ε re f f − 0.258 W h + 0.8
4.17
(1.81 − 0.258) 1.57 + 0.8
(5)
The final patch length value is:

L = Le − 2∆L = 3.98 − 2 × 0.74 = 2.58 mm. (6)

Once the size of the radiating element is determined, the size of the reference plane
also needs to be determined. The size of the reference plane is assumed to be greater
than the radiating element by approximately six thicknesses of the substrate, by both
width and length. The dimensions of the reference plane are determined by the following
relationship [26,27]:
Ls = L + 6h = 2.58 + 6 × 1.57 = 12 mm. (7)
Ws = W + 6h = 4.17 + 6 × 1.57 = 13.59 mm. (8)
In the next step, we determine the inset feed gap from the following relation [21]:
Electronics 2021, 10, 1 6 of 19

4.65 × 10−18 ·c· f r 4.65 × 10−18 × 3 × 105 × 28 × 109

f gap = p = √ = 0.21 mm. (9)
2ε re f f 2 × 1.81

The last step of the numerical antenna design process is to determine the size of the
feed line. By calculating the dimensions of a microstrip feed line with a characteristic
impedance ZC = 50 Ω, we start the determination of auxiliary variables a and b as:
q  
Zc ε r +1 ε r −1 0.11
a= 60 2 + ε r +1 0.23 + εr = 1.158
2
(10)
60π
b= √
Zc ε r
= 7.97

Since parameter a is less than 1.52, the width and length of the feed line is determined
from the following relationships [21,23]:
n h io
r −1
W f = π2 b − 1 − ln(2b − 1) + ε2ε ln(b − 1) + 0.39 − 0.61 ·h
r εr (11)
= 3.07 mm.

L f = 3·h = 3 × 1.57 = 4.71 mm. (12)

At this stage, the process of developing a numerical model of the designed antenna is
complete. The calculations above resulted in the antenna dimensions presented in Table 1.
The design of the developed antenna, as calculated, is shown in Figure 2. These data are be
used to optimize the antenna in terms of bandwidth and miniaturize its dimensions.

Table 1. Dimensions of the preliminary antenna design.

Antenna Component Symbol Dimensions (mm)

Ground plane width W S = We 13.59
Ground plane length L S = Le 12.00
Patch width Wp 4.17
Patch length Lp 2.58
Copper thickness Cu 0.05
Substrate thickness h 1.57
Permittivity Er 2.20
Electronics 2020, 9, x FOR PEER REVIEW 7 of 20
Feed line width Wf 3.07
Feed line length Lf 4.71

Figure 2.
Figure Antenna model
2. Antenna model view,
view, front
front and
and back
back side.
side.

6. Optimization
6. Optimization Process
Process and
and Discussion
Discussion of of Simulation
Simulation Results
Results
Computer simulations can help to determine the electrical parameters and radial
Computer simulations can help to determine the electrical parameters and radial
characteristics of the antenna; in this case, we used FEKO software developed by the Altair
characteristics of the antenna; in this case, we used FEKO software developed by the Altair
company.
company.
The analysis of electrical parameters of the radiating element design solution and
The analysis of electrical parameters of the radiating element design solution and
other antenna elements indicates that there is room to improve the electrical parameters of
other antenna elements indicates that there is room to improve the electrical parameters
of the antenna by reducing VSWR, increasing the width of the operating band, miniatur-
izing antenna dimensions, and increasing antenna gain. For this purpose, the design was
optimized for the parameters mentioned above, using FEKO software from the Altair
company. The calculated parameters of the rectangular patch were used to create a pre-
6. Optimization Process and Discussion of Simulation Results
Computer simulations can help to determine the electrical parameters and radial
characteristics of the antenna; in this case, we used FEKO software developed by the Altair
company.
The analysis
Electronics of1electrical parameters of the radiating element design solution and
2021, 10, 7 of 19
other antenna elements indicates that there is room to improve the electrical parameters
of the antenna by reducing VSWR, increasing the width of the operating band, miniatur-
izing antenna dimensions, and increasing antenna gain. For this purpose, the design was
the antenna by reducing VSWR, increasing the width of the operating band, miniaturizing
optimized for the parameters mentioned above, using FEKO software from the Altair
antenna dimensions, and increasing antenna gain. For this purpose, the design was opti-
company. The calculated parameters of the rectangular patch were used to create a pre-
mized for the parameters mentioned above, using FEKO software from the Altair company.
liminary simulation of the electrical parameters of the developed antenna model. Then,
The calculated parameters of the rectangular patch were used to create a preliminary
the model underwent a process of optimization, with the main assumption that the an-
simulation of the electrical parameters of the developed antenna model. Then, the model
tenna must operate at the resonance frequency of 28 GHz.
underwent a process of optimization, with the main assumption that the antenna must
The FEKO software supports the ability to define the electrical parameters optimiza-
operate at the resonance frequency of 28 GHz.
tion process for the designed antenna and provides a number of options related to opti-
The FEKO software supports the ability to define the electrical parameters optimization
mization. The algorithm of the optimization
process forprocess used by
the designed the FEKO
antenna and software
provides is shown of options related to optimization.
a number
in Figure 3. The algorithm of the optimization process used by the FEKO software is shown in Figure 3.

Figure 3. The algorithm of the optimization process in FEKO software.

Figure 3. The algorithm of the optimization process in FEKO software.
During the optimization process, one of the following optimization methods of FEKO
software must be chosen: simplex (Nelder–Mead), particle swarm optimization (PSO),
or genetic algorithm (GA). For the above methods, there are two options for stop criteria:

Specify maximum number of runs by the solver: The optimization process

ends when the FEKO solver has been run a certain number of times during
the optimization process. In the case of PSO and GA methods, if a full swarm
or generation is not generated within the allowed number of assigned runs,
optimization may be completed before the indicated number of runs by the solver.
If the optimization process is terminated prematurely, due to a reduction in the
number of runs by the solver, the software will provide the optimal solutions
found up to that point, as well as information about the optimization process.
Optimization convergence accuracy (standard deviation): This option allows
adjustment of accuracy levels required for the optimization process. Three op-
tions for selecting the accuracy level are available, i.e., high, normal, and low.
The selected accuracy level of the optimization process modifies the conditions in
which the search algorithm will converge, and the effect depends on the selected
method.
Electronics 2021, 10, 1 8 of 19

In the optimization process of the proposed antenna, the option “specify maximum
number of runs by the solver” was chosen. In the next step, we have to define the pa-
rameters of the antenna model that are to be changed during the optimization process in
order to achieve an optimal solution. The parameters define the variables, which can be
changed during the optimization process. These parameters are local for each optimization
search, and a correct search must contain at least one defined parameter. Any variable
defined in the FEKO software can be used as an optimization parameter, for example,
the physical dimensions of the model, load, and source (amplitude and phase), provided
that no relationship is implied between the optimization parameters in the same search.
During the optimization process, the length and width of the antenna base with the
screen (Ws , We , Ls , and Le ), the length and width of the radiator (Lp and Wp ), the length
and width of the power line (Lf and Wf ) and the length and width of the patch inset (Y0 and
X0 ) were changed. Other structural elements, such as thickness and permittivity of the
substrate, were not changed [28–31].
In the next step, we define the range of parameters to be changed, specifying their
minimum, maximum, and initial values. For each optimization parameter, a minimum value
and a maximum value must be defined, and optionally, an initial value can be given as
well. The initial value will influence the optimization process when using particle swarm
optimization or genetic algorithm. If the starting value is not specified by the user, the value in
the middle of the range of the parameter will be taken as the starting point of the optimization.
In our optimization process, we selected the minimum, maximum, and start values based on
calculated parameters of the rectangular patch and the center frequency.
In the next step, we define an optimization mask. An optimization mask is a set of
user-defined values forming a continuous line. The mask is a criterion to which the optimal
solution must adhere. It is specified that the optimized solution is smaller, equal, or larger
than the mask. When calculating an optimal solution, the target values are compared with
the mask. If the mask criterion is met, the values are added to the value table. We do not
define an optimization mask in our optimization process.
The last step of the optimization process is the selection of the optimization target.
The objectives, which need to be defined, specify the desired state that the optimization
process should try to achieve by changing certain parameters of the antenna model.
For each optimization search, goals should be defined to determine the desired state
that the optimization process should aim to achieve. Many goals can be defined, but a
correct optimization search must include at least one goal. During the optimization process
we define the following goals:
• Impedance goal (input impedance, input admittance, reflection coefficient (S11 ), trans-
mission coefficient, VSWR, return losses, current);
• Near-field goal (E field – electric field, H field – magnetic field, directional component,
coordinate system);
• Far-field goal (E field, antenna gain, directivity, RCS – Radar Cross Section);
• S-parameter goal (coupling coefficient, reflection coefficient, transmission coefficient,
VSWR, return losses);
• SAR (Specific Absorption Rate) goal;
• Power goal (efficiency, active power, power loss);
• Transmission/reflection coefficients goal (reflection, transmission, co-polarization,
and cross-polarization);
• Receiving antenna power goal (efficiency, active power, and power loss).
In our optimization process, we chose an impedance target (reflectance and VSWR)
with a target minimization operator. The impedance target provides optimization related
to the impedance and admittance of any voltage or current source, which is solved within
the FEKO model. The reflectance ratio is calculated in relation to the indicated reference
impedance.
In the case of an impedance target, the reflection factor is calculated directly from the
observed input impedance. Therefore, this value is the active reflection factor, and may
Electronics 2021, 10, 1 9 of 19

differ from the S11 , which was calculated during the calculation of the S parameter in the
multiport model. The voltage standing waveform factor for the observed input impedance
is considered in relation to the indicated reference impedance.
Once the parameters of the optimization process have been defined, we can run
“OPTFEKO” to calculate the optimal solution for the specified parameters. Once the
optimal values for the model are obtained, the FEKO model with the optimal parameters is
created.
The optimization process resulted in a final design of the antenna model, as shown in
Electronics 2020, 9, x FOR PEER REVIEW 10 of 20
Figure 4. Dimensions of this design are shown in Table 2.

Figure4.4.The
Figure Theoptimized
optimizedantenna
antennamodel
modelview,
view,front
frontand
andback
backside
sidewith
withdimensions.
dimensions.
Table 2. Dimensions of the optimized antenna model.
The resulting antenna design was simulated with the use of FEKO software from the
Altair company,
Antennawhich provided results forSymbol
Component the following electrical parameters:
Dimensions (mm) reflection
coefficient, standing wave
Ground plane widthratio, input impedance,
W S = We antenna gain, current
8.40 distribution in
the antenna, and radiation characteristics.
Ground plane length L S = Le 6.20
Patch width Wp 3.66
6.1. Q Factor Patch length Lp 2.14
Copper thickness Cu 0.05
The QSubstrate
factor ofthickness
an antenna is defined in the
h same way as for resonant 1.57 circuits, i.e., the
ratio of energy stored to energy lost during one
Permittivity Er period of vibration. 2.2 The difference is in
the user’s expectations, while for resonant circuits,
Feed line width Wf a high Q factor is usually
1.26 required, the
Feed for
opposite is true lineantennas.
length By determiningLthe
f 3.10to easily estimate
Q factor, it is possible
the bandwidth Inset
offeed gap
an antenna. The larger BWYis0 because of a reduction 0.50in the Q factor of
Width feed gap X0 0.68
the patch resonator, which is due to less energy being stored beneath the patch, and due
to higher radiation. In order to determine the Q factor value in the first step, we analyti-
callyThe resultingthe
determine antenna designofwas
bandwidth the simulated
antenna basedwith the use dimensions
on the of FEKO software
of the from the
proposed
Altair company,
antenna model which provided
(dimensions results
of the for theelement
radiating followingandelectrical parameters:
the thickness of the reflection
substrate),
coefficient, standing wave
using the following ratio, input impedance, antenna gain, current distribution in the
dependencies:
antenna, and radiation characteristics.
𝜀𝑟 − 1 ℎ 𝑊 2.2 − 1 1.57 3.66
𝐵𝑊 = 3.771 ∙ [ 2
] ∙ ∙ ( ) = 3.771 ∙ [ ]∙ ∙( )
6.1. Q Factor (𝜀𝑟 ) 𝜆0 𝐿 2.22 10.7 2.14 (13)
1.2
The Q factor of an antenna = is3.771 [
defined ] ∙ the
in 0,146728 ∙ 1.71028
same way as for=resonant
23% circuits, i.e.,
4.84
the ratio of energy stored to energy lost during one period of vibration. The difference is
in theThe theoretical
user’s bandwidth
expectations, while for forresonant
the antenna, calculated
circuits, a high Q from mathematical
factor relation-
is usually required,
ships,
the is about
opposite 23%.for
is true Theantennas.
BW of theBy microstrip
determining antenna
the Qis inversely proportional
factor, it is possible to to its Q
easily
factor and
estimate theisbandwidth
given by [32]
of the following:The larger BW is because of a reduction in the
an antenna.
Q factor of the patch resonator, which is due(𝑉𝑆𝑊𝑅 to less energy
− 1) being stored beneath the patch,
𝐵𝑊 =
and due to higher radiation. In order to determine the Q factor value in the first step, (14)
𝑄 𝑉𝑆𝑊𝑅
we analytically determine the bandwidth of the antenna based on the dimensions of the
Having determined the theoretical bandwidth BW and taking VSWR = 2 as the limit
value for the bandwidth, after transforming the above Relation (14) the Q factor of the
antenna was determined as:
(𝑉𝑆𝑊𝑅 − 1) (2 − 1) 1
𝑄= = = = 3.07. (15)
𝐵𝑊 𝑉𝑆𝑊𝑅 0.23 2 0.3252
Electronics 2021, 10, 1 10 of 19

proposed antenna model (dimensions of the radiating element and the thickness of the
substrate), using the following dependencies:
   
BW = 3.771· ε r −12 · λh0 · WL
(ε r )
h i h i (13)
= 3.771 × 2.2 −1 × 1.57 × 3.66 = 3.771 1.2 × 0.146728 × 1.71028 = 23%


2.22 10.7 2.14 4.84

The theoretical bandwidth for the antenna, calculated from mathematical relationships,
is about 23%. The BW of the microstrip antenna is inversely proportional to its Q factor
and is given by [32] the following:

(VSWR − 1)
BW = √ (14)
Q VSWR

Having determined the theoretical bandwidth BW and taking VSWR = 2 as the limit
value for the bandwidth, after transforming the above Relation (14) the Q factor of the
antenna was determined as:
(VSWR − 1) (2 − 1) 1
Q= √ = √ = = 3.07. (15)
BW VSWR 0.23 2 0.3252

The analytically low value of the antenna’s Q factor confirms its ability to operate in a
wide frequency range.

6.2. Reflection Coefficient

The base value of the reflection coefficient is assumed to be −10 dB, which means that
10%
Electronics 2020, 9, x FOR PEER REVIEW
of the incident power is reflected, i.e., 90% of the power is received by the antenna,
11 of 20
which is considered to be perfect for mobile communication [24,33]. The proposed antenna
haswhich
a resonance at 28.00
is considered to beGHz with
perfect fora mobile
reflection loss of −22.50
communication dB,The
[24,33]. as shown
proposedin an-
Figure 5.
Thetenna
S11 parameter was obtained
has a resonance by powering
at 28.00 GHz the antenna
with a reflection using dB,
loss of −22.50 the as
edge port.
shown in The
Figureantenna
5. The S parameter was obtained by powering the antenna using the edge
has an operating bandwidth of 5.57 GHz, which gives a relative operating bandwidth of
11 port. The an-
tenna The
19.89%. has an operating bandwidth
bandwidth determined offrom
5.57 GHz, which of
the results gives a relativesimulations
computer operating band-
is slightly
width of 19.89%. The bandwidth determined from the results of
smaller than the theoretical bandwidth determined on the basis of Relation computer simulations
(13),isdue to
theslightly smaller
inset feed usedthan
as athe theoretical
power bandwidth
supply. determined
Nevertheless, on the basis
this method of Relationpower
of supplying (13), to a
due to the inset feed used as a power supply. Nevertheless, this method of supplying
microstrip antenna provides us with better impedance matching (lower reflection factor
power to a microstrip antenna provides us with better impedance matching (lower reflec-
value) for a resonance frequency.
tion factor value) for a resonance frequency.

Figure 5. The reflection coefficient as a function of frequency for the proposed antenna working in the fifth generation
Figure 5. The reflection coefficient as a function of frequency for the proposed antenna working in the fifth generation (5G) system.
(5G) system.

6.3. Voltage Standing Wave Ratio

In the case of a patch antenna, the voltage standing wave ratio (VSWR) should not
be greater than 2 across the entire frequency band. Ideally, this value should equal 1
[24,34]. The voltage standing wave ratio as a function of frequency is shown in Figure 6.
As can be seen in Figure 6, the VSWR value obtained at a resonance frequency of 28.00
GHz equals 1.162, and the VSWR value of 2 was determined at 26.01 and 31.58 GHz re-
Electronics 2021, 10, 1 11 of 19

Figure 5. The reflection coefficient as a function of frequency for the proposed antenna working in the fifth generation
(5G) system.
6.3. Voltage Standing Wave Ratio
6.3.
InVoltage Standing
the case Wave antenna,
of a patch Ratio the voltage standing wave ratio (VSWR) should not be
greater In the 2case
than of a the
across patch antenna,
entire the voltage
frequency standing
band. Ideally, wave
thisratio
value(VSWR)
shouldshould
equalnot
1 [24,34].
Thebevoltage
greater standing
than 2 across
wavetheratio
entire
as frequency
a functionband. Ideally, this
of frequency value should
is shown equal6.1As can
in Figure
be [24,34].
seen inThe voltage
Figure 6, standing
the VSWR wave ratio obtained
value as a function
at aofresonance
frequency isfrequency
shown in Figure 6. GHz
of 28.00
As can be seen in Figure 6, the VSWR value obtained at a resonance frequency
equals 1.162, and the VSWR value of 2 was determined at 26.01 and 31.58 GHz respectively. of 28.00
GHz equals 1.162, and the VSWR value of 2 was determined at 26.01 and 31.58 GHz re-
The above values show that the proposed antenna operates in the whole assumed frequency
spectively. The above values show that the proposed antenna operates in the whole as-
band (LMDS), i.e., from 27.50 to 28.35 GHz.
sumed frequency band (LMDS), i.e., from 27.50 to 28.35 GHz.

Electronics 2020, 9, x FOR PEER REVIEW 12 of 20

Figure 6. The chart of the voltage standing wave ratio as a function of frequency for the proposed antenna working in the
5G system.
Figure 6. The chart of the voltage standing wave ratio as a function of frequency for the proposed antenna working in the
5G system.
6.4. Input Impedance
6.4.
TheInput Impedancedesign assumes that the impedance of the power line should be 50 Ω.
antenna’s
The of
In the case antenna’s design assumes
large discrepancies, it that the impedance
is possible to use aofmatching
the powersystem.
line should be 50 Ω.adding
However,
In thesystem
another case of large discrepancies,
introduces it is possible
additional to use
losses and a matching
generates system. However,
additional addingThe de-
costs [35–37].
another system introduces additional losses and generates additional costs
signed width and length of the feed line in the antenna has a complex input impedance [35–37]. The of
designed width and length of the feed line in the antenna has a complex input
Z = 56.1 + j5.08 Ω at the resonance frequency of 28.00 GHz, while at 27.50 GHz the complex impedance
of Z = 56.1 + j 5.08 Ω at the resonance frequency of 28.00 GHz, while at 27.50 GHz the
input impedance is Z = 51.57 − j10.73 Ω, and at 28.35 GHz the complex input impedance is
complex input impedance is Z = 51.57 − j 10.73 Ω, and at 28.35 GHz the complex input
59.6 − j1.39
Z =impedance Ω. Detailed input impedance values for the proposed antenna as a function
is Z = 59.6 − j 1.39 Ω. Detailed input impedance values for the proposed an-
of frequency are shown
tenna as a function in Figureare
of frequency 7. shown in Figure 7.

Figure 7. The chart of the complex input impedance as a function of frequency for the proposed antenna working in the
Figure 7. The chart of the complex input impedance as a function of frequency for the proposed antenna working in the 5G
5G system (real part, blue line and imaginary part, green line).
system (real part, blue line and imaginary part, green line).
6.5. Antenna Gain
Usually, antenna gain is given in relation to an isotropic antenna and is expressed in
dBi. In some cases, antenna gain is also given in relation to the dipole antenna and ex-
pressed in dBd. The energy gain of an antenna is dependent on its directivity and the
energy loss of the antenna, which results from the material it is made of [35,37]. The pro-
Electronics 2021, 10, 1 12 of 19

6.5. Antenna Gain

Usually, antenna gain is given in relation to an isotropic antenna and is expressed
in dBi. In some cases, antenna gain is also given in relation to the dipole antenna and
expressed in dBd. The energy gain of an antenna is dependent on its directivity and
the energy loss of the antenna, which results from the material it is made of [35,37].
The proposed antenna has a gain of 5.06 dBi at a resonance frequency of 28.00 GHz,
which is considered to be high for compact microstrip antennas. Values of antenna gain as
a function of frequency are shown in Figure 8.
Electronics 2020, 9, x FOR PEER REVIEW 13 of 20

Electronics 2020, 9, x FOR PEER REVIEW 13 of 20

Figure 8. The chart of antenna gain as a function of frequency for the proposed antenna working in the 5G system.
Figure 8. The chart of antenna gain as a function of frequency for the proposed antenna working in the 5G system.
Figure 8. The chart of Current
6.6. antenna gain as a function
Distribution of frequency
in the Antenna for the proposed antenna working in the 5G system.
6.6. Current Distribution in the Antenna
In In a microstrip antenna, the value of the current should be minimal at the end of the
a6.6.
microstrip antenna,inthe
Current Distribution value of the current should be minimal at the end of
the Antenna
radiating element (the edge of the patch). The voltage at the edge of the patch is phase
the radiatingInelement (the
a microstrip edge the
antenna, of the
valuepatch). The voltage
of the current should beatminimal
the edgeat theofend
theofpatch
the is
shifted in relation to the current. Therefore, peak voltage is present at the end of the patch,
phase radiating
shifted in element
relation (the
to edge
the of the patch).
current. The voltage
Therefore, at the
peak edge ofisthe
voltage patch isat
present phase
the end
with current values close to zero [33,37]. A similar situation occurs in the middle of the
of the shiftedwith
patch, in relation to the
current current.
values Therefore, peak voltageAis present atsituation
the end of occurs
the patch,
wave, further down the line, i.e., atclose to zero
the beginning [33,37].
of the patch. similar
The phase-shifted voltage in the
with current values close to zero [33,37]. A similar situation occurs in the middle of the
middle
in relation to the current phase produces fields on the edges of a microstrip antenna. The
of the wave, further down the line, i.e., at the beginning of the patch. The phase-
wave, further down the line, i.e., at the beginning of the patch. The phase-shifted voltage
shifted voltage
current in relation
distribution of thetodeveloped
the current phaseatproduces
antenna frequenciesfields on the
of 27.51 GHz edges of aGHz
(a), 28.0
in relation to the current phase produces fields on the edges of a microstrip antenna. The
microstrip
(b), and
antenna. 28.35
The GHz (c)
current is presentedofinthe
distribution Figure 9.
developed antenna at frequencies of 27.51 GHz (a),
current distribution of the developed antenna at frequencies of 27.51 GHz (a), 28.0 GHz
28.0 GHz(b), (b), and
and 28.35
28.35 GHzGHz(c) is (c) is presented
presented in Figure in9.Figure 9.

(a)
(a)

Figure 9. Cont.
Electronics 2021, 10, 1 13 of 19

Electronics 2020, 9, x FOR PEER REVIEW 14 of 20

(b)

(c)

Figure 9. Surface current distribution for the proposed 5G antenna at different frequencies. (a) 27.51 GHz; (b) 28.0 GHz; (c) 28.35 GHz.

Figure 9. Surface6.7. Radiation

current Characteristics
distribution for the proposed 5G antenna at different frequencies. (a) 27.51 GHz; (b) 28.0 GHz;
(c) 28.35 GHz.
The radiation characteristics show how the antenna radiates energy depending
on direction. It represents
6.7. Radiation a standardized distribution of the electric field, or the rela-
Characteristics
tive distribution of surface power density.
The radiation characteristics show how The
the characteristics
antenna radiates are determined
energy depending on in two
planes, i.e., horizontal
direction. and vertical,
It represents and can
a standardized also be presented
distribution infield,
of the electric a three-dimensional
or the relative dis- (3D)
format [33,34].
tributionThe designed
of surface powerantenna should
density. The have directional
characteristics radiation
are determined in two characteristics.
planes, i.e.,
horizontal andchart
A three-dimensional vertical, andproposed
of the can also be presented
antenna’s in a three-dimensional
radiation characteristics(3D)for format
27.51 GHz
[33,34].
(a), 28.0 GHz The
(b), anddesigned antenna
28.35 GHz (c)should have directional
frequencies is shownradiation
in Figurecharacteristics.
10. Figure 11 A shows
three- the
dimensional chart of the proposed antenna’s radiation characteristics for 27.51 GHz (a),
standardized radiation characteristics of the proposed antenna for 27.51 GHz (blue line),
28.0 GHz (b), and 28.35 GHz (c) frequencies is shown in Figure 10. Figure 11 shows the
28.0 GHzstandardized
(green line), and 28.35
radiation GHz (redofline)
characteristics frequencies
the proposed in the
antenna for polar coordinate
27.51 GHz system
(blue line),
for vertical polarization. Figure 12 shows the standardized radiation characteristics of the
proposed antenna for 27.51 GHz (blue line), 28.0 GHz (green line), and 28.35 GHz (red line)
frequencies in the polar coordinate system for horizontal polarization. The presented
drawings show that the beam width at the –3 dB level for vertical polarization is 115.04◦
with the back lobe being −6 dB, while the beam width at the –3dB level for horizontal
polarization is 72.16◦ with the back lobe being −10 dB.
 28.0 GHz (green line), and 28.35 GHz (red line) frequencies in the polar coordinate system
for vertical polarization. Figure 12 shows the standardized radiation characteristics of the
proposed antenna for 27.51 GHz (blue line), 28.0 GHz (green line), and 28.35 GHz (red
line) frequencies in the polar coordinate system for horizontal polarization. The presented
Electronics 2021, 10, 1 drawings show that the beam width at the –3 dB level for vertical polarization is 14115.04°
of 19
with the back lobe being −6 dB, while the beam width at the –3dB level for horizontal
polarization is 72.16° with the back lobe being −10 dB.

(a)

(b)

(c)

Figure 10. The three-dimensional (3D) view of radiation pattern for the proposed 5G antenna model at different frequencies.
(a) 27.51 GHz; (b) 28.0 GHz; (c) 28.35 GHz.
 Electronics 2020, 9, x FOR PEER REVIEW 16 of 20
Electronics 2020, 9, x FOR PEER REVIEW 16 of 20
Figure
Electronics 10.10,The
2021, 1 three-dimensional (3D) view of radiation pattern for the proposed 5G antenna model at different frequen-
15 of 19
Figure 10.27.51
cies. (a) The three-dimensional
GHz; (b) 28.0 GHz;(3D) viewGHz.
(c) 28.35 of radiation pattern for the proposed 5G antenna model at different frequen-
cies. (a) 27.51 GHz; (b) 28.0 GHz; (c) 28.35 GHz.

Figure 11.
Figure The normalized
11. The normalizedradiation
radiationpatterns
patternsfor
forthe
theproposed
proposedantenna
antennamodel
modeloperating
operatingininthe
the5G
5G
system
Figure for
11. 27.51
The GHz
normalized(blue line),
radiation28.0 GHz
patterns(green
for line),
the and
proposed 28.35 GHz
antenna (red
modelline) in polar
operating coordinates
in
system for 27.51 GHz (blue line), 28.0 GHz (green line), and 28.35 GHz (red line) in polar coordi- the 5G
for vertical
system
nates for polarization.
for vertical
27.51 GHz (blue line), 28.0 GHz (green line), and 28.35 GHz (red line) in polar coordi-
polarization.
nates for vertical polarization.

Figure 12.
Figure The normalized
12. The normalizedradiation
radiationpatterns
patternsfor
forthe
theproposed
proposedantenna
antennamodel
modeloperating
operatingininthe
the5G
5G
system
Figure for
12. 27.51
The GHz (blue
normalized line),
radiation28.0 GHz
patterns(green
for line),
the and
proposed 28.35 GHz
antenna (red
modelline) in polar
operating coordinates
in
system for 27.51 GHz (blue line), 28.0 GHz (green line), and 28.35 GHz (red line) in polar coordi-the 5G
system for horizontal
27.51polarization.
for horizontal
nates for GHz polarization.
(blue line), 28.0 GHz (green line), and 28.35 GHz (red line) in polar coordi-
nates for horizontal polarization.
 7. Comparison of the Proposed Antenna with Other Antennas
The obtained parameter values for the proposed microstrip antenna can be compared
in terms of impedance matching and appropriate bandwidth, with other published results
for the purposes of comparative evaluation. For example, the frequency response of the
Electronics 2021, 10, 1 measured S11 parameter of the proposed antenna may not be the lowest as compared16with of 19
the S11 values obtained for the antennas presented in [18,20,32], but it is relatively low.
Figure 13 shows a comparison of the reflection factor as a function of frequency for
the proposed antenna, and the antennas developed in other works. The corresponding
7. Comparison
frequency of theusing
responses Proposed Antenna
the VSWR with Other
parameter Antennaswith each other, as shown
are compared
The obtained
in Figure 14. parameter values for the proposed microstrip antenna can be compared
in terms
The of impedance
minimum matching
reflection lossand appropriate
value obtainedbandwidth,
in this study with other−22.51
is about published results
dB (VSWR
for the purposes of comparative evaluation. For example, the frequency response
≈ 1.16), while in [20] it is about −26.00 dB (VSWR ≈ 1.10), in [18] it is about −21.00 dB (VSWR of the
measured S
≈ 1.20), and in parameter of the proposed antenna may not be the lowest as compared
11 [32] it is about −18.00 dB (VSWR ≈ 1.28). Moreover, for the antenna proposed with
the S
in this values obtained for the antennas presented in [18,20,32], but it is relatively
11work, the operating band of the antenna (BW) is the largest, amounting to S11 ≤ −10 low.
Figure≤132)shows
dB (VSWR 5.57 GHz,a comparison of the reflection
and is centered exactly at factor
28 GHz,as awhile
function of frequency
in [20] the operatingfor
the proposed
band is 2. 63 GHzantenna,
and isandalsothe antennas
centered developed
exactly in other
at 28 GHz, in [18]works. The corresponding
the operating band is 2.85
frequency responses using the VSWR parameter are compared with
GHz and is also centered at about 28.1 GHz (shift of about 100 MHz), and in [32] each other, as the
shown
op-
in Figure 14.
erating band is 1.07 GHz and is also centered exactly at 28 GHz.

Figure
Figure 13.
13. The
The reflection
reflection coefficient
coefficient as
as aa function
function of
of frequency for the
frequency for the proposed
proposed antenna
antenna and
and antennas
antennas from
from other
other works.
works.
Electronics 2020, 9, x FOR PEER REVIEW 18 of 20

Figure
Figure 14.
14. The
The voltage standing wave
voltage standing waveratio
ratio(VSWR)
(VSWR)asasaafunction
functionofoffrequency
frequencyfor
for the
the proposed
proposed antenna,
antenna, and
and antennas
antennas from
from other works.
other works.

A comparative summary of the proposed antenna with the antennas described in

[18,20,32] is presented in Table 3. It shows that the proposed microstrip antenna in this
work has the best impedance matching bandwidth performance in all cases, especially for
stringent matching conditions (VSWR ≤2, VSWR ≤1.5, and VSWR ≤1.25).

Table 3. Comparisons among the bandwidth achieved in the present work to those achieved in
Electronics 2021, 10, 1 17 of 19

The minimum reflection loss value obtained in this study is about −22.51 dB (VSWR ≈
1.16), while in [20] it is about −26.00 dB (VSWR ≈ 1.10), in [18] it is about −21.00 dB
(VSWR ≈ 1.20), and in [32] it is about −18.00 dB (VSWR ≈ 1.28). Moreover, for the antenna
proposed in this work, the operating band of the antenna (BW) is the largest, amounting to
S11 ≤ −10 dB (VSWR ≤ 2) 5.57 GHz, and is centered exactly at 28 GHz, while in [20] the
operating band is 2. 63 GHz and is also centered exactly at 28 GHz, in [18] the operating
band is 2.85 GHz and is also centered at about 28.1 GHz (shift of about 100 MHz), and in [32]
the operating band is 1.07 GHz and is also centered exactly at 28 GHz.
A comparative summary of the proposed antenna with the antennas described
in [18,20,32] is presented in Table 3. It shows that the proposed microstrip antenna in
this work has the best impedance matching bandwidth performance in all cases, especially
for stringent matching conditions (VSWR ≤2, VSWR ≤1.5, and VSWR ≤1.25).

Table 3. Comparisons among the bandwidth achieved in the present work to those achieved in other published works.

Performance Measure Present Work (Proposed Antenna) Work of [18] Work of [20] Work of [32]
Center frequency 28.00 GHz 28.10 GHz 28.00 GHz 28.00 GHz
VSWR ≤ 1.25 1.15 GHz 0.45 GHz 0.79 GHz NA
BW VSWR ≤ 1.5 2.68 GHz 1.74 GHz 1.57 GHz 0.55 GHz
VSWR ≤ 2 5.57 GHz 2.85 GHz 2.63 GHz 1.07 GHz
Relative BW 19.89% 10.14% 9.39% 3.83%
Gain 5.06 dBi 6.59 dBi 6.37 dBi 6.72 dBi

8. Conclusions
A rectangular microstrip antenna has been proposed for 5G applications in response
to the growing demand for mobile data and mobile devices. This antenna has a resonance
frequency of 28.00 GHz, for which the reflectance is equal to −22.50 dB. The proposed
antenna has a radiation efficiency of 80.18% and the antenna gain for the resonance fre-
quency is 5.06 dB. The results also show that its bandwidth is 5.57 GHz (relative operational
band 19.89%), which is a very good result, much greater than results achieved in other
published works, such as [15,16], where the operating bandwidth of the proposed antennas
is around 1 or 2 GHz [18–20]. The proposed antenna would be a good option for 5G mobile
communication, as it offers the high throughput required. The antenna is very compact
and lightweight, which makes it suitable for devices where space is a major limitation.

Author Contributions: Conceptualization, R.P.; Methodology, R.P. and M.B.; Resources, M.B. and L.N.;
Visualization, R.P. and M.B.; Writing—original draft, R.P., M.B. and L.N.; Writing—review & editing,
L.N. For other cases, all authors have participated. All authors have read and agreed to the published
version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Data sharing not applicable. No new data were created or analyzed in
this study. Data sharing is not applicable to this article.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Andrews, J.G.; Buzzi, S.; Choi, W.; Hanly, S.V.; Lozano, A.E.; Soong, A.C.K.; Zhang, J.C. What Will 5G Be? IEEE J. Sel.
Areas Commun. 2014, 32, 1065–1082. [CrossRef]
2. Roh, W.; Seol, J.-Y.; Park, J.; Lee, B.; Lee, J.; Kim, Y.; Cho, J.; Cheun, K.; Aryanfar, F. Millimeter-wave beamforming as an enabling
technology for 5G cellular communications: Theoretical feasibility and prototype results. IEEE Commun. Mag. 2014, 52, 106–113.
[CrossRef]
Electronics 2021, 10, 1 18 of 19

3. Rappaport, T.S.; Sun, S.; Mayzus, R.; Zhao, H.; Azar, Y.; Wang, K.; Wong, G.N.; Schulz, J.K.; Samimi, M.; Gutierrez, F. Millimeter
Wave Mobile Communications for 5G Cellular: It Will Work! IEEE Access 2013, 1, 335–349. [CrossRef]
4. Schoeman, M.; Marchand, R.; Van Tonder, J.; Jakobus, U.; Aguilar, A.; Longtin, K.; Vogel, M.; Alwajeeh, T. New Features in
Feko/WinProp 2019. In Proceedings of the 2020 International Applied Computational Electromagnetics Society Symposium
(ACES), Monterey, CA, USA, 27–31 July 2020; pp. 1–2.
5. Imai, T.; Kitao, K.; Tran, N.; Omaki, N.; Okumura, Y.; Sasaki, M.; Yamada, W. Development of high frequency band over 6 GHz for
5G mobile communication systems. In Proceedings of the 2015 9th European Conference on Antennas and Propagation (EuCAP),
Lisbon, Portugal, 13–17 April 2015; pp. 1–4.
6. Arya, A.K.; Kim, S.J.; Kim, S. A dual-band antenna for lte-r and 5g lower frequency operations. Prog. Electromagn. Res. Lett. 2020,
88, 113–119. [CrossRef]
7. Zou, Y.; Pan, J. Broadband and High-gain Antenna based on Novel Frequency Selective Surfaces for 5G Application. In Proceed-
ings of the 2019 IEEE 4th Advanced Information Technology, Electronic and Automation Control Conference (IAEAC), Chengdu,
China, 20–22 December 2019; Volume 1, pp. 2488–2491.
8. Taghian, F.; Abdipour, A.; Mohammadi, A.; Roodaki, P.M.; Mohammadi, A. Design and nonlinear analysis of a dual-band Doherty
power amplifier for ISM and LMDS applications. In Proceedings of the 2011 IEEE Applied Electromagnetics Conference (AEMC),
Kolkata, India, 18–22 December 2011; pp. 1–4.
9. Resende, U.C.; Moreira, F.J.S.; Bergmann, J.R. Analysis of omnidirectional antennas with radome operating in LMDS band for
signals of digital TV. In Proceedings of the 2009 SBMO/IEEE MTT-S International Microwave and Optoelectronics Conference
(IMOC), Belem, Brazil, 3–6 November 2009; pp. 83–86.
10. Son, L.T.; Ekman, T.; Orten, P. Performance of the multi-carrier LMDS system at the 28 GHz radio frequency band. In Pro-
ceedings of the 2006 First Mobile Computing and Wireless Communication International Conference, Amman, Jordan, 17–20
September 2006; pp. 150–155.
11. Lee, H.; Kim, T.; Lim, Y. Wide-band microstrip patch antenna design for LMDS band. In Proceedings of the 2000 5th International
Symposium on Antennas, Propagation, and EM Theory. ISAPE 2000 (IEEE Cat. No.00EX417), Beijing, China, 15–18 August 2000;
pp. 709–712.
12. Chin, K.-S.; Chang, H.-T.; Liu, J.-A. Design of LTCC Wideband Patch Antenna for LMDS Band Applications. IEEE Antennas Wirel.
Propag. Lett. 2010, 9, 1111–1114. [CrossRef]
13. Chen, K.-S.; Chu, C.-Y. A propagation study of the 28 GHz lmds system performance with m-qam modulations under rain fading.
Prog. Electromagn. Res. 2007, 68, 35–51. [CrossRef]
14. Sridevi, S.; Mahendran, K. Design of Millimeter Wave Microstrip Patch Antenna for MIMO Communication. Int. Res. J.
Eng. Technol. 2017, 4, 1513–1518.
15. Johari, S.; Jalil, M.A.; Ibrahim, S.I.; Mohammad, M.N.; Hassan, N. 28 GHz Microstrip Patch Antennas for Future 5G. J. Eng.
Sci. Res. 2018, 2, 1–6.
16. Mungur, D.; Duraikannan, S. Microstrip Patch Antenna at 28 GHz for 5G Applications. J. Sci. Technol. Eng. Manag. Adv. Res. Innov.
2018, 1, 5–7.
17. Ahmed, Z.; McEvoy, P.; Ammann, M.J. Comparison of Grid Array and Microstrip Patch Array Antennas at 28 GHz. In Proceedings
of the 2018 IEEE MTT-S International Microwave Workshop Series on 5G Hardware and System Technologies (IMWS-5G), Dublin,
Ireland, 30–31 August 2018; pp. 1–3. [CrossRef]
18. Teresa, P.M.; Umamaheswari, G. Compact Slotted Microstrip Antenna for 5G Applications Operating at 28 GHz. IETE J. Res.
2020, 1–8. [CrossRef]
19. Darboe, O.; Bernard, D.; Konditi, O.; Manene, F. A 28GHz rectangular microstrip patch antenna for 5G applications. Int. J. Appl.
Eng. Res. 2019, 12, 854–857.
20. Kaeib, A.F.; Shebani, N.M.; Zarek, A.R. Design and Analysis of a Slotted Microstrip Antenna for 5G Communication Networks at
28 GHz. In Proceedings of the 2019 19th International Conference on Sciences and Techniques of Automatic Control and Computer
Engineering (STA); Institute of Electrical and Electronics Engineers (IEEE), Sousse, Tunisia, 24–26 March 2019; pp. 648–653.
21. Kiran, T.; Mounisha, N.; Mythily, C.; Akhil, D.; Phani Kumar, T.V.B. Design of microstrip patch antenna for 5G applications.
IOSR J. Electron. Commun. Eng. (IOSR-JECE) 2018, 13, 14–17.
22. Khraisat, Y.S.H. Increasing Microstrip Patch Antenna Bandwidth by Inserting Ground Slots. J. Electromagn. Anal. Appl. 2018,
10, 1–11. [CrossRef]
23. Yang, S.J.; Su, H.F.; Wu, Y.; Zhang, X.Y. Dual-polarized Low-profile Wideband Filtering Antenna for 5G Applications. In Pro-
ceedings of the 2019 Photonics & Electromagnetics Research Symposium-Fall (PIERS-Fall); Institute of Electrical and Electronics
Engineers (IEEE), Xiamen, China, 17–20 December 2019; pp. 295–299.
24. Derneryd, A. A theoretical investigation of the rectangular microstrip antenna element. IRE Trans. Antennas Propag. 1978,
26, 532–535. [CrossRef]
25. Muhammad, S.; Shehu, Y.; Ya’u, I.; Abubakar, A.S. Design of Single Feed Dual-Band Millimeter Wave Antenna for Future 5G
Wireless Applications. Sci. World J. 2019, 14, 84–87.
26. Desai, A.; Bui, C.D.; Patel, J.; Upadhyaya, T.; Byun, G.; Khang, N.T. Compact Wideband Four Element Optically Transparent
MIMO Antenna for mm-wave 5G Applications. IEEE Access 2020, 8, 1. [CrossRef]
Electronics 2021, 10, 1 19 of 19

27. Arya, A.K.; Kim, S.J.; Park, S.; Kim, D.-H.; Hassan, R.S.; Ko, K.; Kim, S. shark-fin antenna for railway communications in lte-r, lte,
and lower 5G frequency bands. Prog. Electromagn. Res. 2020, 167, 83–94. [CrossRef]
28. Chayono, R.; Futter, P.; Castany, J.S. Characteristic mode analysis of smart phone antenna using HW FEKO. In Proceedings of the
2016 International Symposium on Antennas and Propagation (ISAP), Okinawa, Japan, 24–28 October 2016; pp. 230–231.
29. Jakobus, U.; Aguilar, A.; Attardo, E.; Schoeman, M.; Van Tonder, J.; Longtin, K. Review of selected new features in FEKO 2018.
In Proceedings of the 2018 International Applied Computational Electromagnetics Society Symposium (ACES), Institute of
Electrical and Electronics Engineers (IEEE), Denver, CO, USA, 25–29 March 2018; pp. 1–2.
30. Almajali, E.; McNamara, D.; Lee, D. Using electromagnetic simulation code FEKO as a numerical laboratory in antenna
engineering. In Proceedings of the 2010 14th International Symposium on Antenna Technology and Applied Electromagnetics &
the American Electromagnetics Conference, Ottawa, ON, Canada, 5–8 July 2010; pp. 1–4.
31. Pozar, D.M.; Schaubert, D.H. Microstrip Antennas: The Analysis and Design of Microstrip Antennas and Arrays; IEEE Press: New York,
NY, USA, 1995.
32. Goyal, R.K.; Modani, U.S. A Compact Microstrip Patch Antenna at 28 GHz for 5G wireless Applications. In Proceedings of the
2018 3rd International Conference and Workshops on Recent Advances and Innovations in Engineering (ICRAIE), Jaipur, India,
22–25 November 2018; pp. 1–2.
33. James, J.R.; Hall, P.S. Handbook of Microstrip Antenna; Institution of Engineering & Technology: London, UK, 1989.
34. Kara, M. Closed-form expressions for the resonant frequency of rectangular microstrip antenna elements with thick substrates.
Microw. Opt. Technol. Lett. 1996, 12, 131–136. [CrossRef]
35. Qing, X.; Chen, Z.N. Comments on “Comments on ‘Measuring the Impedance of Balanced Antennas by an S-Parameter Method’”.
IEEE Antennas Propag. Mag. 2010, 52, 171–172. [CrossRef]
36. Kang, J.-S.; Kim, J.-H.; Park, J.-I. Comparison of antenna parameters of R-/S-band standard gain antennas. In Proceedings of the
2015 International Workshop on Antenna Technology (iWAT), Seoul, South Korea, 4–6 March 2015; pp. 369–371.
37. De Silva, S.; Belostotski, L.; Okoniewski, M. Modeling and measuring of antenna array s-parameters and radiation efficiency.
In Proceedings of the 2017 IEEE International Symposium on Antennas and Propagation & USNC/URSI National Radio Science
Meeting, San Diego, CA, USA, 9–14 July 2017; pp. 2293–2294. [CrossRef]