See discussions, stats, and author profiles for this publication at: [Link]

net/publication/384470316

Design of a high-gain, low-ECC 2x2 MIMO antenna for 28 GHz 5G wireless

communication system

Article in Physica Scripta · October 2024

DOI: 10.1088/1402-4896/ad81bb

CITATIONS READS

0 48

2 authors:

Mohit Pant Leeladhar Malviya

Mahakal Group Of Institutes Shri G. S. Institute of Technology and Science, Indore (M.P.)-452003
11 PUBLICATIONS 155 CITATIONS 97 PUBLICATIONS 1,460 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Leeladhar Malviya on 18 October 2024.

The user has requested enhancement of the downloaded file.

 Phys. Scr. 99 (2024) 115520 [Link]

PAPER

Design of a high-gain, low-ECC 2x2 MIMO antenna for 28 GHz 5G

wireless communication system
RECEIVED
4 August 2024
REVISED
17 September 2024
ACCEPTED FOR PUBLICATION
Mohit Pant∗ and Leeladhar Malviya
30 September 2024 Shri G. S. Institute of Technology and Science, Department of ECE, Indore, 23, Sir M. Visvesvaraya Marg, Vallabh Nagar, Indore, M.P.
PUBLISHED 452003, India
∗
14 October 2024 Author to whom any correspondence should be addressed.
E-mail: ermohitpant@[Link] and ldmalviya@[Link]

Keywords: 5G, MIMO, Gain, ECC, Antenna, TARC

Abstract
In this paper, a 2 × 2 MIMO antenna with inset feed is designed for the frequency Range 2 (FR2)
millimeter wave frequencies such as n257 (26-29.5 GHz), n258 (24.25-27.5 GHz), and n261
(27.5-28.35 GHz). The MIMO antenna is designed and fabricated using low loss Rogers RT/duroid
5880 substrate having dimension of 1.4λ0×2.8λ0×0.073λ0, where λ0 is the free-space wavelength
operating at 28 [Link] defected ground structure (DGS) minimized the surface waves current. Thus
DGS are used to increased the gain and the isolation of the proposed MIMO antenna design. The gain
of the MIMO antenna reported is 10.3 dBi with 97.01% radiation efficiency at 28 GHz resonant
frequency. The -10 dB impedance bandwidth lies between 26.95-29.50 GHz, which covers the 5G
frequency bands. The isolation is 28 dB and envelope correlation coefficient (ECC) is 10 −5,
respectively, in the complete operating band.

1. Introduction

Fifth-generation mobile communication has seen revolutionary developments and growth in recent years. It can
fulfill the new demands that arise due to the exponential increase in users. There will be around 3 billion 5G
subscribers across the globe by the end of 2025 [1, 2]. There are new 5G applications generated every day, such as
drive-less cars, smart cities, industry automation, live gaming, tele-education, smart healthcare, and virtual
reality offices [3, 4]. These applications are possible due to the unique features of 5G, for example, the massively
high data rate (20 Gbps), low latency (10 ms), high mobility (upto 500 km/hr), high spectral efficiency (0.074-6.1
b/s/Hz), and high connection density, 2000 active users/km2 [5, 6]. The main pillar needed to fulfill such an
application is a high-bandwidth MIMO antenna. The sub-6 GHz 5G frequency band is highly crowded [7].
Thus, it is essential to move to the mmWave frequency band to manage the ultra high bandwidth requirements
[8]. The exceptionally short wavelengths of mmWave frequency signals allow for the fabrication of physically
compact antenna arrays or MIMO antenna arrays. The implementation of low-cost fabrication technology
results in significant cost savings for these antenna systems [9]. The mmWave antenna is used for the 5G
frequency bands such as 27.5-28.35 GHz, 24.25-24.45 GHz, 24.75-25.25 GHz, 27.5-28.35 GHz, 24.25-
27.5 GHz, 27.5-29.5 GHz, and 27.5-29.5 GHz, which will be used in different 5G wireless communication
systems such as vehicle to-everything (V2X) communication, streaming of 4K and 8K video, 3D video in high
mobility environments, massive IoT, process automation, industrial automation, remote surgery, and many
more [10]. The limitations of 5G frequencies are high path loss and low penetration power, due to which the
transmitted signal strength becomes weaker. To overcome these limitations, a high-gain and high-isolation
multiple input, multiple output (MIMO) antenna array can be employed in the transceiver. The main challenges
in the available literature to improve the performance of 5G MIMO antennas include boosting gain, improving
radiating efficiency, and minimizing mutual coupling between elements of the antenna array [11]. The
performance improvement methods include defective ground structures (DGS), decoupling structure,
frequency-selective surfaces (FSS), and artificial magnetic conductors (AMC) [12] However, DGS allows for

© 2024 IOP Publishing Ltd. All rights, including for text and data mining, AI training, and similar technologies, are reserved.
Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

effective mutual coupling reduction without requiring a significant increase in the physical separation between
antenna elements. This is particularly advantageous in mmWave applications where space is limited, such as in
mobile devices or compact wireless modules [13].
In antenna arrays and multi-element antennas, DGS is a crucial technique for minimizing mutual coupling
and increasing the gain, especially in microstrip and PCB-based designs [14, 15]. The reliability of DGS depends
on meticulous design and optimization to meet the demands of the particular antenna system and operating
frequency band. The specific dimension slots or patterns, depending on the resonant frequencies, are etched
from the ground plane to design the DGS. These can have any shape that meets the requirements, including slits,
slots, meander lines, and other geometries. This modification reduces surface waves and enhances antenna
performance parameters such as gain, bandwidth, and isolation [16]. It offers the potential to boost system
performance, reliability, and efficiency, making it a valuable approach for 5G wireless communication systems.
However, the shape, design, placement, and optimization of DGS structure need to be properly configured to
meet the particular requirements of the MIMO system.
A four port MIMO antenna with DGS was designed on the Rogers R04350B substrate (òr=3.66,
h = 0.76 mm, tan (d )=0.0009) of dimensions 30 × 35 × 0.76 mm3. The design incorporates a feed network
based on a T-junction power combiner/divider, and the ground plane is made defective with a rectangular,
circular, and zigzag-shaped slotted structure to enhance the radiation characteristics of the antenna. The gain
reported was 8.3 dBi with 80% radiation efficiency at 28 GHz frequency. The ECC observed was < 0.01 and the
isolation was greater than 17 dB in the complete operating band. The design complexity was high as multiple
shapes and different sizes of slots were etched from the ground layer [17].
To improve the isolation and gain, a 2 × 2 MIMO antenna with DGS was designed for 28 GHz 5G
applications on Rogers RT/Duroid 5880 substrate (òr=2.2, h = 1.575 mm, tan (d )=0.0009) of dimensions
30 × 35 × 0.76 mm3 in the operating band of 26.16-29.72 GHz. The slot was cut from the radiating elements to
increase the bandwidth. The edge-to-edge distance between two elements was 1.2 λ. The multiple slots were
etched into the ground to form DGS. The gain observed was 7.1 dBi with 90% radiation efficiency. The isolation
reported was > 30 in the complete operating band. The design complexity was high as multiple slots were made
in the ground plane [18].
A 4-port MIMO antenna optimized for operation in the millimeter-wave band for 5G communication was
designed on a Rogers RT/Duroid 5880 substrate (òr=2.2, h = 0.79 mm, tan (d )=0.0009) of dimensions
25 × 25 × 0.79 mm3. The antenna is designed with a square-shaped defected ground structure (DGS) and
incorporates circular and rectangular slots to optimize the antenna impedance bandwidth. The isolation of
greater than 30 dB was observed in the complete operating band of 25.28-28.02 GHz [19].
A dual-band, two-element MIMO antenna for 5G wireless communications was designed on Rogers
RO4003 substrate (òr=3.55, h = 0.203 mm, tan (d )=0.0009) of dimensions 30 × 12 × 0.203 mm3. The designed
utilized orthogonal diversity to obtained desired isolation. The peak gain observed was 5.3 dBi at 38 GHz
frequency [20].
MIMO antenna with other mutual coupling reduction techniques such as EBG, AMC, and FSS were
employed to enhance the gain, bandwidth and isolation [21–24]. To increase gain and bandwidth, a mmWave
MIMO antenna with a 5 × 5 EBG structure and DGS was designed. At 28 GHz, 6 dBi of gain was measured [25].
However, in terms of gain, impedance bandwidth, isolation, and efficiency, these antennas fall short
significantly.
There are a number of designs available in the past literature that integrated DGS techniques into MIMO
antenna systems. A four port dual band MIMO antenna using orthogonal diversity was designed on Rogers
TMM4 substrate (relative permittivity (òr)=4.5, thickness (h)=1.52 mm, and loss tangent (tan (d ))=0.0020) of
dimension 17.6 × 17.6 mm2. A 3 × 5 EBG structure was used to reduce the mutual coupling. The isolation was
greater than 15 dB at 28 GHz and 38 GHz [26]. A dual band MIMO antenna was fabricated for 5G application of
dimension 60 × 60 mm2 on Rogers RT5880 dielectric material (òr=2.2, h = 0.508 mm, tan (d )=0.0009). The
DGS was used as mutual coupling reduction technique to enhance the isolation [27].
A dual band two port MIMO antenna was designed on FR4 substrate of dimension 14 × 28 × 1.6 mm3 to fit
into modern handset devices. The design consist of T-shaped radiating element with inverted I-shaped slots as
DGS providing a dual-band response. The gain obtained was 6.2 dBi with 96% radiation efficiency at 28 GHz.
The isolation was greater than 20 dB in the operating band lies from 26-42 GHz [28].
To reduce the mutual coupling, a dual band four port MIMO antenna was designed on Rogers TMM4
substrate (òr=4.5, h = 1.52 mm, tan (d )=0.0020) of dimension 17.76 × 17.76 × 1.52 mm3 design specially for
5G smartphone operating in 28/38 GHz bands. The decoupling network was used to increase the isolation [29].
An AMC-based MIMO antenna with a defected ground structure was designed on a Rogers RO 4003
substrate having dimensions 47 × 47 × 4.7 mm3. The antenna covered a -10 dB frequency band of 26.4-
31.5 GHz and realized a gain of 10.0 dB with 86% radiating efficiency. The isolation was greater than 25 dB, and
the ECC reported was 0.005 in the complete operating band. The edge-to-edge distance between the antennas

2
Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

was 4.5 mm (0.385 λ0) at 26 GHz. The high-profile 7 × 7 AMC structure was chosen to improve isolation and
gain; therefore, the design complexity was noticeable [30].
A four-element multilayered MIMO antenna of dimensions 24 × 24 × 4.258 mm3 was designed on a Rogers
RT/duroid 5880 substrate. To enhance the bandwidth, a 2 × 2 irregular-shaped metamaterial layer was placed
at the bottom of the MIMO configuration. The operational bandwidth observed was 23.3-28.8 GHz, and the
gain reported was 8.9 dBi with 80% radiating efficiency. The ground of the MIMO antenna was defective to
achieve the desired impedance bandwidth. The design complexity is quite high due to the employment of
multiple slot insertion in circular radiating patches along with defected ground and 2 × 2-unit array of meta-
surface cells [31].
A 2 × 4 MIMO antenna array was designed for 28/37 GHz 5G mmWave band applications on RT/duroid
5880 substrate (òr=2.2, h = 0.79 mm, tan (d )=0.0009) having dimensions 43.62 × 43.62 × 0.4 mm3. The
multiple shaped DGS was employed to improve the gain. Therefore, the design complexity is high. The gain
obtained was 7.9 dBi at 28 GHz resonant frequency. The -10 dB impedance bandwidth was 27.50-
28.35 GHz [32].
To overcome these problems 2 × 2 MIMO antenna with DGS technique is proposed in this paper to
improve the isolation, antenna gain, and bandwidth. The exiting designs achieve smaller dimensions, there is
often a trade-off in terms of radiation efficiency or gain. The proposed design offers the compact size, highest
gain (10.3 dBi) with high isolation, which indicates an improvement. In this paper, a compact, high gain, high
isolation MIMO antenna array is designed for 28 GHz mmWave 5G frequency band. The spatial diversity and
DGS are employed to enhanced the gain and reduce the mutual coupling, thus improves the performance
parameters of the MIMO antenna. The proposed design is less complex as same shape and equal dimensions of
DGS are used to implement the design.

2. 2 × 2 MIMO antenna Design

The two port MIMO antenna design is utilized for 5G applications [33]. In this part, the design of a 2 × 2 MIMO
antenna array, its fabricated hardware design, additional measurements, and MIMO parameters are discussed in
detail. The MIMO antenna array is designed on RT/Duroid 5880 substrates (òr=2.2, h = 0.79 mm, and
tan d =0.0009) with a dimension of 1.4λ0×2.8λ0×0.073λ0mm3, where λ0 is the free-space wavelength at
28 GHz. The Rogers RT/duroid 5880 is chosen for 28 GHz antenna designs due to its various advantages such as
low dielectric constant (òr=2.2), low loss tangent (tan d =0.0009), low water absorption (< 0.02%), and high
thermal stability. These features making it ideal for high-frequency and high-performance applications such as
28 GHz [34].
To improve the performance parameters of the MIMO antenna, a DGS with multiple slots is employed on
the ground layer. The 14 slots, each having a dimension of 0.7 × 2.5 mm2 separated by a distance 1.9 mm, are
etched from the ground, which helps to improve the gain and the isolation of the proposed antenna. The
schematic front and back views of the design are shown in figures 1 (a) and (b). The optimized dimensions are
shown in table 1.

3. Equivalent circuit model of proposed design

The proposed design’s equivalent circuit model is shown in figure 2. It is considered that the radiating element of
the proposed antenna is a parallel combination of RLC, as shown by Ra, La, and Ca. A parallel combination of the
Li and Ci series with C1 represents the inset feed slot. Similarly, a parallel combination of the Lu and Cu series with
C2 represents the upper slot of the patch antenna. The DGS is represented by equivalent Rdgs, Ldgs, and Cdgs. The
admittance of the radiating element circuit is calculated as given in equation (1) [35].

1 1 ⎞
Y= + j 2p ⎛⎜Ca f - 2 ⎟ (1)
Ra ⎝ 4p La f ⎠

where, Ra, La, and Ca are the resistance, inductance, and capacitance values of radiating element components as
shown in the circuit model. The equivalent impedance is calculated as given in equation (2).

1
Z= (2)
1
Ra (
+ j 2p Ca f -
1
4p 2L a f )

3
Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

Figure 1. Schematic view of proposed 2 × 2 MIMO antenna: (a) Front (b) Back.

Figure 2. Equivalent circuit model of proposed design.

The scattering parameters of the circuit model are calculated by using equations (3) to (8).

1
S12 = (3)
ZZ 0-1
1
1+ 2

2Z 0
S12 = -1
(4)
1
⎣ a (
2Z 0 + ⎡ R + j 2p Ca f - 4p 2L
1
a f ) ⎤
⎦

2Z 0
S12 = -1
(5)
(
2Z 0 + j 2p Ca f -
1
4p 2L a f )
4
Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

Figure 3. Design steps of proposed 2 × 2 MIMO antenna.

2Z 0 1
∣S12∣ = = (6)
2 2
4Z 02 + ( wc
Ca (w 20 - w 2c ) )
wc
Ca = (7)
2Z 0 (w 20 - w c2)
2Z 0 (w 20 - w c2)
La = (8)
wc w 20
where, ωc, is the angular frequency at 3 dB point and ωo is the angular frequency at 28 GHz. The inductance and
capacitance values are calculated with the help of above equations. The resistance Ra in the circuit model can be
calculated by using equations (9) to (10). The calculated circuit model component values for the design MIMO
antenna are shown in table 2. The coupling coefficients are represented by K12, K21, K1d, and K2d [29, 36].
2Z 0 2Z 0
S12∣ w = w 0 = = (9)
2Z 0 + Ze 2Z 0 + R a
2Z 0 (1 - S12∣w = w 0 )
Rp = (10)
S21∣w = w 0
The resonant frequency fr is calculated by using 11.
1
fr = (11)
2p L a C a

4. Results and discussion

4.1. Design steps and results

All the design steps of the proposed antenna array are shown in figure 3. The MIMO antenna array is fabricated
on Rogers RT/Duroid 5880 substrate (òr=2.2, h = 0.79 mm, and tan d =0.0009) of dimensions
1.4λ0×2.8λ0×0.073 mm3, where λ0 is the free-space wavelength at 28 GHz.
Initially, a series feed patch antenna with a quarter-wave transformer is designed with a patch dimension of
3.1 × 4.5 mm2. In Step I, the return loss does not go below -10 dB, suggesting that this initial stage doesn’t
provide acceptable impedance matching. However, this stage may still be considered as part of the design
process because it’s an early step toward achieving the desired performance. The progression in the following
steps indicates that the design evolves, improving the return loss characteristics to meet the desired specifications
(below -10 dB in Step II and beyond). This plot helps demonstrate how the design is refined at each stage to
achieve better performance in terms of return loss, with Step 1 serving as the baseline for further steps.
In the second step, the patch antenna array excited with a series-parallel feed is designed, which resonates at
27.5 GHz and covers the frequency band of 26.73-28.88 GHz. The antenna array gain is 8.60 dBi. After that, for
better impedance matching, an inset feed is employed, and an upper slot is used to enhance the bandwidth. The
frequency range covered is 26.82-29.50 GHz.
The gain simulated was 9.66 dBi with 96.98% radiation efficiency. To further increase the gain and for better
impedance matching, DGS is employed. The ground is modified, then multiple slots are used. The gain reported

5
Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

Figure 4. 2 × 2 MIMO antenna: S-parameters of each design step.

Table 1. Optimized dimensions of 2 × 2 MIMO antenna.

Parameters L W fl1 fl2 fl3 fw1/fw2 fw3

Value (in mm) 15 30 7 1.75 1.2 0.7 0.8

Parameters fl4 fw4 dm1 dm2 pw1 pl1 uw1
Value (in mm) 1.9 0.20 14.2 6.2 3.4 4.1 0.5
Parameters ul1 iw1 il1 L1 W1 gl1 gw1
Value (in mm) 0.5 0.5 0.6 12.5 21.6 2.2 1.2

Table 2. Calculated circuit model parameters for the design MIMO antenna.

Component Value Component Value Component Value

Lf 10.4 pH La 11.4 pH Ca 4.5 pF

Component Value Component Value Component Value
Ra=Rdgs 50 Ω Li 8.5 pH Ci 0.35 pF
Component Value Component Value Component Value
C1 2.2 pF Lu 7.5 pH Cu 2.1 pF
Component Value Component Value Component Value
Cdgs 7.9 pF Ldgs 8.7 pH C2 0.25 pF

was 10.1 dBi with a radiation efficiency of 96.5% at 28 GHz. In step-V, MIMO antennas are employed by using
the spatial diversity technique. The edge-to-edge distance between two antennas is 0.5λ0. The antenna achieved
a gain of 10.3 dBi with 97.01% radiation efficiency at 28 GHz. The S-parameters of all the design steps are shown
in figure 4.

4.2. Parametric analysis

This section discusses the parametric analysis of the proposed MIMO antenna array, which helps determine the
necessary dimensional parameter for the proposed antenna, which resonates at the desired frequency, such as
28 GHz. The important geometrical parameters taken for the parametric analysis are the patch length (pl1),
patch width (pw1), inset length (il1), and upper slot length (ul1).
The behaviors of S-parameters, reflection, and transmission coefficient are analyzed by varying the width of
the radiating elements, pw1, from 4.1 mm to 4.5 mm. The S11 curve shifts the left side towards the resonant
frequency as pw1 increases, while shifting the right side as pw1 decreases. At pw1 = 4.3, the desired resonant
frequency of 28 GHz is achieved. The variation of S-parameters with respect to patch width is shown in
figure 5 (a).
The CST Microwave is used to evaluate the antenna’s performance under different parameter settings, and
simulation results such as return loss, bandwidth, and impedance matching are obtained. The patch length (pl1)

6
Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

Figure 5. Parametric sweep: Impact of patch width on the S-parameters (b) Impact of patch length on the S-parameters.

Figure 6. Parametric sweep: (a) Impact of upper slot length on the S-parameters (b) Impact of inset feed length on the S-parameters.

Figure 7. Fabricated 2 × 2 MIMO antenna with Southwest 2.92 mm end launch connectors :(a) Front view (b) Back view.

is varied from 3.30 mm to 3.50 mm. At pl1 = 3.40 mm, the proper impedance matching is observed, and the
antenna is resonant at 28 GHz while other parameters are kept constant. The variation of S-parameters with
respect to patch length is shown in figure 5 (b).
The proper design and placement of slots play an important role in maximizing the bandwidth enhancement
while maintaining desired antenna characteristics such as radiation pattern, gain, and impedance matching. The
parametric analysis is explored for the upper slot length (ul1) through simulation to optimize the antenna’s
bandwidth performance. The variation of S-parameters with respect to upper slot width is shown in figure 6 (a).

7
Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

Figure 8. Fabricated 2 × 2 MIMO antenna: Measurement setup with VNA.

Figure 9. Fabricated MIMO antenna: Measurement setup in anechoic chamber.

Figure 10. 2 × 2 MIMO antenna: Simulated and measured S-parameters.

8
Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

Figure 11. 2 × 2 MIMO antenna: Snapshot of returnloss port from VNA.

Figure 12. 2 × 2 MIMO antenna: 2 × 2 MIMO antenna: Simulated and measured S-parameters (S12).

To analyze the effect of inset feed length on an antenna, parametric studies are performed. By systematically
varying the feed length (il1) and observing its impact on the antenna’s performance metrics, such as resonant
frequency and impedance matching, it can be properly determined. The il1 is varied from 0.2 mm to 1.2 mm
with an interval of 0.2 mm. As the inset feed length increases, the resonant frequency of the antenna becomes
lower, and vice versa. The variation of S-parameters with respect to patch length is shown in figure 6 (b).

5. Experimental results

The proposed two-port MIMO antenna array in the 28 GHz frequency band is fabricated using the lithography
process. The 2.92 mm end launch connectors from Southwest Microwave are used for measuring the port
performance of the prototype antenna shown in figure 7. The proposed MIMO antenna array is validated by the

9
Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

Figure 13. 2 × 2 MIMO antenna: Snapshot of measured S-parameters (S12) from VNA.

Figure 14. 2 × 2 MIMO antenna: Gain and efficiency.

Agilent Technologies vector network analyzer (VNA)-N5247A with a frequency range of 10 MHz to 67 GHz to
calculate S-parameters as shown in figure 8. The calibration is completed using the open-short-load technique.
Figure 9 shows the fabricated MIMO antenna and the measurement setup in the anechoic chamber. The
simulated and measured reflection coefficients of the proposed MIMO antenna array are compared in figure 10
and snapshot of the reflection coefficient is shown in figure 11. The simulated S-parameters -10 dB impedance
bandwidth lies between 27.13-29.44 GHz, providing a 2.31 GHz bandwidth, while the measured operating band
lies between 26.95-29.50 GHz with a 2.55 GHz bandwidth.
The proposed antenna covered the majority of the 5G mmWave frequency band declared by the Federal
Communications Commission (FCC). The minor discrepancies between simulated and measured results arise
due to various factors such as manufacturing tolerances, lossy substrates, environmental effects, cable losses, and
calibration errors.

10
Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

Figure 15. Simulated and Measured radiation patterns in XY-plane at 28 GHz when port-1 is excited.

Figure 16. 2 × 2 MIMO antenna: Simulated and measured radiation patterns in XY-plane at 28 GHz when port-2 is excited.

The simulated and measured S-parameter curves, |S12| are illustrated in figures 12 and 13. The -10 dB
frequency lies between 26.95-29.50 GHz, and |S11| is -45.18 dB at 28 GHz and |S12| is 45.24 dB at 28 GHz. The
isolation (|S21| 28 dB) in the complete operating band. The snapshot of the transmission coefficient, S12 from
VNA, is shown in figure 13.
The gain-efficiency graph for the proposed MIMO antenna array is shown in figure 14. The gain of 2 × 2
MIMO antenna is 10.3 dBi at 28 GHz. The gain varies from 9.58-0.35 dBi in the operating band. The radiation
efficiency of 97.01% and total efficiency of 90.98%, respectively, are achieved at 28 GHz. The radiation efficiency
varies from 96.12%-99.63% and the total efficiency varies from 73.52%-96.05%, respectively, in the
operating band.
The far-field gain of the proposed MIMO antenna is carried out in an anechoic chamber in the presence of
the standard horn antenna and power meter. Then, with the help of the FRIIS equation as given in equation (12),
gain is measured. The gain is measured at θ=0◦ in E-field. Due to port and cable coupling losses, slight changes
in simulated and measured gains are observable [37].

PT G T GR l2
PR = (12)
(4pR)2

Where PR is the power at the receiving antenna, PT is the power at the transmitting antenna, GT is the gain of the
transmitting antenna, GR is the gain of the receiving antenna λ is the wavelength, and R is the distance between
the antennas

11
Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

Figure 17. Surface current distribution of 2 × 2 MIMO antenna: Without DGS.

Figure 18. Surface current distribution of 2 × 2 MIMO antenna: With DGS.

The E-field radiation patterns are shown in figure 15. In the case of the E-plane (f=0◦), it is found that the
main lobe magnitude and direction for the MIMO antenna array are 8.85 dBi and 0◦, respectively. The half
power beam width (HPBW) is 39.5◦, and the side lobe level (SLL) is -17.7 dB, which is achieved due to the
employment of series-fed patch elements. The measured main lobe magnitude is 8.68 dBi with a 3◦ main lobe
direction. The 3-dB angular width is 57.3◦ and SLL is -10.4◦. The main lobe magnitude and direction for
simulated H-field (f=90◦) are 10.5 dBi and 22◦. The HPBW is 60.7◦, and the SLL=-16.7 dB as shown in
figure 16. The measured main lobe magnitude is 10.3 dBi at a direction of 29◦ and the 3 dB angular width is 54.9◦.
To reduce the inter element coupling between two arms of the feeding network, multiple slots of DGS are
used. Thus, due to these approaches, a lower value of surface current is concentrated on the unexcited port. The
current concentrates on the feed line and on the patch elements of the excited port, and current is also found on
the elements of the non-excited port due to the mutual coupling. To reduce the mutual coupling, multiple slots
of DGS are employed. The maximum current concentrates on the feedlines and the radiating elements. A
negligible amount of current is seen in the other port. The surface current distribution without DGS is shown in
figure 17 and with DGS is shown in figure 18.

6. MIMO diversity performances

In the proposed design diversity performance parameters of the MIMO antennas are calculated such as ECC,
CCL, MEG, and TARC. The envelope correlation coefficient (ECC)of the MIMO antenna, can be
mathematically related to the S-parameters of the MIMO antenna by equation (13) [38].

12
Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

Figure 19. 2 × 2 MIMO antenna: Simulated and Measured ECC.

Figure 20. 2 × 2 MIMO antenna: Simulated and Measured ECC (b CCL).

N
å Si*,nSn*,j
n= 1
ECC = (13)
N
 ⎡1 -
⎢ å Si*,nSn*,k⎤⎥
k = (i, j ) ⎣ n= 1 ⎦

where i (1 → N) and j (1 → N) are antenna ports, n is the number of radiating elements, and Si, n and Sn, j are
scattering parameters of antenna elements. In the proposed design, N=2.
For proper MIMO operation, the ECC value lies between 0.3-0.4. In our proposed work, the simulated The
ECC value is 5.06 × 10−8 at 28 GHz and less than 0.0002 in the entire operating band, and the measured value of
ECC is 3.06 × 10−8 at 28 GHz and less than 0.0005 in the entire operating band, as shown in figure 19.
The channel capacity loss (CCL), measured in bps/Hz, is considered an important performance metric for
MIMO systems. Its theoretical value is less than 0.4 bps/Hz. Mathematically, it is calculated with the help of
S-paramaters and expressed from the equations (14) to 17 [39].
CCL = - log2∣y R∣ (14)

r r
y R = ⎡ r11 r12 ⎤ (15)
⎣ 21 22 ⎦

rii = 1 - (∣Sii∣2 - (∣Sij ∣2 ) (16)

13
Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

Figure 21. 2 × 2 MIMO antenna: Simulated and measured MEG.

Figure 22. 2 × 2 MIMO antenna: TARC with different set of excitation angles at ports.

rij = - (Sii*Sij + S*ji Sij ) (17)

The simulated and measured values of CCL for the complete operating band are shown in figure 20.
The mean effective gain (MEG) is used to assess and compare the directional performance of different
antennas in a MIMO configuration. By using equations (18) to (19), MEG can be defined as.
N
MEGi = 0.5⎡1 -
⎢ å ∣Sij∣2 ⎤⎥  -3 dB (18)
⎣ k=1 ⎦

- 3 dB  MEGi - MEGj < 12 dB (19)

The mean effective gain of another performance indicator for MIMO antennas is the MEG. The MEG
should be below -3 dB for good MIMO performance [36, 37]. The suggested MIMO antennaʼs MEG lies below
-6 to -9 dB. as shown in figure 21 (a) for the gaussian medium, for the whole band, which is more than sufficient
for a high diversity performance in a MIMO antenna system.
Total active reflection coefficient (TARC) is a very useful MIMO performance metric. For the N-port set of
MIMO antennas, if θ is an independent and identically distributed Gaussian phase of a random process, then
TARC can be defined by the equation (20) [40, 41].

14
 Phys. Scr. 99 (2024) 115520
Table 3. Comparative analysis of the proposed design with the existing ones.

Ref No Size (mm3) Frequency band/bands (GHz) Gain (dBi) Radiation Efficiency (%) Isolation (dB) ECC Description Design Complexity

[17] 3.3λ0 ×2.8λ0 26.2-29.72 7.1 90 30 <0.0005 4-port MIMO antenna with High
×0.13λ0 multiple DGS
[18] 5.3λ0×2.2λ0 26-30 5.2 90 20 <10−5 2-port MIMO antenna with stubs High
×0.66λ0 and DGS
[19] 3.27λ0×2.8λ0 27.5-28.35 8.1 82 20 <0.0004 2-port MIMO antenna Moderate
×0.073λ0
[20] 1.20λ0×1.52λ0 23.6-24.1 6.0 80.5 37 <0.24 2-port MIMO antenna with EBG High
×0.02λ0 and DGS
15

[21] 2.68λ0×1.75λ0 27.58-28.4 10.2 90 35 0.0001 2-element MIMO antenna array High
×0.05λ0 using SRR
[22] 2.8λ0×2.8λ0 27.5-29.50 8.65 81.5 24 0.00041 2 × 2 MIMO antenna array with High
×0.14λ0 multiple CSRR
[28] 1.3λ0×2.61λ0 28/38 7 90 40 0.0015 2 × 2 MIMO antenna with Low
×0.14λ0 DGS
[29] 1.65λ0×1.65λ0 28/38 8.9 90 > 30 0.0015 2 × 2 MIMO antenna with Moderate
×0.14λ0 decoupling structure
[30] 4.38λ0×4.38λ0 26.4-31.5 10 80 > 25 0.005 2 × 2 MIMO antenna with AMC High
×0.43λ0 and DGS
[31] 2.24λ0×2.24λ0 23.34-28.8 8.9 80 > 20 0.005 2 × 2 MIMO antenna with AMC High
×0.39λ0 and DGS
[32] 4λ0×4λ0 27.5-28.35 7.8 76 > 20 0.005 2 × 2 MIMO antenna with AMC High
×0.03λ0 and DGS
Proposed work 1.40λ0×2λ0 26.95-29.50 10.3 97.01 45.24 10−7 2 × 2 MIMO antenna with Low
×0.074λ0 DGS

M Pant and L Malviya

Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

n n 2
TARC = N -1 2 å å Sik e jqk - 1 (20)
i=1 k=1

∣S12 + S11 e jq∣2 + ∣S21 + S22 e jq∣2

TARC = (21)
2
For N = 2, the above expression can be written as equation (21) [42]. As, can be seen from figure 22, the value
of θ (excitation angle) is varied from 0° to 180° and TARC values are calculated for different values of θ. The
TARC results are appropriate for θ=0° and θ=180° because they cover the -10 dB impedance bandwidth.
The comparison of various MIMO antenna designs, including details on their dimensions, frequency bands,
peak gain, radiation efficiency, isolation, ECC, descriptions, and design complexity are shown in table 3. In
MIMO antenna design, there’s often a trade-off between design complexity and performance metrics like gain,
isolation, radiation efficiency, and ECC. The antenna designs with metamaterial such as EBG, AMC, and
decoupling structures are generally considered complex. These designs involve intricate layouts and fabrication
process. The complex designs often achieve higher gains due to the use of advanced structures that improve the
antenna’s directivity and reduce losses. For example, antennas design in [21] and [30] with SRR and AMC has
gain above 10 dBi. However, simpler designs may have lower gains as seen in references [18] and [20], which is
lower compared to the complex designs. The complex designs often feature techniques to improve isolation
between antenna elements, such as, CSRR, SRR, EBG, and AMC. High isolation as in reference [20, 21, 28, 29]
and in the proposed design are greater than 30 dB, crucial in MIMO systems to reduce interference between
multiple antenna elements, improving overall system performance.
The DGS employed in the earlier designs would not deliver the desired outcomes at 28 GHz frequency. Also,
they acquired much spacing compared to the proposed design. As a result, we have come to the conclusion that
the proposed design showed the lowest dimension, higher gain, high efficiency, extremely high isolation, and
lowest ECC when compared to the studied references.

7. Conclusion

In conclusion, the implementation of the DGS with a 2 × 2 MIMO antenna is presented. DGS offers improved
antenna performance by suppressed surface waves, reduced mutual coupling, and improved gain. Also, it offers
a structured and comprehensive analysis of the proposed antenna’s capabilities, including simulated and tested
results, to validate its potential for 5G communications. The gain of the MIMO antenna achieved is 10.3 dBi with
97.01% radiation efficiency at 28 GHz. The bandwidth of the MIMO antenna lies between 26.95-29.50 GHz.
The paper emphasizes the importance of the proposed antenna’s performance parameters, such as ECC, MEG,
TARC, and CCL, in evaluating its suitability for 5G applications. The ECC value observed is less than 10−5 in the
complete operating band. Future research may focus on developing massive MIMO systems using the proposed
design that operate in the mmWave bands and overcome issues like beamforming and high path loss.

Data availability statement

The data cannot be made publicly available upon publication because no suitable repository exists for hosting
data in this field of study. The data that support the findings of this study are available upon reasonable request
from the authors.

ORCID iDs

Mohit Pant [Link]

References
[1] Agiwal M, Roy A and Saxena N 2016 Next generation 5G wireless networks: A comprehensive survey IEEE Communications Surveys &
Tutorials 18 1617–55
[2] Hakeem S A A, Hussein H H and Kim H 2022 Vision and research directions of 6G technologies and applications Journal of King Saud
University-Computer and Information Sciences 34 2419–42
[3] Li J, Chen L and Chen J 2021 Enabling technologies for low-latency service migration in 5G transport networks [Invited] IEEE/OSA
Journal of Optical Communications and Networking 13 A200–10
[4] Parkvall S, Dahlman E, Furuskar A and Frenne M 2017 NR: The New 5G Radio Access Technology IEEE Communications Standards
Magazine 1 24–30
[5] Gupta P, Malviya L and Charhate S V 2019 5G multi-element/port antenna design for wireless applications: a review Int. J. Microwave
Wireless Technolog. 11 918–38

16
 Phys. Scr. 99 (2024) 115520 M Pant and L Malviya

[6] Fuentes M et al 2020 5G New Radio Evaluation Against IMT-2020 Key Performance Indicators IEEE Access 8 110880110880–96
[7] Pant M and Malviya L 2023 Design, developments, and applications of 5G antennas: a review Int. J. Microwave Wireless Technolog. 15
156–82
[8] Rao L, Pant M, Malviya L, Parmar A and Charhate S V 2020 5G beamforming techniques for the coverage of intended directions in
modern wireless communication: in-depth review Int. J. Microwave Wireless Technolog. 13 1–24
[9] Bogale T E, Wang X and Le L B 2017 mmWave communication enabling techniques for 5G wireless systems: A link level perspective in
MmWave Massive MIMO (Academic Press) 195–225
[10] Malviya L and Gupta P 2021 Millimeter Wave High-Gain Antenna Array for Wireless Applications IETE Journal of Research 69 (5)
2645–54
[11] Hmamou A, Kiouach F, Das S, El Ghzaoui M, Islam T and Madhav B T P 2024 A MIMO Antenna Array Featuring Dual Wideband and
High Gain for 5G NR n257/n258/n260/n261 Bands Applications Phys. Scr. 99 095506
[12] Kumar P, Singh A K, Kumar R, Sinha R, Mahto S K, Choubey A and Al-Gburi A J A 2024 High isolated defected ground structure based
elliptical shape dual element MIMO antenna for S-band applications Progress in Electromagnetics Research C 143 67–74
[13] Arya A K, Kartikeyan M V and Patnaik A 2010 Defected ground structure in the perspective of microstrip antennas: a review Frequenz
64 79–84
[14] Kumar P, Ali T and Pai M M 2021 Electromagnetic metamaterials: A new paradigm of antenna design IEEE Access 9 18722–51
[15] Roges R, Sharma S, Malik P K, Islam T, Asha S and Das S 2024 A compact circularly polarized dual port MIMO antenna with DGS and
parasitic patch for modern wireless communication and IOT applications Phys. Scr. 99 055515
[16] Malviya L, Parmar A, Solanki D, Gupta P and Malviya P 2020 Highly isolated inset-feed 28 GHz MIMO-antenna array for 5G wireless
application Procedia Computer Science 171 1286–92
[17] Khalid M, Iffat Naqvi S, Hussain N, Rahman M, Fawad, Mirjavadi S S, Khan M J and Amin Y 2020 4-Port MIMO antenna with defected
ground structure for 5G millimeter wave applications Electronics 9 71
[18] Hussain M, Mousa Ali E, Jarchavi S M R, Zaidi A, Najam A I, Alotaibi A A and Ghoneim S S 2022 Design and characterization of
compact broadband antenna and its MIMO configuration for 28 GHz 5G applications Electronics 11 523
[19] Guler C and Bayer Keskin S E 2023 A novel high isolation 4-Port compact MIMO antenna with DGS for 5G applications Micromachines
14 1309
[20] Sabek A R, Ali W A and Ibrahim A A 2022 Minimally coupled two-element MIMO antenna with dual band (28/38 GHz) for 5G wireless
communications Journal of Infrared Millimeter, and Terahertz Waves 43 335–48
[21] Esmail B A and Koziel S 2023 High isolation metamaterial-based dual-band MIMO antenna for 5G millimeter-wave applications AEU-
Int. J. Electron. Commun. 158 154470
[22] Khan D, Ahmad A and Choi D Y 2024 Dual-band 5G MIMO antenna with enhanced coupling reduction using metamaterials Sci. Rep.
14 96
[23] Mahnoor K, Naqvi S I, Hussain N, Rahman M, Mirjavadi S S, Khan M J and Amin Y 2020 4-Port MIMO antenna with defected ground
structure for 5G millimeter wave applications Electronics 9 1
[24] El Hadri D, Zakriti A, Zugari A, El Ouahabi M and El Aoufi J 2020 High isolation and ideal correlation using spatial diversity in a
compact MIMO antenna for fifth-generation applications International Journal of Antennas and Propagation 2020 1–102740920
[25] Iqbal A, Basir A, Smida A, Mallat N K, Elfergani I, Rodriguez J and Kim S 2019 Electromagnetic bandgap backed millimeter-wave
MIMO antenna for wearable applications IEEE Access 7 111135111135–44
[26] Al-Gburi A J A 2024 5G MIMO antenna: compact design at 28/38 GHz with metamaterial and SAR analysis for mobile phones Prz.
Elektrotech. 2024 171–174
[27] Cuneray K, Akcam N, Okan T and Arican G O 2023 28/38 GHz dual-band MIMO antenna with wideband and high gain properties for
5G applications AEU-Int. J. Electron. Commun. 162 154553
[28] Thakur E, Gupta A, Abdulhameed M K, Khaleel A D and Abdullah Al-Gburi A J 2024 Microstrip Antenna with Two Elements and
Defected Ground Structure for 5G Mobile Applications at 28/38 GHz Progress in Electromagnetics Research C 146 177–85
[29] Elabd R H and Al-Gburi A J A 2024 Low mutual coupling miniaturized dual-band quad-port MIMO antenna array using decoupling
structure for 5G smartphones Discover Applied Sciences 6 189
[30] Ibrahim A A and Ali W A 2021 High gain, wideband and low mutual coupling AMC-based millimeter wave MIMO antenna for 5G NR
networks AEU-Int. J. Electron. Commun. 142 153990
[31] Sehrai D A, Asif M, Shah W A, Khan J, Ullah I, Ibrar M, Jan S, Alibakhshikenari M, Falcone F and Limiti E 2021 Metasurface-based
wideband MIMO antenna for 5G millimeter-wave systems IEEE Access 9 125348125348–57
[32] Aghoutane B, Das S, Ghzaoui M E, Madhav B T P and El Faylali H 2022 A novel dual band high gain 4-port millimeter wave MIMO
antenna array for 28/37 GHz 5G applications AEU-Int. J. Electron. Commun. 145 154071
[33] Pant M and Malviya L 2022 Fifth generation MIMO antenna array with diversity and mutual coupling reduction technique 2022 IEEE
11th International Conference on Communication Systems and Network Technologies (CSNT) (Indore, India, 23-24 April 2022) (IEEE)
pp 36–41
[34] Rogers Corporation, publication no. 92-101 and 92-125, 2018. [Online]. Available: [Link]
[35] Elabd R H and Al-Gburi A J A 2024 Super-compact 28/38 GHz 4-port MIMO antenna using metamaterial-inspired EBG structure
with SAR analysis for 5G cellular devices Journal of Infrared, Millimeter, and Terahertz Waves 45 35–65
[36] Elabd R H and Al-Gburi A J A 2024 Ultra-Compact 4-Port MIMO antenna with defected ground structure and SAR analysis for 28/38
GHz 5G mobile devices J. Electromagn. Waves Appl. 1–26
[37] Hamdan S, Mohamed H A and Hamad E K 2022 Design of High Isolation Two-port MIMO Two-element Array Antenna Using Square
Split-Ring Resonators for 5G Applications International Journal of Microwave and Optical Technology 17 (4) 339–46
[38] Singh M, Singh S and Islam M T 2022 CSRR loaded high gained 28/38GHz printed MIMO patch antenna array for 5G millimeter wave
wireless devices Microelectron. Eng. 262 111829
[39] Biswas A K and Chakraborty U 2019 Compact wearable MIMO antenna with improved port isolation for ultra-wideband applications
IET Microwaves Antennas & Propagation 13 498–504
[40] Malviya L, Panigrahi R K and Kartikeyan M V 2018 Four element planar MIMO antenna design for long-term evolution operation
IETE Journal of Research 64 367–73
[41] Chouhan S, Panda D K, Gupta M and Singhal S 2018 Multiport MIMO antennas with mutual coupling reduction techniques for
modern wireless transreceive operations: A review Int. J. RF Microwave Comput. Aided Eng. 28 e21189
[42] Chae S H, Oh S and Park S 2007 Analysis of mutual coupling, correlations, and TARC in WiBro MIMO array antenna IEEE Antennas
Wirel. Propag. Lett. 6 122–5

17

View publication stats