1 Introduction

The globe is approaching the fifth generation (5G) era due to the wireless communication industry’s fast expansion [1]. With this backdrop, wireless communication has to meet unprecedented demands for data rate, capacity, and latency. The MIMO antenna has proven to be an essential component of 5G communication systems due to its benefits in terms of increased Improvements in capacity, reduced delays, durability against fading caused by multiple paths, and better data rate [2]. Nonetheless, a progressive reduction in the spacing between antenna components is caused by 5G terminals or platforms that have significant space constraints.

Quality of the antenna, counting impedance BW, radiation shape, and efficiency, might worsen once the antenna component arrangement is smaller than half wavelength in distance, as per the antenna array theory, which states that the MC amid antenna components cannot be minimal [3,4,5]. As a result, MC among antenna parts has always been a problem when designing small MIMO antennas, particularly when space is limited.

In recent decades, a number of efficient decoupling strategies have been planned to reduce the MC between antenna parts [6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44]. These approaches may be broadly categorized into three groups based on the decoupling mechanism [11]. The first kind involves directly reducing the current or wave connection across antenna parts in order to destroy the coupling channel. A common example is the Defected Ground Structure (DGS) [12,13,14,15], which has a band-reject response and can effectually stop surface wave spread by causing disturbances to the current spreading on the antenna ground. However, the main disadvantage of the DGS is that its Front-to-Back Ratio (FBR) is usually high. Thankfully, because of their distinct electromagnetic properties at particular frequencies, different meta-structures have been developed, such as Single Negative metamaterial (SNG) [16, 17], Antenna Decoupling Surface (ADS) [18, 19], Frequency Selective Surface (FSS) [20, 21], Polarization Conversion Surface (PCS) [22], meta-surface construction [23, 24], and EBG building [25,26,27,28], which offer new insights into repressing the connection current or wave between components of antenna. To lessen the coupling, the suggested meta-surface structure was used in between MIMO antenna elements in [20].

The meta-surface structure is thought to function as a spatial band stop filter, therefore suppressing the elements’ spatial electromagnetic movement. The meta-surface isolation wall’s expensive cost and limited BW, however, are drawbacks. In order to address the MC problem in large-scale antenna array organizations, a unique antenna decoupling surface was created in [22] that counteracts coupling waves by producing out-of-phase reflected waves. However, the antenna profile was much enlarged since the decoupled areas of the antenna were positioned around half a wavelength over the antenna’s surface.

A micro-base-station antenna with a polarization of 1 × 2 was constructed in [24] using a meta-surface isolation wall as its foundation. The unique isolation obstacle made of two meta-surfaces connected to the ground plane was employed to lessen interface between micro-base-station arrays. But in order to achieve the same isolation, the meta-surface structure comes at a far greater cost than a standard copper patch. A unique fence-type decoupling building was illustrated in [27], mostly consisting of many slots of identical size.

By acting as a band-stop filter, the many slits efficiently reduced the coupling current among two ports. Though, the manufacturing process is complex and the inserted decoupling assembly impacts the impedance matching at small frequency.

The second category of decoupling approach operates by canceling the initial connection route between elements and adding a new coupling track through the use of a Neutralization Line (NL) [29,30,31,32], or Decoupling Network (DN) [33,34,35], Parasitic Element (PE) [36,37,38,39].

Antenna elements are directly linked to both NLs and DNs, which are made up of lumped components and microstrip circuits. These elements can supply the decoupling current with an identical amplitude and reverse phase to offset the link current. But both NLs and DNs contribute extra insertion loss and make it harder to match impedance in antenna circuits as the interconnector, which reduces antenna efficiency [35]. The parasitic element, which is positioned close to antenna elements without a physical connection, can offer an alternative to the aforementioned strategy for creating an indirect coupling channel.

In order to offset the connection current, an induced current with the same amplitude and reverse phase arises in the parasitic element when the antenna is stimulated. Nevertheless, the initial radiation shape of the antenna is distorted by the field emitted by the generated current in the parasitic element [38]. The majority of decoupling structures that have been documented up to this point usually need additional distance among antenna components or straight up over the surface of the antenna, thus invariably adds to the complexity of the scheme and antenna profile.

Furthermore, the insertion effect is always present when decoupling structures are used, which lowers the antennas’ overall performance, including pattern, radiation efficiency, and impedance match.

Consequently, a third type of decoupling style, in which the MIMO antenna shows excellent natural isolating following design, as the decoupling basis —has been utilized to address the aforementioned drawbacks of depending on decoupling structures [40,41,42,43,44,45,46,47,48,49].

Based on this idea, typical representative decoupling techniques comprise, form diversity [40, 41], polarization multiplicity [42, 43], self-decoupling [44, 45] and the scheme of distinguishing approaches [46]. These techniques have the unique benefit of leaving the antenna act essentially unaffected because no added construction is placed in or around the radiating assemblies. Nonetheless, there are strict restrictions on antenna configurations, feeding techniques, and layout due to polarization or pattern variety. The self-decoupling approach lacks adaptability and an organized design guideline, despite being more inventive and flexible than polarization or shape variety. This is because the self-decoupling effect is typically achieved by utilizing an antenna’s inherent property.

To lessen MC, parasitic components using polarization variety were utilized in [43]. Broadband antennas are not suited for the polarization diversity’s decoupling effect, though. A weak field-based method of self-decoupling was put out in [45]. The neighboring element must be positioned in the region of the feeble field created by the two fields—the radiating patch and the feeding structure—that cancel each other out in order for the mechanism to function. However, this method of decoupling is limited to inset fed patch antenna arrays.

This work proposes and develops a two-element Orthogonal Printed Elliptical Slot Antenna (OPESA) array with good isolation. A further analysis of the EBG structure’s operation reveals that the isolated branch and six rings on the vertical sheet of the building are used to prevent MC among the antenna components. The isolated branch is incorporated into the antenna substrate, whereas the vertical EBG sheet is situated in the center line of the antenna top layer.

The rest of this essay is arranged as follows: The single element Scheme I’s design details are given in Sect. 2. Section 3 presents the single element Scheme II’s design details. In Sect. 4, we talk about the MIMO antenna, covering antenna arrangement, strategy processes, the decoupling idea, and theoretical study of the simulation findings. In Sect. 5, the impact of a few critical decoupling structure factors on antenna performance is examined. Time domain analysis is introduced and described in Sect. 4. Section 5 provides a thorough comparison with earlier, comparable research as well as a commentary. Finally, Sect. 6 presents a succinct conclusion.

2 Designed millimeter wave broadband antenna scheme I

The basic shape of the broadband Printed Elliptical Slot Antenna (PESA) is seen in Fig. 1. Every aspect of the antenna design parameter is displayed in both the top and bottom views. The Scheme antenna has a dielectric constant of \({\epsilon }_{r}\) = 4.3 and is built on a 15 × 15 mm2 (Wx × Wy) FR4 with a thickness of 1.6 mm. An elliptically designed slit carved off the ground plane of center (P2) (representing a Ground Plane Aperture (GPA)) is positioned non-concentrically inside an elliptically molded radiating patch of center (P1). A 50-ohm close proximity-feed microstrip line, positioned on the substrate’s opposite side as seen in Fig. 2, is used to excite the patch. Defected Microstrip structure (DMS) in the form of an inverted L is imprinted in the feed line to enhance impedance matching, as demonstrated in Fig. 3a.

Fig. 1
figure 1

Millimeter wave wideband antenna Scheme I (PESA), geometric configuration: a top view; b bottom view

Full size image
Fig. 2
figure 2

Simulated reference impedance of proposed Scheme I

Full size image

As seen in Fig. 3b–d, the antenna is parametrically studied for key design parameters (R1x, L1, and W6). Table 1 provides an overview of the optimal antenna design parameters based on these parametric research and optimization.

Fig. 3
figure 3

Simulated return loss of designed antenna. a With and without DMS, b Different value of R1x, c different value of L1 and d different value of W6

Full size image
Table 1 The optimized parameters of proposed millimeter wave antenna Scheme –I (units in mm)
Full size table
Fig. 4
figure 4

Simulated return loss of designed antenna

Full size image

The full-wave electromagnetic simulator CST is utilized to estimate and optimize the performance of the suggested antenna. Using a 50 ohm port impedance. As shown in Fig. 4, the results indicate that the simulated return loss BW (|S11| < −10 dB) spans 22.5 GHz to 34 GHz, covering the 28 GHz bands. The proposed antenna Scheme I’s simulated radiation shapes at a frequency of 28 GHz are displayed in Fig. 5. As it can be seen in the image, the antenna exhibits omni-directional forms with a maximum observed gain of 6.30 dBi.

Fig. 5
figure 5

Simulated radiation shapes of proposed antenna Scheme I, at 28 GHz.

Full size image

3 Proposed millimeter wave broadband antenna scheme II

This portion presents the design of the Orthogonal Printed Elliptical Slot Antenna (OPESA) for operating at 28/38 GHz. The knowledge behindhand the recommended antenna is to increase gain and broaden the BW by using two components from Scheme I (PESA) with a 100 ohm proximity-feed microstrip line that is 90° angled toward one another. As seen in Fig. 6, the suggested OPESA is printed on the same side of a FR4 substrate with a 4.3 dielectric constant and a 0.025 loss tangent. Figure 7a illustrates how the OPESA is stimulated using a 50 ohm proximity-feed microstrip line. The top and bottom views are displayed, together with a detailed explanation of the geometric design factors. Table 2 provides the optimum design parameters.

Fig. 6
figure 6

Structural architecture of proposed millimeter wave wideband antenna Scheme II (OPESA), a top view, and b bottom view

Full size image
Fig. 7
figure 7

a Simulated reference impedance of proposed Scheme II (OPESA). b The equivalent circuit of UWB antenna, c The equivalent circuit of proposed antenna and d ADS and CST outcomes

Full size image
Table 2 The suggested millimeter wave antenna scheme II’s optimum variables (in millimeters)
Full size table

Scheme II and planned Scheme I are compared in Fig. 8. The simulated return loss BW (|S11| < −10 dB) for Scheme I can be shown in Fig. 8a as ranging from less than 25 GHz to 34 GHz, covering the frequency bands (28 GHz), but also from less than 25 GHz to greater than 40 GHz, covering the bands of 28/38 GHz. The simulated gain is displayed in Fig. 8b. Scheme I has a maximum gain of 6.5dBi, whereas Scheme II has a maximum gain of 9.5dBi.

The simulated efficiency is displayed in Fig. 8c. Scheme I has a maximum efficiency of 87.9%, whereas Scheme II has a maximum efficiency of 91.5%. It is observed that the Scheme II covers the dual tone frequencies of 28 and 38 GHz and improves gain and efficiency. Table 3 presents a summary of the antenna Scheme I and Scheme II data, including the estimated radiation efficiency, gain, and BW. It was observed that Scheme II outperforms Scheme I.

Fig. 8
figure 8

Simulated parameters of proposed Scheme I and Scheme II. a |S11|, b Gain, and c Efficiency

Full size image
Table 3 Scheme I and Scheme II contrast
Full size table

The LC resonant circuit is a reasonable approximation for a narrowband antenna. Broadband antennas consist of a cascade of LC resonators, whose overlap** resonant frequencies form the whole wideband resonances of the UWB antenna. The equivalent circuit of the ultra-wideband antenna is drawn using the Advanced Design System (ADS) software. Figure 7b displays the lumped element of the circuit. Figure 7c displays the equivalent circuit of proposed antenna.

As seen in Fig. 7d, the ideal simulation and the similar circuit’s simulation results are contrasted. The Lo and Co components for the transmission line of the suggested antenna in the analogous circuit design are found using Eqs. (1–2) [50]. The inductance and capacitance for the lumped circuit components are computed using Eqs. (3–4). Using ADS software, the antenna equivalent circuit is drawn once the component values have been combined and adjusted. The obtained findings for the total resonant frequencies are shown in the figure to be in good agreement between the CST and ADS software. The circuit achieves the following final results for the component circuit: Lo = 0.192 nH, Co = 0.093 pF, R1 = 57.96 Ω, L1 = 1.686 nH, C1 = 1.94 pF, R2 = 63.67 Ω, L2 = 0.919 nH, C2 = 0.87 pF, R3 = 45.94 Ω, L3 = 0.186 nH, C3 = 1.49 pF, R4 = 61.19 Ω, L4 = 0.151 nH, C4 = 0.83 pF.

$${L}_{o}=100 h\left(4\sqrt{\frac{{W}_{f}}{h}}-4.21\right)$$
(1)
$${C}_{o}={W}_{f}\left[\left(9.5{\epsilon }_{r}+1.25\right)\frac{{W}_{f}}{h}+5.2{\epsilon }_{r}+7\right]$$
(2)
$$L=\frac{img\left({Z}_{11}\right)}{2\pi f}$$
(3)
$$C={\left[{\left(2\pi f\right)}^{2}L\right]}^{-1}$$
(4)

where \({W}_{f}\), \(h\), \({\epsilon }_{r}\) is the feed width, substrate thickness, and dielectric constant, respectively.

The antenna is constructed, and Fig. 9a–b display a shot of the finished antenna structures. Figure 10a displays the reflection coefficient |S11| curves that were measured and simulated. The observed results demonstrate high agreement among the simulated and measured results, and the measured return loss BW (|S11| < −10 dB) spans from less than 25 GHz to more than 40 GHz, indicating millimeter wave operations with dual-band at 28/38 GHz. Figure 10b–c display the efficiency and gain. The antenna’s maximum efficiency and gain are determined to be 91.5% and 9.5 dBi, respectively.

The 3D and 2D radiation characteristics for frequencies of 28 and 38 GHz, respectively, are displayed in Figs. 11 and 12. The E Plane and H Plane radiation outlines, both measured and simulated, are displayed. The antenna displays the usual radiation characteristics of a monopole. It can be seen from according to the simulated results, the recommended antenna emits radiation in all directions. There is good agreement between the simulated and observed radiation forms. Figure 1313 demonstrates the corresponding current distribution of the proposed antenna without and with defected ground plane. It visualized that the currents are densely populated and flowing in different directions and this makes coupling between the two antennas causing limited bandwidth as shown in Fig. 13a–b. However, as shown in Fig. 13c–d, using the defected ground plane, the current flows through the transmission feed line and less currents is flowing between the two antennas as it isolates the coupling between them.

Fig. 9
figure 9

Fabricated Scheme II (OPESA): a top view and b bottom view

Full size image
Fig. 10
figure 10

Simulated and measured parameters, a return loss, b Gain and c Efficiency

Full size image
Fig. 11
figure 11

3D radiation design at frequency a 28 GHz, b 38 GHz.

Full size image
Fig. 12
figure 12

2D radiation shape at a 28 GHz, b 38 GHz.

Full size image
Fig. 13
figure 13

Surface current distribution of proposed antenna without defected ground plane at a 28 GHz, b 38 GHz. Surface current distribution of proposed antenna with defected ground plane. at c 28 GHz, d 38 GHz.

Full size image

4 MIMO antenna arrangement and mutual coupling

The construction of a 2-port MIMO antenna and the way to increase element separation are the main topics of this section. We offer a method of MC reduction with an EBG structure. The dimensions of the suggested design for the array that composed of two-port MIMO are 51 × 30 × 1.6 mm.

As seen in Fig. 14a–b, duplicates of the one antenna unit (Scheme II) from the preceding sector are placed 5 mm apart and perpendicularly to each other. The FR4 substrate is used to build the suggested antenna array. A CST 2019 model of the antenna array is used. Without the use of isolation measures, Fig. 15 illustrates the MC between the ports in a dual-port MIMO array. It is evident that the target band’s isolation is insufficient.

Fig. 14
figure 14

An orthogonal MIMO antenna geometry. a Top view, b Bottom view

Full size image
Fig. 15
figure 15

MC among ports without any Decoupling Methods

Full size image

4.1 Isolation enhancement using the EBG structure

One of the main objectives of 5G millimeter-wave applications is to increase MIMO characteristics, and the EBG structure is proposed to improve these properties. As seen in Fig. 16, the vertical EBG sheet in the MIMO antenna arrangement is made up of six rings that are integrated on the top side. The proposed MIMO structure has the following dimensions: 51 × 30 × 1.6 mm.

In order to design the suggested EBG, first, a Ring-shaped EBG unit cell is designed as displayed in Fig. 16c. The dimensions of the proposed EBG unit cell EBG structures of certain dimensions (\(\text{r}\)= 0.4 mm, \(\text{D}\)= 1.2 mm, \(\text{W}\)= 1.3 mm, and \(\text{g}\)= 0.05 mm). A standard 1.6 mm thickness FR4 material (ε = 4.3) is chosen as the substrate. The suggested EBG unit cell is analyzed by utilizing the CST. Figure 16d, presents the S- parameter (phase in degrees) of the Ring unit cell. It is shown that the phase shift between \(\pm {90}^{0}\) involving two resonance frequency 28 GHz and 38 GHz. Figure 16e shows the characteristics, such as permeability, refractive index, and effective permittivity. From Fig. 4, the permittivity is smaller than unity at the frequency band 20–40 GHz. Consequently, printing the array of unit cells, the antenna’s gain can be enhanced.

The MIMO antenna array is manufactured by the photolithographic process, as shown in Fig. 17, after being simulated using the commercial CST 2019. Figure 1818 shows the computed and observed characteristics of the suggested MIMO antenna structure.

As seen in Fig. 18a, the MIMO antenna construction functions as a millimeter-wave 5G antenna, spanning the spectrum from less than 25 to more than 40 GHz. Due to surface waves and space, there is a greater MC between the antennas because of their close proximity. EBG structures of certain dimensions (\(r\)= 0.4 mm, \(D\)= 1.2 mm, \({K}_{3}\)= 1.87 mm, and \({K}_{4}\)= 4.93 mm) are added in order to lessen this.

By efficiently capturing the waves and transforming them into surface currents, these components lessen MC. The isolation between the antennas is seen to vary over the working spectrum, from − 40 dB to − 91 dB, as Fig. 18b illustrates.

As seen in Fig. 18c, d, respectively, the system’s efficiency is around 95% and its gain fluctuates between 8.5 and 11.5 dBi along the operational band. As seen in Fig. 18e, f, the values of DG and ECC are around 10 and less than 1.8 × 10−4, respectively. Figure 18g shows that during the whole band region, the MIMO ‘s total active reflection coefficient (TARC) is less than − 30 dB. Figure 18h displays the CCL (channel capacity loss) of the proposed MIMO array with the EBG insertion.

The planned range of 0.3 bits/s/Hz is reached by the CCL. The little discrepancy among the simulated and measured values might be the result of a soldered connection. But the variance is still within the allowed variety. Figure 19 displays the surface current disseminations. Before adding EBG the current of antenna1 affected on antenna 2. After the EBG is installed, the current is engrossed around the structure of the EBG, causing the remaining antenna parts to sustain less damage and reducing MC.

Fig. 16
figure 16

An orthogonal MIMO antenna ‘s geometry with EBG structure. a Top view, b Bottom view, c EBG unit cell, d Simulated S- Parameter (phase in degrees) of the proposed ring unit cell, and e The characteristics of proposed Ring unit cell: refractive index (n), permittivity and permeability

Full size image
Fig. 17
figure 17

The fabricated Scheme of MIMO antenna ‘s geometry with EBG structure. a Top view, b Bottom view and c Measurement setup

Full size image
Fig. 18
figure 18

Simulated and measured performance of the planned MIMO structure. a Return loss b MC, c Efficiency, d Gain, e ECC, f DG, g TARC, and h CCL

Full size image
Fig. 19
figure 19

Surface current circulation of projected MIMO antenna without EBG for port 1 at a 28 GHz and b 38 GHz. And with EBG for port 1 at c 28 GHz and d 38 GHz.

Full size image

The radiation pattern in the E- and H-plane at 28 GHz and 38 GHz frequencies is displayed in Fig. 20. As seen in Fig. 20, the major lobes are oriented at phi = 90 degrees and are perpendicular to each other at 28 GHz and 38 GHz, correspondingly. The patterns of the E-plane are 90 degrees out of phase. On the other hand, all antennas exhibit the same E-plane patterns. This demonstrates the ability of the recommended antenna to offer form variety, guaranteeing reception free from interference.

Fig. 20
figure 20

Measurement setup: a schematic diagram, b antenna module in measurement. c Simulated and measured E-plane and H-plane radiation shape of planned MIMO antenna system after inserting EBG

Full size image

4.2 MIMO antenna performance evaluation

A number of factors are used to assess the MIMO antenna’s effectiveness, and the Envelope Correlation Coefficient (ECC) is one of the most important measures. One important dynamic feature used to measure MIMO antenna system performance is ECC. ECC may be determined using Eq. (5) depending on the S-parameters.

$$ECC=\frac{{\left|{S}_{11}*{S}_{12}+{S}_{21}*{S}_{22}\right|}^{2}}{\left(1-{\left|{S}_{11}\right|}^{2}-{\left|{S}_{21}\right|}^{2}\right)*(1-{\left|{S}_{22}\right|}^{2}-{\left|{S}_{12}\right|}^{2})}$$
(5)

In which the reflection coefficients at Ports 1 and 2 are signified by S11 and S22, individually. The Isolation and Insertion inefficiencies can be expressed by S12 and S21, accordingly.

Where the reflection coefficients at Ports 1 and 2 are denoted by S11 and S22, respectively. The Isolation and Insertion losses are represented by S12 and S21, respectively. The results of ECC using the PE approach are roughly less than \({1.8\times 10}^{-4}\), as shown in Fig. 18f. The transmission power loss in the MIMO scheme while utilizing a diversity mechanism is measured using a metric called Diversity Gain (DG). The following equation from reference [5] can be used to calculate DG:

$$DG=10\sqrt{1-{\left(ECC\right)}^{2}}$$
(6)

Figure 18e indicates that DG is around 10 dB.

An additional metric that evaluates the coupling between ports and confirms a MIMO antenna’s functionality is TARC (Total Active Reflection Coefficient). The square root of the total reflected power divided by the square root of the entire incident power is known as TARC. TARC may be computed using Eq. (7) as follows:

$$\varGamma =\sqrt{\frac{{\left|{s}_{11}+{s}_{12}{e}^{j\theta }\right|}^{2}+{\left|{s}_{21}+{s}_{22}{e}^{j\theta }\right|}^{2}}{2}}$$
(7)

θ marks the phase angle of input. which ranges from 0 to 180 degrees with a step increment of 30 degrees. Ports 1 and 2’s reflection coefficients are denoted by S11 and S22, individually. The TARC values are shown in Fig. 18g.

The fourth performance measure, channel capacity loss (CCL), is a vital indicator to assess MIMO systems. The Eq. (8) that follows is used to compute it:

$$C loss=-{log}_{2}\left({{\Psi }}^{R}\right)$$
(8)

where \({{\Psi }}^{\text{R}}\), which may be expressed as follows, is the receiving antenna’s correlation matrix:

$${{\Psi }}^{R}=\begin{array}{cc}{{\Psi }}_{ii}& {{\Psi }}_{ij}\\ {{\Psi }}_{ji}& {{\Psi }}_{jj}\end{array}$$
(9)

where,

$${{\Psi }}_{ii}=1-{(\left|{S}_{ii}\right|}^{2}-{\left|{S}_{ji}\right|}^{2})$$
(10)
$${{\Psi }}_{ij}=-\left({S}_{ii}*{S}_{ji}+{S}_{ij}*{S}_{jj}\right)$$
(11)
$${{\Psi }}_{ji}=-\left({S}_{jj}*{S}_{ji}+{S}_{ij}*{S}_{ii}\right)$$
(12)
$${{\Psi }}_{jj}=1-{(\left|{S}_{jj}\right|}^{2}-{\left|{S}_{ij}\right|}^{2})$$
(13)

To achieve the needed criterion, as shown in Fig. 18h, CCL must be less than 0.4 bits/s/Hz. Because our plan is less than this amount, it meets CCL.

4.3 Evaluation of the antenna achievement in time domain

The antenna’s time-domain presentation is studied in detail, including Group Delay (GD), S21 phase, and forward transmission coefficient (S21). The simulated time-domain conformation is shown in Fig. 21 in three different configurations: side-to-side, face-to-side, and face-to-face. Two identical antennas are situated 36 mm apart in each of these setups. In these configurations, one antenna functions as the transmitter (Tx) and the other as the receiver (Rx).

The GD, S21 phases, and S21 magnitudes are displayed in Fig. 22. Figure 22a makes it evident that the antenna configurations facing each other and side by side have S21 values below − 40 dB at 28/38 GHz, whereas the configuration facing each other has an S21 value over − 50 dB at 28 GHz and reaching − 60 dB at 38 GHz.

In Fig. 22b, the S21 phase is also displayed to help understand the linearity characteristics within the intended band procedure. It is obvious from the phase curves that the antenna behaves linearly in the desired frequency range in a variety of orientations. Figure 22c presents the GD results to support this observation. At 28 GHz and 38 GHz, the GD is around 0.24 ns and 0.15 ns, respectively.

Fig. 21
figure 21

Configuration for time-domain analysis in different simulation setups

Full size image
Fig. 22
figure 22

An examination of the suggested antenna’s time-domain capability. a Transmission forward coefficient, b S21 phase, and c GD.

Full size image

5 Evaluation in light of related works

Although wideband-frequency band MIMO antennas with excellent isolation have been produced, it is difficult to give up compact dimensions due to design factors including low profile and enhanced gain. This research presents the introduction of a 5G Millimeter-Wave wide-band MIMO antenna array. The 5G Millimeter-Wave wireless terminal is intended for use with the suggested structure. Regarding its small size, number of ports, frequency range, MC methods, MC, gain, efficiency, ECC, and DG, the suggested MIMO antenna array stands out for its exceptional qualities and performance. When compared to related work, its performance is better. Table 4 displays a comparative evaluation of the MIMO antenna topologies that were given, highlighting their respective competitive advantages.

Table 4 The performance assessment by contrasting the suggested method with earlier research initiatives
Full size table

6 Conclusion

We have presented a two-port MIMO array in our work that stands out for having high port isolation. In order to reduce MC between the antennas, we used the EBG technology, which we have carefully assessed for efficacy. We ran simulations with CST version 2019 to maximize the radiating components’ size. The proposed MIMO antenna achieved remarkable isolation values of – 90 dB throughout the whole spectrum, notably at 38 GHz. It was specially intended to work at double tone frequencies, 28/38 GHz.

The proposed MIMO antenna achieved remarkable isolation values of – 90 dB throughout the whole spectrum, notably at 38 GHz. It was specifically planned to work at two resonance frequencies, 28/38 GHz. Because more capacitance was added to the circuit, the EBG approach was essential in improving the isolation between antennas. As a result, it demonstrated strong DG, a high efficiency rating of 91%, and a top gain of 11.5 dBi, demonstrating the technique’s effectiveness.

To fully assess our antenna design’s performance in terms of diversity and highlight the exceptional quality of the suggested two-port MIMO antenna, we took a number of MIMO parameters out of the measurement and simulation data, such as ECC, DG, and CCL. Positive trends were regularly observed in the findings of these investigations inside the two working frequency ranges. This infers that our suggested antenna construction has potential for use in millimeter-wave applications for 5G.