A Sub-6GHz Two-Port Crescent MIMO Array Antenna for 5G Applications

Abstract

The fifth generation of wireless communication (5G) technology is becoming more innovative with the increasing need for high data rates because of the incremental rapidity of mobile data growth. In 5G systems, enhancing device-to-device communication, ultra-low latency (1 ms), outstanding dependability, significant flexibility, and data throughput (up to 20 Gbps) is considered one of the most essential factors for wireless networks. To meet these objectives, a sub-6 5G wideband multiple-input multiple-output (MIMO) array microstrip antenna for 5G Worldwide Interoperability for Microwave Access (WiMAX) applications on hotspot devices has been proposed in this research. The 1 × 4 MIMO array radiating element antenna with a partial ground proposed in this research complies with the 5G application standard set out by the Federal Communications Commission. The planned antenna configuration consists of a hollow, regular circular stub patch antenna shaped like a crescent with a rectangular defect at the top of the patch. The suggested structure is mounted on an FR-4 substrate with a thickness “h” of 1.6, a permittivity “${\epsilon }_r$” of 4.4, and a tangential loss of 0.02. The proposed antenna achieves a high radiation gain and offers a frequency spectrum bandwidth of 3.01 GHz to 6.5 GHz, covering two 5G resonant frequencies “$f_r$” of 3.5 and 5.8 GHz as the mid-band, which yields a gain of 7.66 dBi and 7.84 dBi, respectively. MIMO antenna parameters are examined and introduced to assess the system’s performance. Beneficial results are obtained, with the channel capacity loss (CCL) tending to 0.2 bit/s/Hz throughout the operating frequency band, the envelope correlation coefficient (ECC) yielding 0.02, a mean effective gain (MEG) of less than −6 dB over the operating frequency band, and a total active reflection coefficient (TARC) of less than −10 dB; the radiation efficiency is equal to 71.5%, maintaining impedance matching as well as good mutual coupling among the adjacent parameters. The suggested antenna has been implemented and experimentally tested using the 5G system Open Air Interface (OAI) platform, which operates at sub-6 GHz, yielding −67 dBm for the received signal strength indicator (RSSI), and superior frequency stability, precision, and reproducibility for the signal-to-interference-plus-noise ratio (SINR) and a high level of positivity in the power headroom report (PHR) 5G system performance report, confirming its operational effectiveness in 5G WiMAX (Worldwide Interoperability for Microwave Access) application.

1. Introduction

The wireless network is suggested from the point of view of the Internet of Things (IoT), which offers higher data rates, more reliability, and broader network coverage when utilizing new generations of wireless communications. It can also handle a great deal of gigabits of data in seconds. These 5G systems present increasing usage of wireless and smart devices; this new network is rapidly evolving from a plan into a significant market demand [1,2,3]. By concentrating on the antennas employed in 5G applications supported in wireless communication systems, 5G communication differs from earlier protocols in several performance parameters. Regarding latency, data rate, energy efficiency of networks, and connection density, the 5G protocol offers a benefit [4,5,6,7]. In addition, more precise communication methods covered by the spectrum have been altered compared to its previous protocols to achieve these aims [8,9,10,11]. Another issue with the additional equipment needed for 5G is finding areas that can support more network towers and stations, which becomes difficult due to the rising number of access points. Another factor to consider is the aesthetic impact of these base stations, because it makes sense to utilize the existing infrastructure [12,13,14,15]. Sub-1 GHz as a low-band, sub-6 GHz as a mid-band, and the mm-waves as a high band are the three frequency bands that comprise the 5G spectrum [16,17,18]. While 6 GHz is a supported and licensed spectrum and serves as the primary classes’ separation point, the sub-6 GHz range is chosen to be studied. The operation of this band is intended to meet performance requirements incompatible with sub-6 GHz designs [19,20].

The envelope correlation coefficient (ECC) of sub-6 GHz, gain, and efficiency are all attained for 5G MIMO antennas [21], making them suitable for mobile devices and applications of vehicles to everything (V2X). In [22], the research presents a wideband inverted-F antenna. In [23], a small, multiband monopole antenna at sub-6 GHz with a multiband antenna that supports 5G is proposed for a vehicle rooftop. A compact antenna design for automotive use that enables 4-channel MIMO at sub-6 for 5G applications is presented in [24], resulting in acceptable compromises for MIMO performance, VSWR, pattern, and gain. In [25], a small wideband printed antenna with a rectangular slot is partially grounded. The design strategy utilizing a partially slotted ground is significantly more adaptable for enhancing antenna performance.

The Nyquist–Shannon theorem estimates single-input, single-output (SISO) channel capacity for both the transmitter and receiver sides. The MIMO wireless technology is used nowadays and has become the most powerful wireless technology with equal antenna spacing for line-of-sight geometry, which presents a more significant channel capacity while keeping equal spacing between antennas as a distance function compared with SISO systems [26]. MIMO antenna applications are preferred in 5G communication systems because of their high data rates, low latency, and several connected nodes [27,28]. The proper size, intense isolation, reflection coefficient (S11) of less than 20 dB, ECC of less than 0.5, and gain in the range of 3 dB to 12 dB are the main characteristics of the 5G MIMO antenna [29]. The dual-polarization technique is used to minimize the size of MIMO antennas to half compared with the single-polarization technique, avoiding the large sizing of MIMO antennas that is required to enhance channel capacity and add additional antennas at the transmitter and receiver [30,31]. It uses electromagnetic band gap (EBG) structures [32] to provide high MIMO isolation; even so, the design is complex and has a negative impact on the radiation pattern [33].

These techniques depend on extra-structural decoupling, which increases the MIMO array’s design footprint and complexity to a great extent and may even have an adverse effect on radiation’s far-field qualities. Antenna array self-decoupling has been the subject of more recent research [34]. Rather than relying on outside structures for decoupling, this strategy minimizes coupling the antenna’s elements mutually. In [35], inverted-F antennas and dipole antennas are examined in the presented approach by exciting them simultaneously, which can eliminate the mutual coupling effect. In [36], the two-port MIMO antenna exhibits a mutual coupling value of approximately −18 dB, which is not within an acceptable range. A common ground technique paired with a stub is employed to enhance this mutual coupling, achieving a lower mutual coupling of −27 dB. In [37], two methods are presented for estimating the angle of arrival (AoA) of multiple sources affected by mutual coupling. The first method is designed for situations with many coupling parameters and formulates an optimization problem to minimize a specific cost function while using constraints to prevent false peaks. The second method enhances the initial algorithm for better AoA performance but complicates the issue by introducing multi-dimensional challenges. In contrast, the first method uses alternating minimizations to streamline the solution process. In [38], a practical implementation of massive MIMO is showcased, emphasizing its potential to improve spectral efficiency through adaptive electronic beam control in its elevation and azimuth dimensions. In [39], a 12-port massive MIMO system for future 5G and sub-6 GHz applications is investigated. All twelve antenna elements are integrated effectively, achieving good reflection coefficients of 6 dB and acceptable isolation levels exceeding 12 dB. However, the total efficiency is approximately 40%.

In [40], a microstrip patch antenna (MPA) with a star shape offering a radiation pattern as a conical shape is simple in design, operating at a narrow-band low frequency and offering 3 dB gain. The inset-fed patch antenna uses the field’s coupling to limit its opposition from the radiating antenna element [41]. In [42], a monopole single-element antenna design was the foundation for fabricating a compact four-element array of λ/4-spaced with a 41.8% bandwidth (BW). Nevertheless, there was a two dBi gain limit and significant cross-polarization to provide a patch antenna design with dual operating frequency bands. Ref. [43] provides a ground-defect structure for usage in sub-6 GHz 5G applications, designing a 2 × 2 antenna aimed at impedance of BW enhancement.

In [44], for 5G applications, a wideband antenna design proposes a microstrip array with FSS that operates at 2.3 GHz and 12.4 dBi. In [45], for sub-6 GHz, a 4 × 4 antenna design provides an L-shaped strip that operates at 0.2 GHz and 5.1 dBi. In [46], an M-shaped strip is provided that operates at 0.26 GHz and 1.6 dBi for mobile applications. In [47], an M-shaped strip is proposed that operates at 0.2 GHz and 4.5 dBi for mobile applications. In [48], a 4 × 4 MIMO system proposes a printed Yagi–Uda technique that operates at 0.3 GHz and five dBi. In [49], a 4 × 4 MIMO design provides two operating frequency bands using the circular quasi-Yagi technique at 1.1 GHz and 7 dBi. In [50], a MIMO array antenna design provides a self-decoupling technique that operates at 4 GHz and 3 dBi, and Ref. [51] offers a microstrip MIMO technique that operates at 0.2 GHz and is used for 5G smartphone applications.

The proposed work presents a novel approach to designing wideband MIMO array antennas that are cost-effective and of proper size. These antennas are intended to serve WiMAX 5G applications and operate at sub-6 GHz, offering high efficiency, high gain, and effective measured MIMO parameters. Our contribution can be summed up as follows:

• The proposed MIMO array antenna was designed based on designing the unit cell antenna of a hollow regular circular stub patch antenna in the shape of a crescent and a rectangular defective shape at the upper part of the patch using Microwave Studio CST for simulation.
• A 1 × 2 array and a 1 × 4 array were designed based on a unit cell and lead to high gain, and to achieve wideband beam width, the proposed MIMO array antenna was designed to cover 3.5 GHz and 5.8 GHz for WiMAX 5G systems.
• A MIMO array antenna was fabricated and measured in the Microstrip Circuits Department Laboratories at the Electronics Research Institute (ERI), which nearly matched the simulated results.
• The MIMO array antenna presented a fractional frequency band of 3.01 GHz to 6.51 GHz, which was noticeably more meticulous than in previous research. In the interim, the efficiency went to 71.5%, and a high gain of 7.84 dBi and 7.66 dBi was obtained at 5.8 GHz and 3.5 GHz, respectively. Effective measured MIMO parameters were CCL tending to 0.2 bit/s/Hz, the ECC yielding 0.02, a MEG of less than −6 dB over the operating band, and a TARC of less than −10 dB.
• The fabricated MIMO array antenna was experimentally tested using the 5G system built on the OAI platform, which operates at sub-6 GHz, confirming its operational effectiveness in 5G applications over the MIMO dipole antenna.

This paper’s organization is as follows: Section 1 demonstrates an introduction to 5G and sub-6 GHz applications and provides a literature review of previous works. Section 2 introduces a simulated parametric study of the proposed antenna and design procedures, starting with a single-patch radiating element, 1 × 2 and 1 × 4 array antennas, and reaching the proposed MIMO array design. The results of the fabrication of the proposed MIMO array on FR-4 substrate are verified and compared with the simulated results in Section 3. Section 4 discusses the implementation of the proposed MIMO array antenna on a 5G system and a performance evaluation. In Section 5, the paper’s conclusion is provided.

2. The Antenna Design and Configuration

A low-cost FR-4 substrate, which had a 1.6 mm thickness, a 0.02 loss tangent, and a 4.4 relative permittivity, was used to construct the proposed MIMO array antenna design, which operates at sub-6 GHz. Microwave Studio CST simulated the design between 2.84 GHz and 6.33 GHz. The CST software 2019 measured the proposed antenna’s principal features, such as the reflection coefficient, frequency response, polarization, BW, and gain.

The MIMO antenna array design methodology passed four main steps to reach the proposed design with the best simulated and fabricated results. First is the unit cell design, second is the 1 × 2 array design, third is the 1 × 4 array antenna design, and finally, as a fourth step, is the MIMO array design, considered the proposed design.

2.1. Design of the Single-Patch Radiating Antenna Element

The single-patch schematic diagram is shown in Figure 1. The single patch relies on a hollow, regular, circular stub patch antenna in the shape of a crescent with a cut-out circle with a rectangular defective shape at the top of the patch. The cutting of a regular circular stub essentially enhances the antenna’s reflection coefficient and the aperture area with an impedance matching 50 ohms. A standard methodology is used to design the circular patch antenna. Equation (1) in [50] can be used to compute the radius r.

$$r={\displaystyle \frac{\mathrm{F}}{\sqrt{1+\frac{2\mathrm{h}}{\pi F{\epsilon }_r}\left[\mathrm{l}\mathrm{n}\right(\frac{\pi F}{2h})+1.7726]}}}$$

(1)

where $F=(8.791\times {10}^9)/f_r\sqrt{{\epsilon }_r}$, while $h,$ ${\epsilon }_r$, and $f_r$ denote the configuration’s thickness, relative permittivity, and resonance frequency. The low-cost FR4 substrate with a 1.6 mm thickness, a 4.4 relative permittivity, and a 0.02 tangent loss is the fundamental structure for the proposed single-patch element design. The design simulation and performance optimization were completed using “CST Microwave Studio”.

Table 1. The single-antenna dimensions.

Antenna Parameter	Dimension (mm)	Antenna Parameter	Dimension (mm)	Antenna Parameter	Dimension (mm)
Ls (single)	46	Lg	15	t	1.45
Ws (single)	30	Wc	7	Lt	2
Lf	16.65	r1	12	Wf	2.91
Lc	5.113	r2	6

shows the suggested single-element design and its actual dimensions.

Figure 2 shows the simulation results of the S-parameters that are essential for understanding the input–output relationships between ports in any electrical system. In a multi-port network, the parameter SNM represents the power transferred from Port M to Port N, making it crucial for analyzing system performance. An S11 value of 0 dB indicates that all power is reflected from the antenna, resulting in no radiation. On the other hand, an S11 value of −10 dB reveals that when 3 dB of power is supplied to the antenna, −7 dB is reflected power. Consequently, the remaining power is either accepted by the antenna or lost. A well-designed antenna should minimize these losses to ensure most of the delivered power is effectively radiated.

Furthermore, understanding the voltage standing wave ratio (VSWR) is vital, as it correlates directly with S11 and is a key indicator of antenna performance. The reflected power from the antenna, represented by S11, plays a significant role in maximizing efficiency, making it a critical factor in successful antenna design. The amount of power reflected from the antenna (S11) curve of the unit cell antenna starts at 2.84 GHz and rises to 6.33 GHz, with a BW of 3.49 GHz, which covers 3.5 GHz and 5.8 GHz. Figure 3 presents the gain over frequency. It is up to 2 dBi over the operating frequency range, with the maximum value reaching 4.9 dBi at 5.77 GHz. Figure 4a and Figure 4b show the radiation pattern at two operating frequencies, 3.5 GHz and 5.8 GHz, respectively, with the omnidirectional figure-eight shape reaching its maximum value at 3.35 dB at 183° and 5.2 dB at 150° at 3.5 GHz and 5.8 GHz, respectively. The slightly unsymmetric radiation between the back and broadside radiation, as shown in Figure 4a, refers to the partial ground design necessary to increase BW in our design. When fabricating a microstrip antenna (MSA) on a substrate with a partial ground plane, surface waves propagate along the ground plane until they encounter its edges. The waves reflect and diffract at these edges, resulting in slightly higher back radiation patterns than broadside radiation.

The optimized single-patch antenna methodology starts with iteration (1) in Figure 5a as a conditional microstrip patch radiating antenna in a circular configuration that operates from 2 GHz to 5.30 GHz. The traditional central circle is cut off in iteration (2), as shown in Figure 5b. In Figure 5c, a tapering inverted triangle section is added at the top of the feed line as iteration (3) to achieve a wider BW from 2 GHz to 5.47 GHz. A rectangular shape is served from the top of the conditional antenna element and added to the cut circular shape created at iteration (4) to promote the antenna’s impedance matching, as seen in Figure 5d. In Figure 5e, iteration (5) presents the proposed patch antenna by hollowing a regular circular stub with a crescent shape, leading to a wide radiating range, improved reflection coefficient, and matching antenna performance. The impedance BW for the –10 dB reference value is extended from 2.84 GHz to 6.33 GHz while keeping the antenna’s footprint size, as demonstrated in Figure 2. Utilizing a wide BW achieved by a partial ground antenna design of Lg = 34.6 mm, Figure 6 shows the evolution of the s11 curve reflection coefficient in five iterations to the proposed unit cell antenna realization, reaching the most appropriate design with a maximum frequency range of 2.84 GHz to 6.33 GHz under −10 dB and with matching impedance, guaranteeing high gain.

The performance of the proposed antenna depends on several factors. A parametric analysis is conducted to optimize the suggested single-cell antenna. Figure 7a illustrates how the width of the rectangular cutout W_c promotes the antenna’s impedance matching and affects the antenna’s reflection coefficient response. The impedance matching and stability can be improved by expanding the cutout rectangle width from 5 mm to 7 mm. A width of 7 mm is the best, with S11 starting at 2.83 GHz and ranging up to 6.36 GHz, compared with 5 mm, which is near −10 dB.

Nevertheless, a 9 mm width reduces the antenna’s BW. Figure 7b shows the S11 response affected by the length variation of the rectangular cutout Lc; 5.11 mm, which is a moderate length, achieves more matching impedance with a broader-frequency BW starting at 2.83 GHz compared with 3.11 mm starting at 3.2 GHz or 7.11 mm, which is near −10 dB. Figure 7c shows that increasing the radius of the circular cutout shape r₂ by increasing the cutout radius from 5 mm to 6 mm leads to a wider BW, but by increasing it to 7 mm, the S11 response becomes more rippled up to −10 dB at 5.7 GHz and splits the wide band into two separated bands. Figure 7d shows that the width of the outer circular shape is “t”. When the width is 0.45 mm, the impedance match is over −10 dB, giving two narrow bands, 2.7 GHz to 3.5 GHz and 5 GHz to 6 GHz. The matching impedance drops to −13.5 dB when the width is increased to 1.45 mm, and the BW widens. That is better than raising the width to 2.45 mm, which presents a more stable response under −10 dB. Increasing the width to 2.45 mm yields a band that starts from 3 GHz to 5.5 GHz, so 1.45 mm is the optimum width.

2.2. Design of Antenna Array Arrangement

The radiating antenna element with a single patch builds array patches of 1 × 2 and 1 × 4 to achieve a wider band of impedance bandwidth. Next, a 1 × 4 array structure antenna is combined with a reflecting one to construct a MIMO array antenna to pick up high gain and large BW. Utilizing the mathematical equations of the microstrip antenna as a basis for design, the 1 × 2 and 1 × 4 array antennas are built. The hollow, regular circular stub patch antenna in the crescent shape is developed at the 3 GHz resonant operating frequency. To upgrade the antenna’s performance, a 1 × 2 array arrangement is constructed that comprises two hollow regular crescent circular antennas connected to a transformer via a dual-direction power connector splitter utilizing a 1/4 impedance, as seen in Figure 8.

Table 2. The proposed 1 × 2 array antenna dimensions.

Antenna Parameter	Dimension (mm)	Antenna Parameter	Dimension (mm)	Antenna Parameter	Dimension (mm)
Ls (1 × 2 array)	66	Lg (1 × 2 Array)	26.33	W2	9.36
Ws (1 × 2 array)	66	Lp	16.65	W3	2
Wf (1 × 2 array)	2.91	W1	10	d	31.63

shows the exact dimensions in mm of the 1 × 2 array antenna design, accompanied by its geometric size, which can be described as the substrate’s height (h), width (Ws), and length (Ls) of 66 × 66 × 1.6 mm³.

Figure 9 shows the simulation results of the reflection coefficient of the 1 × 2 array antenna starting at 2.83 GHz and ranging up to 6.36 GHz, with a 3.53 GHz frequency band covering 3.5 GHz and 5.8 GHz. Figure 10 shows the gain over frequency up to 3 dBi over the operating frequency range, with the maximum value reaching 8.1 dBi at 5.9 GHz. Figure 11a and Figure 11b show the radiation pattern at two operating frequencies, 3.5 GHz and 5.8 GHz, respectively, with the omnidirectional eight of the figure-eight shape reaching its maximum value at 4.42 dB at 183° and 9.87 dB at 10° at 3.5 GHz and 5.8 GHz, respectively.

In the meantime, as seen in Figure 12, a 1 × 4 array antenna was constructed using an FR-4 substrate and quadruple patches of radiated elements joined to a quadruple-way power divider to increase antenna gain. In the 1 × 4 array antenna design, half of the wavelength (λ/2) distance between the two patches is kept. The actual dimensions of the 1 × 4 array antenna design are 140 × 66 × 1.6 mm³, as shown in

Table 3. The proposed 1 × 4 array antenna dimensions.

Antenna Parameter	Dimension (mm)	Antenna Parameter	Dimension (mm)	Antenna Parameter	Dimension (mm)
Ls (1 × 4 array)	66	Lf (1 × 4 Array)	11.84	W5	17.91
Ws (1 × 4 array)	140	Wf	2.91	W6	10
Lg (1 × 4 array)	34.6	W4	17.91	L	12.29

.

Figure 13 shows the simulation results of the reflection coefficient of the 1 × 4 array starting at 3.01 GHz and ranging up to 6.51 GHz with a BW of 3.5 GHz, which covers 3.5 GHz and 5.8 GHz. The gain over frequency is presented in Figure 14 and is up to 4 dBi over the operating frequency range, with the maximum value reaching 9.259 dBi at 5.7 GHz. Figure 15a and Figure 15b show the radiation pattern at two operating frequencies, 3.5 GHz and 5.8 GHz, with the omnidirectional figure-eight shape reaching its maximum value at 8.82 dB at 200° and 9.90 dB at 31° at 3.5 GHz and 5.8 GHz, respectively.

2.3. Design of MIMO Array Antenna Arrangement

The proposed MIMO antenna array, shown in Figure 16, has a geometric size of 140 × 135 × 1.6 mm³ and a reflection coefficient of the radiated antenna from 3.01 GHz to 6.5 GHz at 3.5 and 5.8 GHz resonant frequencies. The gain reaches 7.66 dBi and 7.84 dBi, respectively, and a 50 ohm impedance feedline. The array antenna air gap is S = λ/4 and S = 16.88 mm. The proposed design is paramount in offering mutual coupling reduction among the opposite neighboring antennas, allowing a MIMO system that reduces the form factor. The ground of the two opposite 1 × 4 arrays is linked, forming a connected ground using a 1 mm thin copper line in the middle of the ground added to a 2 mm thin glazed line of copper that aids in surface mitigation wave propagation and reduces undesired mutual coupling among the two 1 × 4 radiating patches. It is vital to subside the formation of a height conceivable at the antenna’s port excitation caused by standing waves of induced voltage that might damage antenna connectors.

Figure 17a shows the reflection coefficients of the S11 and S22 simulations at a wide operating range starting at 3 GHz and ranging up to 6.5 GHz, with a BW of 3.5 GHz. Figure 17b shows the transmission coefficients of the power transferred from Port 2 to Port 1 (S12) and the S21 simulation results under −20 dB over the operating range.

Figure 18a and Figure 18b show the MIMO radiation pattern at 3.5 GHz and 5.8 GHz, with the omnidirectional figure-eight shape due to it being partially grounded, reaching its maximum value at 7.7 dB at 193° and 7.42 dB at 12° at 3.5 GHz and 5.8 GHz, respectively. Figure 19 shows the gain and wideband MIMO array antenna that tends to 7.84 dBi with a good deal of matching impedance and minimizing the antenna elements’ mutual coupling.

The proposed MIMO array fabricated antenna design has A 65.8 mm gap among the radiating elements’ ports. The distance between two opposite radiating patch elements is 7.266 mm. The opposing 1 × 4 array antenna is achieved by self-decoupling. The proposed MIMO array antenna design is shown in

Table 4. The proposed MIMO array antenna dimensions.

Antenna Parameter	Dimension (mm)	Antenna Parameter	Dimension (mm)	Antenna Parameter	Dimension (mm)
Ls (MIMO)	135	Lg2	6.5	Wg3	1
Ws (MIMO)	140	Wg1	1
Lg1	14.7	Wg2	12

.

3. Measurement Results Discussion

A low-cost substrate made of FR-4 was utilized to build the proposed MIMO array design, as demonstrated in Figure 20, with a geometric size of 140 × 135 × 1.6 mm³ for the physical dimensions. FR-4 substrates are dependable for low-frequency applications (below 6 GHz) because of their low-loss dielectric properties. They are cost-effective, thanks to affordable materials and efficient processing. Their mechanical robustness ensures durability, while easy fabrication enhances production efficiency, making FR4 a reliable choice. The antenna measurements were propagated to the receiving side when the transmitted RF signals were collected.

The experiment was conducted practically in a room with anechoic properties to assess the antenna’s gain and radiation pattern behavior measurements, as seen in Figure 21. S11 was adjusted by the analyzer’s vector in the network. Furthermore, the equation, namely, Friis’ [44], was used to calculate the gain.

3.1. Reflection and Transmission Coefficient Results

A wideband MIMO array antenna was fabricated, and the reflection coefficients S11 and S22 are shown in Figure 22a and Figure 22b, respectively. They start from 2.9 GHz and range up to 6.1 GHz, with a BW of 3.2 GHz, which covers 3.5 GHz and 5.8 GHz and is acceptable compared with simulation results. The transmission coefficient responses S12 and S21 of the MIMO array antenna shown in Figure 22c,d lie under −20 dB over the operating range, which is the same as the simulation results.

The analyzer’s vector network was used to measure the results, and simulations were obtained using CST software. Simultaneously, the sub-6 GHz band was covered by the reflection coefficient response derived from the wideband MIMO-array antenna simulation and measurement; the simulated and measured frequency bands covered by the MIMO array antenna’s reflection and transmitted coefficient responses were very similar.

3.2. Radiation Pattern and Realized Gain Results

The radiation pattern and the gain were measured in the Microwave Research Laboratory at the Microstrip Circuits Department (NRC), National Research Center, Egypt. A horn transmitting antenna (Tx) was used to demonstrate the transmitting antenna and test the gains of the fabricated antenna’s design. On the other hand, the proposed antenna acted as a receiving antenna (Rx) and maintained a distance of one and a half meters between the Tx and the Rx. The simulated and measured radiation patterns were very similar to each other.

The MIMO array antenna’s simulated and measured radiation pattern is presented in Figure 23a,b. The fabricated antenna’s radiation pattern was measured at two operating frequencies, 3.5 GHz and 5.8 GHz, with the omnidirectional figure-eight shape reaching its maximum value at 7.7 dB at 192° and 7.43 dB at 12° at 3.5 GHz and 5.8 GHz, respectively, which is close to the simulation results.

The realized gain for the MIMO array antenna is presented in Figure 24 as the simulated and measured comparison results. The fabricated antenna’s measured gain was 7.66 dBi and 7.54 dBi at 3.5 GHz and 5.8 GHz, respectively, which are very close to each other. Furthermore, the variation in impedance matching at different operating range frequencies related to the fluctuation in gain. The MIMO array antenna attained a maximum efficiency of 71.5% for the radiation efficiency response, indicating that the proposed antenna is appealing for antenna technology improvement.

3.3. MIMO Antenna Parameter Results

It is observable that the dependency of the latest wireless communication systems on MIMO antenna structures increased system performance. Isolation is a must to reduce the lack of an external decoupling network and guarantee MIMO performance. The following effective parameters that control MIMO’s performance should be considered [52,53,54].

3.3.1. The Envelope Correlation Coefficients

The ECC parameter measures the variation of the diversity factor, which reveals the correlation among antennas in the MIMO system. The ECC can be calculated as a function of S-parameters, as in Equation (2). Figure 25a demonstrates the comparison between simulation and measurement, showing that both are similar and that both yielded 0.02, which is a good achievement, as it tended to zero and led to an effective antenna.

$${\mathsf{\rho }}_{eij}={\displaystyle \frac{{S^{*}}_{ii}{S^{*}}_{ij}+{S^{*}}_{ji}{S^{*}}_{jj} }{(1-{\left|{\mathrm{S}}_{ii}\right|}^2-{\left|{\mathrm{S}}_{ij}\right|}^2)(1-{\left|{\mathrm{S}}_{ji}\right|}^2-{\left|{\mathrm{S}}_{jj}\right|}^2)}}.$$

(2)

3.3.2. The Diversity Gain

The DG is a performance measurement of a MIMO antenna’s quality and reliability. The DG comes from the ECC, which is good when it nears 10 dB over the operating frequency band. The DG can be calculated, as evident in Equation (3). Comparing the simulation and measurement, Figure 25b shows that both were comparable and yielded 10 dB, which resulted in an effective antenna.

$$\mathrm{DG}=10 \sqrt{1-{\left|{\mathsf{\rho }}_{eij}\right|}^2}.$$

(3)

3.3.3. The Channel Capacity Loss

The CCL refers to the practical minimum received losses of the MIMO system; it should be less than 0.4 bit/s/Hz to be an acceptable standard loss. The CCL can be estimated using Equations (4)–(9). Figure 25c demonstrates the comparison of the CCL between the simulation result, which tended to be 0.2 bit/s/Hz, and the measurement result, which was less than 0.2 bit/s/Hz over the operating frequency band, which is better than the simulated results.

$$\mathrm{CCL}=-{\mathrm{l}\mathrm{o}\mathrm{g}}_2\left(\det \left(a\right)\right).$$

(4)

$$\mathrm{a}=\left|\begin{array}{cc}{\mathsf{\xi }}_{11}& {\mathsf{\xi }}_{12}\\ {\mathsf{\xi }}_{21}& {\mathsf{\xi }}_{22}\end{array}\right|,$$

(5)

$${\mathsf{\xi }}_{11}=1-\left({\left|{\mathrm{S}}_{11}\right|}^2+{\left|{\mathrm{S}}_{12}\right|}^2\right),$$

(6)

$${\mathsf{\xi }}_{12}=-\left({S^{*}}_{11}{\mathrm{S}}_{12}+{S^{*}}_{21}{\mathrm{S}}_{12}\right),$$

(7)

$${\mathsf{\xi }}_{21}=-\left({S^{*}}_{22}{S}_{21}+{S^{*}}_{12}{\mathrm{S}}_{21}\right),$$

(8)

$${\mathsf{\xi }}_{22}=1-\left({\left|{\mathrm{S}}_{22}\right|}^2+{\left|{\mathrm{S}}_{21}\right|}^2\right),$$

(9)

3.3.4. The Mean Effective Gain

The MEG tends to equal the total adequate average power received compared to the total average power received by an isotropic antenna. The MEG can be calculated from Equation (10). The MEG should be less than −3 dB to be an acceptable standard for MIMO performance. The comparison between the simulation result, typically −6.9 dB, and the measurement result, which was less than −6 dB over the working frequency band, is shown in Figure 25d. These results are almost identical, which means the MIMO system is effective.

$${\mathrm{M}\mathrm{E}\mathrm{G}}_i=\left(1-\sum _{j=1}^M{\left|{\mathrm{S}}_{ij}\right|}^2\right).$$

(10)

3.3.5. Total Active Reflection Coefficient

The total active reflection coefficient (TARC) is the square root of the ratio of the total power reflected to the incident power. The TARC can be calculated from Equation (11). The TARC should lie under −10 dB, which means the antenna design is effective. Figure 25e compares the simulation and measurement: Both were similar and less than −10 dB.

$$TARC=\sqrt{{\displaystyle \frac{{\left|\left({{\mathrm{S}}_{11}+\mathrm{S}}_{12 }{e}^{j\theta }\right)\right|}^2+{\left|\left({{\mathrm{S}}_{21}+\mathrm{S}}_{22 }{e}^{j\theta }\right)\right|}^2}{2}}}.$$

(11)

The proposed MIMO array antenna stands in comparison with previous works regarding the design area, frequency band, gain, ECC, target, and design technique, as shown in

Table 5. A comparison between the proposed MIMO array antenna and previous works.

Ref. No.	Area	Freq. Band	Gain	ECC	Substrate	Target	Technique
44	136 × 55	2.3	12.4	-	Fr-4	-	Microstrip array with FSS
45	150 × 75	0.2	5.1	0.01	Fr-4	5G	L-shaped strip
46	150 × 75	0.26	1.6	0.3	Fr-4	5G	M-shaped strip
47	145 × 75	0.2	4.5	0.16	Fr-4	5G	M-shaped strip
48	154 × 154	0.3	5		Fr-4	NM	Printed Yagi–Uda
49	263 × 263	1.1	7	0.159	Fr-4	GSM/UMTS/EDGE	Circular quasi-Yagi
50	80 × 80	4	3	0.016	Fr-4	5G/Sub-6GHz	Self-decoupling
51	165 × 85	0.2	-	0.38	Fr-4	5G/Sub-6GHz	Microstrip MIMO
This work	140 × 135	3.5 5.8	7.66 7.84	0.02	Fr-4	5G/Sub-6GHz	Microstrip MIMO array

. The MIMO array is well suited for sub-6 GHz frequencies in 5G applications due to its high gain, broad coverage, and acceptable antenna size demonstrated in the low-cost prototype. The previous works referenced in [44,51] and summarized in Table 5 operate within a single-frequency band. This gives an advantage to our design, which operates on dual-frequency bands, achieving a maximum gain of 7.84 dBi. Reference [44] exhibits a high gain, operating at a low frequency of 2.3 GHz. In contrast, our design functions at both 3.5 GHz and 5.8 GHz.

4. 5G Application Implementation Based on Open Air Interface Platform

The proposed MIMO array antenna has been implemented and experimentally tested using the 5G system OAI platform, which operates at sub-6 GHz, in the Communication and Microwave Research Lab within the Computer, Electrical, and Mathematical Sciences and Engineering (CEMSE) Division at King Abdullah University of Science and Technology in the Kingdom of Saudi Arabia. The Open Air Interface (OAI) is an open-source experimental project that serves as a software platform for 5G applications in LTE (OFDMA) cellular networks. It makes use of software-defined radio (SDR) for implementation.

The OAI platform comprises hardware and software components and is intended to seamlessly integrate with third-party radio frequency (RF) platforms such as the universal software radio peripheral (USRP). In addition to the RF platform, a general-purpose processor or server is required. On the software side, both the OAI EPC and eNB run on a 64-bit Linux PC. The source code needs to be cloned to the computers, followed by the execution of related compilation scripts. Two UEs—essentially, smart mobile phones with SIM cards—are employed as user equipment [55]. A proposed MIMO array antenna testbed using the OAI platform produces a set of telemetry performance parameters for comparison with a dipole antenna already in use on the 5G system on the OAI platform, achieving superior performance.

4.1. 5G Indoor System Implementation Configuration

The main components of EUTRAN include base stations, access points, and mobile terminals, while EPC is the core network. The setup allows the USRP to operate as a transmitter, receiver, or both, and is specifically designed to support different frequency bands within a specified range based on hardware specifications. The system is built using the USRP along with analog-to-digital converters and vice versa, a field-programmable gate array (FPGA), and an antenna port connected to a radio frequency (RF) front end. The proposed MIMO array antenna is connected indoors to the Tx and Rx ports. Subsequently, the computer is linked to the USRP antenna via a USB cable, and an Ethernet cable is connected to the router for Internet access, as shown in Figure 26. Once the system is set up, it is connected to the USRP, configured, and compiled, and finally, the system is executed. Lastly, mobile phones are connected to cellular networks.

Table 6. Environmental parameters of the 5G experimental test.

Environmental Parameters	Configuration	Environmental Parameters	Configuration
Frequency band	N77	SNR	10 dB
Operating frequency	3.5 GHz–5.8 GHz	User equipment noise floor (UE NF)	5 dB
Channel bandwidth	40 MHz	Antenna mode	MIMO
Line of sight	2 m	Operating temperature	−10 °C to 40 °C
gNB transmit power	10 dB	Duplexing mode	TDD

shows the main environmental parameters of the 5G experimental test based on the OAI platform.

4.2. 5G System Performance Evaluation

The performance of the proposed MIMO array antenna in 5G applications is evaluated to showcase its advantages in terms of various telemetry parameters [55,56].

4.2.1. The Received Signal Strength Indicator

The RSSI calculates the average linear value of the overall received power (in watts), which comprises the power from co-channel serving and non-serving cells and neighbor channel interference. Figure 27a compares the proposed MIMO array antenna and the dipole antenna that already operates on the 5G system. The MIMO array antenna has a maximum peak yield of −67 dBm, with −80 dBm for the dipole antenna. The RSSI of −67 dBm represents the minimum signal strength needed for applications that depend on consistently reliable and timely delivery of data packets.

The signal is categorized as very good for VoIP/VoWi-Fi and streaming video 5G applications, but on the other hand, the RSSI of −80 dBm represents the minimum signal strength required for basic connectivity. At this level, packet delivery may be unreliable, so the signal is classified as not good.

4.2.2. Signal-to-Interference-Plus-Noise Ratio

The SINR is the quality measurement representing the ratio between the desired signal’s power and the power of the interference plus noise. The line-locking arrangement depicted in Figure 27b shows superior frequency stability, precision, and reproducibility for the proposed MIMO array antenna compared to the dipole antenna.

4.2.3. Power Headroom Report

The 5G networks utilize the power headroom report (PHR), which enables the gNB to evaluate the UE’s available power margin. When the PHR value is in the negative range, it indicates that the UE is currently transmitting power above the permissible limit. According to Figure 27c, the suggested MIMO array antenna shows superior performance compared to the dipole antenna due to its high level of positivity.

5. Conclusions

The 5G wireless communication technology is required due to the substantial requirement for a higher data rate for mobile communications up to 1 GHz. Several new technologies and methods have been implemented, including MIMO transmission and high-gain antenna arrays used to increase data speeds in 5G. This paper discusses a 140 × 135 × 1.6 mm³ low-cost FR-4 antenna structure of 1 × 4 MIMO arrays that operates from 3.01 GHz to 6.5 GHz at 3.5 and 5.8 GHz resonant frequencies, producing a gain of 7.84 dBi at 5.8 GHz and a radiation efficiency of 71.5%. Over its operational band, the MIMO array exhibits an ECC equal to 0.01 and a CCL of 0.02 bits/s/Hz, which means an effective radiating antenna. Finally, the MIMO array is implemented experimentally in the 5G system based on the OAI platform, confirming its operational effectiveness in the 5G WiMAX application. Because of these properties, the suggested MIMO array antenna is appropriate for 5G communication systems. In summary, the results indicate that the paper’s design is undoubtedly promising for implementing the wideband array antenna design in 5G sub-6 GHz applications.