A Dielectric Resonator Antenna with Enhanced Gain and Bandwidth for 5G Applications

In this paper, a dielectric resonator antenna (DRA) with high gain and wide impedance bandwidth for fifth-generation (5G) wireless communication applications is proposed. The dielectric resonator antenna is designed to operate at higher-order TEδ15x mode to achieve high antenna gain, while a hollow cylinder at the center of the DRA is introduced to improve bandwidth by reducing the quality factor. The DRA is excited by a 50 Ω microstrip line with a narrow aperture slot. The reflection coefficient, antenna gain, and radiation pattern of the proposed DRAs are analyzed using the commercially available full-wave electromagnetic simulation tool CST Microwave Studio (CST MWS). In order to verify the simulation results, the proposed antenna structures were fabricated and experimentally validated. Measured results of the fabricated prototypes show a 10-dB return loss impedance bandwidth of 10.7% (14.3–15.9GHz) and 16.1% (14.1–16.5 GHz) for DRA1 and DRA2, respectively, at the operating frequency of 15 GHz. The results show that the designed antenna structure can be used in the Internet of things (IoT) for device-to-device (D2D) communication in 5G systems.


Antenna Design and Analysis
The configuration of the proposed DRAs with dimensions of length (a), width (b), and height (d) is shown in Figure 1a,b operating at 15 GHz. The length, width, and height of the designed structure are represented as a × b × d = 0.2 λ × 0.2 λ × 1 λ. The DRA is made of an ECCOS-TOCK HiK material with a dielectric constant (ε r ) of 10 and loss tangent (tan δ) of 0.002. The Rogers™ RT/Duroid 5880 substrate with a permittivity of 2.2 and a loss tangent (tan δ) of 0.0009 is used. The thickness of the Rogers substrate is 0.254 mm. Each DRA is mounted on a 20 mm × 20 mm = 1λ × 1λ ground plane and excited by a 50 Ω standard microstrip line with an aperture slot in the ground plane. The ground plane is printed on the top side of the substrate. It is important to mention here that a microstrip feedline is used due to the ease of fabrication. The slot length l s , width w s and stub length S are adjusted to match individual antennas. The detailed optimized dimensions of the proposed antenna structures are listed in Table 1. All dimensions are in millimeters (mm). In each case, stub length S was adjusted to optimize the matching impedance of individual DRAs. The resonant frequencies, f o , of the TE x δnm mode can be predicted using a dielectric waveguide model (DWM) [29]. The wave numbers k x , k y , and k z can be deduced by solving the following transcendental equations: Sensors 2020, 20, 675 3 of 12 where k 2 x + k 2 y + k 2 z = ε r k 2 o , where c is the velocity of light, ε r is the relative permittivity of the DRA, k o is the free space wavenumber, and m and n are half-wave field variations along the yand z-directions, respectively. The symbols k x , k y , and k z represent the wave numbers in the x-, y-, and z-directions, respectively.
Sensors 2020, 20, 675 3 of 12 Where k + k + k = ε k , (2) k = , k = , k = , (3) (3) where c is the velocity of light, is the relative permittivity of the DRA, k is the free space wavenumber, and and are half-wave field variations along theand -directions, respectively. The symbols k , k , and k represent the wave numbers in the -, -, and -directions, respectively.     Figure 2 compares the simulated reflection coefficients |S 11 |of the DRA operating in higher-order (TE x δ15 ) mode without a cylindrical hole (DRA1) and that of the DRA with a cylindrical hole (DRA2).
It can be seen from Figure 2 that the DRA operating in higher-order (TE x δ15 ) mode without a cylindrical hole obtained an impedance bandwidth of 1.6 GHz (10.6%), ranging from 14.3 GHz to 15.9 GHz. The DRA operating in higher-order mode (TE x δ15 ) with a cylindrical hole at the center achieved a comparatively wider impedance bandwidth of 2.6 GHz (17.4%), operating from 14.3 GHz to 16.9 GHz. The bandwidth of DRA2 was relatively larger than DRA1 because a hollow cylindrical hole was drilled at the center of the DRA2, which reduced the radiation Q-factor of the antenna; therefore, the impedance bandwidth was enhanced. Figure 3a,b present the electric field (E-field) distribution in the XY plane for DRA1 (without a cylindrical hole) and DRA2 (with a cylindrical hole), respectively. The electric field distribution was stronger in DRA2 compared to DRA1. The cylindrical hole at the center of DRA2 strengthened the electric field near the center of the DRA. This helped in increasing the bandwidth and efficiency of the DRA. The simulated magnetic fields (H-fields) of both antennas are plotted in Figure 4a,b, respectively, at the operating frequency of 15 GHz. Figure 4a,b show the magnetic field distribution in higher-order TE x δ15 mode.  Figure 2 compares the simulated reflection coefficients |S11|of the DRA operating in higherorder (TE ) mode without a cylindrical hole (DRA1) and that of the DRA with a cylindrical hole (DRA2). It can be seen from Figure 2 that the DRA operating in higher-order (TE ) mode without a cylindrical hole obtained an impedance bandwidth of 1.6 GHz (10.6%), ranging from 14.3 GHz to 15.9 GHz. The DRA operating in higher-order mode (TE ) with a cylindrical hole at the center achieved a comparatively wider impedance bandwidth of 2.6 GHz (17.4%), operating from 14.3 GHz to 16.9 GHz. The bandwidth of DRA2 was relatively larger than DRA1 because a hollow cylindrical hole was drilled at the center of the DRA2, which reduced the radiation Q-factor of the antenna; therefore, the impedance bandwidth was enhanced. Figures 3a,b present the electric field (E-field) distribution in the XY plane for DRA1 (without a cylindrical hole) and DRA2 (with a cylindrical hole), respectively. The electric field distribution was stronger in DRA2 compared to DRA1. The cylindrical hole at the center of DRA2 strengthened the electric field near the center of the DRA. This helped in increasing the bandwidth and efficiency of the DRA. The simulated magnetic fields (H-fields) of both antennas are plotted in Figures 4a,b, respectively, at the operating frequency of 15 GHz. Figures 4a,b show the magnetic field distribution in higher-order ( ) mode.

Measurement Results and Discussion
In this section, the simulated and measured results are analyzed and discussed in detail. The

Measurement Results and Discussion
In this section, the simulated and measured results are analyzed and discussed in detail. The performance of the antenna prototypes was designed and simulated using the commercial threedimensional (3D) electromagnetic (EM) Computer Simulation Technology (CST) Microwave Studio

Measurement Results and Discussion
In this section, the simulated and measured results are analyzed and discussed in detail. The performance of the antenna prototypes was designed and simulated using the commercial three-dimensional (3D) electromagnetic (EM) Computer Simulation Technology (CST) Microwave Studio software. Based on the parameters given in Table 1, prototypes of the proposed DR antennas were fabricated and tested to validate the simulated results. The photographs of the fabricated proposed DR antennas are shown in Figure 5. The reflection coefficients were measured using a vector network analyzer (VNA), while antenna gain and the radiation patterns were measured in an anechoic chamber. The simulated and measured results of reflection coefficients S 11 of the DRA prototypes are depicted in Figure 6. It can be seen from Figure 6 that DRA1 obtained a simulated and measured −10-dB impedance bandwidth of 10.6% and 10.7%, respectively. On the other hand, DRA2 achieved a simulated and measured −10-dB bandwidth of 17.4% and 16.1%, respectively. The slight difference between the simulated and measured results can be attributed to fabrication imperfections.

Measurement Results and Discussion
In this section, the simulated and measured results are analyzed and discussed in detail. The performance of the antenna prototypes was designed and simulated using the commercial threedimensional (3D) electromagnetic (EM) Computer Simulation Technology (CST) Microwave Studio software. Based on the parameters given in Table 1, prototypes of the proposed DR antennas were fabricated and tested to validate the simulated results. The photographs of the fabricated proposed DR antennas are shown in Figure 5. The reflection coefficients were measured using a vector network analyzer (VNA), while antenna gain and the radiation patterns were measured in an anechoic chamber. The simulated and measured results of reflection coefficients S11 of the DRA prototypes are depicted in Figure 6. It can be seen from Figure 6 that DRA1 obtained a simulated and measured −10-dB impedance bandwidth of 10.6% and 10.7%, respectively. On the other hand, DRA2 achieved a simulated and measured −10-dB bandwidth of 17.4% and 16.1%, respectively. The slight difference between the simulated and measured results can be attributed to fabrication imperfections.   For comparison, the measured and simulated results of the reflection coefficient of DRA1 without a cylindrical hole and DRA2 with a cylindrical hole are given in Table 2.

without Cylindrical Hole (DRA1)
The simulated and measured reflection coefficients versus frequency plots of the mode without a cylindrical hole (DRA1) are represented in Figure 7. It can be seen from Figure 7 that the proposed antenna attained simulated and measured −10-dB impedance bandwidths of 10.6% and 10.7%, respectively. The plot of the simulated and measured antenna gain and radiation efficiency as a function of frequency is depicted in Figure 8. With reference to the plot, the simulated and measured For comparison, the measured and simulated results of the reflection coefficient of DRA1 without a cylindrical hole and DRA2 with a cylindrical hole are given in Table 2. 3.1. TE x δ15 without Cylindrical Hole (DRA1) The simulated and measured reflection coefficients versus frequency plots of the TE x δ15 mode without a cylindrical hole (DRA1) are represented in Figure 7. It can be seen from Figure 7 that the proposed antenna attained simulated and measured −10-dB impedance bandwidths of 10.6% and 10.7%, respectively. The plot of the simulated and measured antenna gain and radiation efficiency as a function of frequency is depicted in Figure 8. With reference to the plot, the simulated and measured antenna gains were 10.5 dBi and 10.4 dBi, respectively. As presented in Figure 8, the measured and simulated radiation efficiencies were 97% and 95%, respectively.
Sensors 2020, 20, 675 7 of 12 antenna gains were 10.5 dBi and 10.4 dBi, respectively. As presented in Figure 8, the measured and simulated radiation efficiencies were 97% and 95%, respectively.   The simulated three-dimensional (3D) radiation pattern of the proposed antenna is illustrated in Figure 9. Figure 10 shows the simulated and measured normalized radiation pattern of the proposed structure in the E-and H-planes at 15 GHz. Figure 10 shows the normalized radiation pattern for DRA1 along the H-plane where the half-power beam width was 49.2° in the major lobe and the radiated power in the side lobe level was −11.8dB. The major lobe is located at 0°. In the E-plane, the half-power beam width in the major lobe was 46.4° and the radiated power in the side lobe level was −9.3db.  The simulated three-dimensional (3D) radiation pattern of the proposed antenna is illustrated in Figure 9. Figure 10 shows the simulated and measured normalized radiation pattern of the proposed structure in the E-and H-planes at 15 GHz. Figure 10 shows the normalized radiation pattern for DRA1 along the H-plane where the half-power beam width was 49.2 • in the major lobe and the radiated power in the side lobe level was −11.8dB. The major lobe is located at 0 • . In the E-plane, the half-power beam width in the major lobe was 46.4 • and the radiated power in the side lobe level was −9.3db.  The simulated three-dimensional (3D) radiation pattern of the proposed antenna is illustrated in Figure 9. Figure 10 shows the simulated and measured normalized radiation pattern of the proposed structure in the E-and H-planes at 15 GHz. Figure 10 shows the normalized radiation pattern for DRA1 along the H-plane where the half-power beam width was 49.2° in the major lobe and the radiated power in the side lobe level was −11.8dB. The major lobe is located at 0°. In the E-plane, the half-power beam width in the major lobe was 46.4° and the radiated power in the side lobe level was −9.3db.

3.2.
with Cylindrical Hole (DRA2) Figure 11 demonstrates the simulated and measured reflection coefficients of the proposed antenna. With reference to Figure 11, the proposed antenna structure achieved simulated and measured bandwidths (S11<-10) of 17.4% (14.3−16.9 GHz) and 16.1% (14.1−16.5GHz), respectively. The slight difference in the measured and simulated results occurred because of the fabrication of the

TE x
δ15 with Cylindrical Hole (DRA2) Figure 11 demonstrates the simulated and measured reflection coefficients of the proposed antenna. With reference to Figure 11, the proposed antenna structure achieved simulated and measured bandwidths (S11<−10) of 17.4% (14.3-16.9 GHz) and 16.1% (14.1-16.5GHz), respectively. The slight difference in the measured and simulated results occurred because of the fabrication of the DRA during its assembly process. Figure 12 shows the simulated and measured gain and efficiency of DRA2. It can be seen from Figure 12 that the simulated and measured antenna gains were 10.5 dBi and 10.4 dBi, respectively, while the simulated and measured radiation efficiencies were 98% and 96%, respectively. The simulated three-dimensional (3D) radiation pattern of DRA2 is demonstrated in Figure 13. Figure 14 shows the simulated and measured normalized radiation pattern of DRA2 in the E-plane and H-plane at 15 GHz. Figure 14 shows the normalized radiation pattern for DRA2 with a cylindrical hole along the H-plane, where the half-power beam width was 54.1 • in the major lobe, and radiated power in the side lobe level was −13.6 dB. In the E-plane, the half-power beam width in major lobe was 48.6 • , and the radiated power in the side lobe level was -10.4dB. The major lobe was located at 0 • . Figure 10. Simulated and measured normalized radiation pattern in the E-plane and H-plane of DRA1 at 15 GHz. Figure 11 demonstrates the simulated and measured reflection coefficients of the proposed antenna. With reference to Figure 11, the proposed antenna structure achieved simulated and measured bandwidths (S11<-10) of 17.4% (14.3−16.9 GHz) and 16.1% (14.1−16.5GHz), respectively. The slight difference in the measured and simulated results occurred because of the fabrication of the DRA during its assembly process. Figure 12 shows the simulated and measured gain and efficiency of DRA2. It can be seen from Figure 12 that the simulated and measured antenna gains were 10.5 dBi and 10.4 dBi, respectively, while the simulated and measured radiation efficiencies were 98% and 96%, respectively. The simulated three-dimensional (3D) radiation pattern of DRA2 is demonstrated in Figure 13. Figure 14 shows the simulated and measured normalized radiation pattern of DRA2 in the E-plane and H-plane at 15 GHz. Figure 14 shows the normalized radiation pattern for DRA2 with a cylindrical hole along the H-plane, where the half-power beam width was 54.1° in the major lobe, and radiated power in the side lobe level was -13.6 dB. In the E-plane, the half-power beam width in major lobe was 48.6°, and the radiated power in the side lobe level was -10.4dB. The major lobe was located at 0°.    In Table 3, the simulated and measured results of the proposed DRA prototypes are summarized. An expression that shows the relationship between the bandwidth (BW) and radiation Q-factor (Q) of the DRA is as follows [30]:

with Cylindrical Hole (DRA2)
Equation (5) defines the Q-factor in terms of VSWR and bandwidth (BW). With reference to Equation (5), the radiation Q-factor is inversely proportional to the bandwidth. Thus, the equation    In Table 3, the simulated and measured results of the proposed DRA prototypes are summarized. An expression that shows the relationship between the bandwidth (BW) and radiation Q-factor (Q) of the DRA is as follows [30]: Equation (5) defines the Q-factor in terms of VSWR and bandwidth (BW). With reference to Equation (5), the radiation Q-factor is inversely proportional to the bandwidth. Thus, the equation In Table 3, the simulated and measured results of the proposed DRA prototypes are summarized. An expression that shows the relationship between the bandwidth (BW) and radiation Q-factor (Q) of the DRA is as follows [30]: Equation (5) defines the Q-factor in terms of VSWR and bandwidth (BW). With reference to Equation (5), the radiation Q-factor is inversely proportional to the bandwidth. Thus, the equation clearly shows that the radiation Q-factor of the antenna is reduced and, thus, the impedance bandwidth enhanced. In Table 4, the comparison of volume-to-surface ratio of the DRA with and without a cylindrical hole is given. Finally, a performance comparison between the proposed DRA and previously published work was carried out [31][32][33], as given in Table 5. From Table 5, it can be found that the proposed antenna structure exhibits a wider bandwidth, higher gain, and higher radiation efficiency relative to the aforementioned work. The proposed structure shows better performance compared to previous work.

Conclusions
A high-gain and wideband dielectric resonator antenna was designed, simulated, fabricated, and experimentally verified. The proposed structure achieved a wide bandwidth and high gain operating in higher-order mode using a new approach of putting a cylindrical hole at the center of the DRA. The DRAs were designed at the operating frequency of 15 GHz. The DRAs were fabricated and measured to validate the proposed design concept. Measured results of the fabricated antenna prototypes showed an impedance bandwidth of 10.7% from 14.3-15.9GHz and 16.1% from 14.1-16.5GHz with a high gain of 10.4dBi for DRA1 and DRA2, respectively. The measured and simulated results of the DRA were in good agreement. Furthermore, the results show that the designed antenna is suitable for future 5G communication applications.