Junho Park, Ahmed Abdelmottaleb Omar, Jonghyun Kim, Jaehyun Choi, Beakjun Seong, Jongwoo Lee, Wonbin Hong
IMAGE LICENSED BY INGRAM PUBLISHING
The limited beam-scanning angles of planar phased arrays are one of the most critical challenges for millimeter-wave (mm-wave) wireless communication systems. This article proposes a new class of antenna topology, denoted as a stackable patch antenna (SPA), that achieves near-spherical beam-steering coverage for mm-wave applications. The proposed SPA consists of two identical rectangular patch antenna elements. By separately controlling the magnitude and phase of the excitation ports, the proposed antenna can radiate in multiple directions that are normal or parallel to the patch elements. In addition, the proposed structure has the advantage of enabling end-fire radiation when the structure is encompassed by the full metallic frame of a device. This is accomplished without any deformation or modification of the metal frame.
As an example of a mm-wave device, the metal frame of a user device is modeled and fabricated. Near-isotropic beamforming coverage is achieved by placing two SPA packages diagonally at each opposing corner of the mobile devices, resulting in maximizing the cumulative distribution function (CDF) specified by the 3rd Generation Partnership Project (3GPP) for proof-of-concept purposes. The measured beam-scanning performances in the normal and parallel directions to the patch are ±43° (peak gain: 8.4 dBi) and ±33° (peak gain: 5.1 dBi), respectively. No distinctive nulls in the end-fire direction exist despite the presence of the metallic frame. The feasibility of the SPA is further verified in indoor stationary scenarios. The access point (AP) based on the SPA improves the received power CDF corresponding to the lower percentile range by more than 3 dB compared to the reference scenario incorporating five APs with transceivers.
Beamforming antenna technology is imperative to combat the large propagation loss at the mm-wave spectrum, for example, in user equipment (UE) attempts to connect to APs [1]. To compensate for the narrow analog beamwidth of mm-wave, beam steering can be used to enable a wider angular coverage for mm-wave wireless devices. Naturally, the integration of phased-array antennas and improvement of beam coverage have been extensively studied and presented for mm-wave stationary and mobile applications.
Recently, an advanced packaging strategy integrated a multiple mm-wave beamforming antenna-in-package (AiP) within UE to mitigate hand blockage effects and maximize the beam coverage. In [2], two mm-wave AiP modules are conformally arranged along the top and bottom bezel area of a cellular device to maximize the beam-steering angles in the end-fire direction (in the direction parallel to the plane containing the antenna elements). The corner and the lateral area of the UE can be utilized to incorporate end-fire antennas [3], [4], [5], [6], [7], [8], [9], [10], [11]. The planar folded slot antennas [8] with a height profile of less than 1/9 ${\lambda}_{0}$ are proposed for a multilayer substrate within UE. The horizontal metal strip line-coupled dipoles and vertical magnetoelectric monopoles are complementally integrated to achieve an end-fire radiation with dual polarization [10]. In [11], the vertically polarized electric dipole is formed to generate end-fire beams and fit at the edge of UE. In addition, the phased arrays can be integrated inside the back cover of cellular devices to enable broadside beamforming and mitigate the antenna placement restrictions due to the real estate issues within UE [12].
However, most of the solutions have not taken into account the metallic frame of the UE [8], [13], [14], [15], [17], [18], [19]. The metal frame of a wireless device is important for its mechanical stability, and it also functions as a protective frame of the device. However, from the vantage point of the antenna, the metal frame functions as an obstacle for the end-fire radiation mode. When the metal frame is placed in front of conventional end-fire antennas, the radiation patterns are severely distorted and reflected. This becomes extremely problematic for mm-wave beamforming antennas since the antenna beam direction becomes out of sync with the corresponding beamforming algorithm. The end-fire radiation plays an important role in stabilizing the communication link [20].
Various solutions have been studied to overcome the blockage of metallic frames in mm-wave mobile devices. One solution [21], [22] is to etch mm-wave slot arrays directly on the metal frame. The three-element phased array can be integrated into a UE by forming a 20 × 3.5-mm2 window in the metal frame [23]. In [24], a 25 × 10 × 6-mm aperture is cut into the metal frame to place a mm-wave module. In [25], numerous longitudinal slots are cut out of the metal frame to utilize it as an electromagnetic coupling structure so that horizontally polarized bow tie arrays can radiate through the frame with enhanced beam-steering gain. Similarly, two rows of grooves filled with dielectric material were formed on the frame to overcome the main beam blockage [26]. However, these solutions require modification of the metal frame, which increases the complexity of the fabrication of the metal frame and reduces its mechanical strength. In addition, the integration of phased-array antennas within the metallic frame can cause an undesirable short circuiting effect, which significantly degrades the performance of the sub-6-GHz antennas [27].
The wide-beam lens and antenna designs for mm-wave stationary applications have received great interest. A gradient index lens is presented to overcome the effective aperture size reduction caused by beam steering, which is an intrinsic property of all types of planar arrays [28]. The fabricated prototype demonstrates a maximum gain of 21.2 dBi and a wide scanning range up to ±58° within a 3.1-dB scan loss. In [29], a dipole-based antenna topology including a substrate-integrated cavity is proposed for wide-angle beam-scanning applications in the H-plane. The experimental results show that the proposed antenna achieves solutions with a wide beamwidth of more than 180° and a wide beam-scanning ability up to 65°, with a low scanning loss of 3.7 dB and a low sidelobe level of −15 dB. However, most studies have not taken into account the compatibility with current beamforming phased-array AiP architectures and scalability for implementing spherical beamforming antenna systems for both stationary and mobile applications.
Setting the challenges surrounding the metal frame aside, achieving a spherical beam-steering antenna for mm-wave mobile and stationary applications remains one of the most critical challenges within the wireless community at present. This article proposes a new class of antenna topology, denoted as the SPA, that achieves near-spherical beam-steering coverage for mm-wave applications. As illustrated in Figure 1, the SPA concept can be applied to two stacked compact AiPs within mobile applications, contributing to the enhancement of spherical beam coverage, which fundamentally eliminates the need for any deformation of the metal frame. In addition, the stationary application (e.g., fixed wireless infrastructure) can incorporate the SPA-based AP, which not only improves the beam coverage but also reduces the number of APs as well as the cost of installing APs.
Figure 1. The SPA concept for the realization of spherical beam coverage of mm-wave mobile and stationary applications. (a) A 1 × 4 phased array SPA is integrated within a present-day mobile device featuring a metal frame to enable end-fire radiation, which can significantly contribute to achieving spherical coverage of the mobile device. (b) The SPA concept is applied to stationary applications featuring a large-scale phased array of more than 16 antenna elements. The APs consisting of the 4 × 4 phased arrays with SPA technology efficiently improve the beam coverage, which is explained in the “Feasibility of the Spa in Indoor Stationary Scenarios” section.
The SPA element consists of two identical microstrip patch antennas, which operate at the fundamental TM01 mode. A single microstrip element using a low-temperature cofired ceramic (LTCC) process is devised as an example and illustrated in Figure 2(a). The multilayer printed circuit board fabrication process may be utilized as well. The LTCC package contains 10 stacked layers, and each layer features a thickness of 100 μm with relative permittivity ${\varepsilon}_{r} = {5.9}$ and loss tangent tan ${\delta} = {0.002}$. Silver metal layers feature a thickness of 10 μm. The diameters of the vias and capture pads are ${100}\,{\mu}{\text{m}}$ and ${150}\,{\mu}{\text{m}}$, respectively. The patch is excited by a vertical probe, which is protruding through the ground plane. Strip-line topology is employed to guide the electromagnetic wave from the feeding port to the patch. The strip line is guided by electrical walls created by vias to reduce the coupling effect with other elements when forming the array configuration. The resonant frequency of the patch is realized by controlling the length of the patch ${(}{l}_{p}{)}$ and is designed to operate at 28 GHz for exemplification. The impedance of the strip line is determined by the width of the line ${(}{l}_{f}{)}$ and the height of the substrate ${(}{h}_{f}{)}$. Two identical patches are arranged symmetrically with a separation distance ${(}{d}_{s}{)}$ in the z-direction, as shown in Figure 2(b).
Figure 2. (a) The single microstrip element. (b) A cross section of the SPA concept. (c) The basic structure of the SPA array (ws = 5 mm, wp = 1.92 mm, ls = 5.4 mm, lp = 1.92 mm, hs = 0.54 mm, hf = 0.43 mm, lf = 0.12 mm, and ds = 4.9 mm).
Two important parameters affect the radiation characteristics of the SPA: one is the relative permittivity of the substrate, and the other is the separation distance ${(}{d}_{s}{)}$. The relative permittivity of the LTCC ${(}{\varepsilon}_{r} = {5.9}{)}$ is sufficient for the fringing fields to be concentrated in the substrate at the radiating edges [30]. Moreover, the thickness of commercially available recent mobile devices is typically fewer than 10 mm, resulting in an electrically small separation distance ${(}{d}_{s}{)}$. For clarity, in this article, the far-field radiation directions of the SPA are defined and classified into the broadside radiation (±z) and the end-fire radiation/out-of-phase mode (±x and ±z), and the operating frequency is 28 GHz.
A 1 × 4 phased array SPA is illustrated in Figure 2(c). The separation between the antenna elements is 5 mm (0.47 ${\lambda}_{0}$ at 28 GHz), which is smaller than a half wavelength to avoid grating lobes during beam scanning. The proposed antenna can achieve various radiation patterns by adjusting the magnitude and phase of the signal applied at each port. Figure 3(a) and (b) shows the simulated 3D radiation plot of the simple 1 × 4 patch array when the top/bottom four ports are activated with in-phase excitation. These broadside radiation modes are defined as the top broadside and bottom broadside modes, respectively.
Figure 3. Simulated 3D radiation patterns of the (a) top broadside mode, (b) bottom broadside mode, and (c) out-of-phase mode.
When activating all ports and when a phase difference of 180° is applied between the top and bottom patch elements, end-fire radiation is accomplished, and the simulated 3D radiation plot is shown in Figure 3(c). Although the phase difference between the top and bottom patches is 180°, bidirectional broadside radiation can be achieved. This is due to the fact that the ground planes between the patch elements prevent electric currents that are out of phase from being mutually canceled. In addition, the phase difference of the fringing fields at the radiating edges is near zero, resulting in vertically polarized (z-axis) end-fire radiation (please refer to [19, Figure 3]). This radiation mode is defined as the out-of-phase mode/end-fire radiation. The maximum realized gain shown in Figure 3(a) and (b) is 10.9 dBi in the +z-and –z-directions, respectively, while the maximum realized gain in Figure 3(c) is 8 dBi in the ±z-direction. The intensity of the electric currents induced on the patches is identical in all modes. However, there is a gain difference between the broadside and out-of-phase modes in the broadside direction. This is attributed to the out-of-phase mode, in which a fraction of the fringing fields contributes to radiation in the end-fire direction instead of the broadside direction, resulting in a relatively reduced realized gain compared to the broadside mode.
It is worth mentioning that each antenna can be designed independently, regardless of the presence of the antenna on the opposite layer because the two patch antennas on the top and bottom layers are electrically isolated through the ground plane. The input impedance (Zin) at the center of the patch is ideally zero so that the coaxial feeding point is offset along the x-axis at the center for ${50}{-}{\Omega}$ impedance matching. Consequently, the fringing field at the radiating edges in the +x-axis direction, which is near the feeding point, is stronger than the –x-axis. Logically, it is advantageous to place the feeding point of the patch in the +x-direction. The realized gains of the out-of-phase mode in the +x- and –x-axes are 6.3 and 3 dBi, respectively.
A parametric study regarding the effect of ${d}_{s}$ is conducted to further examine the SPA concept for metal frame devices. To study the effect of the separation distance of an SPA, ${d}_{s}$ is varied from 0 to 7.5 mm. The separation distance has an influence on the radiation pattern in the out-of-phase mode, as shown in Figure 4. When ${d}_{s}$ equals 0 mm, the most ideal dimension of the SPA, an omnidirectional radiation pattern can be achieved in the E-plane (x–z-plane). As ${d}_{s}$ increases, the omnidirectional radiation pattern in the E-plane is deformed. The fringing fields from the top and bottom patches are vertically polarized, which are considered as two vertically polarized element arrays separated by distance ${d}_{s}$ [16]. Considering that the thickness of present-day mobile devices ranges from approximately 7 (0.65 ${\lambda}_{o})$ to 8 mm (0.74 ${\lambda}_{o})$, the configured distance between the patch elements remains within the realistic and practical range.
Figure 4. The variation of the radiation pattern in the E-plane in the out-of-phase mode according to the separation distance (ds).
To compare the radiation characteristic of the SPA in free space with the metal frame-integrated situation, an antenna package is designed, as shown in Figure 5(a). To enhance the spherical coverage, four patch elements are arranged along the y-axis (SPA 1) to achieve end-fire radiation along the x-axis, and the other four elements are arranged along with the x-axis (SPA 2) to achieve end-fire radiation along the y-axis. The metal frame is designed to be 74 × 149 × 7.6 mm in size, emulating the real-life chassis of present-day user devices. The antenna package is symmetrically duplicated in the z-axis to establish the SPA configuration and integrated with the metal frame, as illustrated in Figure 5(b).
Figure 5. Illustration of (a) 3-D view of the proposed SPA package. (b) The top view of the proposed SPA package. (c) The SPA packages integrated into metal frame. (Some part of the metal frame and 8 ports cable on the bottom side are removed for a clear view, ds = 4.9 mm.)
The simulated E-field distribution of the SPA in the out-of-phase mode without the metal frame is plotted in Figure 6(a). As mentioned earlier, the fringing fields induced by patches at the top and bottom are summed in the end-fire direction, resulting in vertically polarized end-fire radiation. Figure 6(b) shows the E-field distribution in the end-fire direction when the metal frame is included. As the fringing field propagates along the x-axis, it encounters the metal frame. However, the SPA can generate the radiation in the end-fire direction without any deformation of the metal frame by synthesizing the fringing field beyond the metal frame.
Figure 6. The simulated E-field distribution of the SPA under the (a) free space condition and (b) metal frame integrated condition.
The radiation property of the proposed SPA is studied according to the distance between the antenna array and the metal frame (df) as well as the width of the metal frame (wf), as illustrated in Figure 7. The values of the realized gain in Figure 7(b) and (c) are extracted at the end-fire direction in the end-fire mode. It can be seen from the realized gain in Figure 7(b) that, as the distance between the metal frame and the antenna “df” increases, the realized gain in the end-fire direction increases. This is attributed to an increment in fringing fields as the ground plane in front of the antenna increased. A distance of 1.7 mm is chosen as a reasonable distance. A smaller distance may be selected on the cost of realized gain in the end-fire direction.
Figure 7. The simulated effect of two parameters on the end-fire realized gain. (a) The simulation model. (b) The effect of changing the distance between the antenna array and the metal frame. (c) The effect of changing the width of the metal frame.
Regarding the width of the metal frame, it is confirmed from Figure 7(c) that the realized gain of the proposed array is proportional to wf. This is mainly attributed to the thickened metal frame improving the coupling intensity between fringing fields from the SPA. In addition, the beam shape, not shown here, is well maintained regardless of the width of the metal frame. Although a thick metal frame can improve the realized gain of the proposed SPA mode, a thin metal frame design is preferred in present-day mm-wave devices. It is worth mentioning that the ±broadside realized gain almost does not change in both simulations.
Conventional planar antenna arrays inherently exhibit a quasi-hemisphere spatial coverage. Therefore, isotropic spherical coverage cannot be achieved even if two planar antenna arrays are placed on the front and back sides of the device. Moreover, conventional broadside antenna arrays cannot achieve radiation in the forward direction of the metal frame, i.e., the ±x- and ±y-axes. A quasi-isotropic spherical coverage can be achieved by using the SPA concept and placing two SPA modules on diagonally opposing sides/edges of a mobile device, as shown in Figure 8(a). One of the two antenna modules can be activated using the antenna diversity method. It is worth mentioning that L-shaped metal blockages are employed behind antenna arrays to mimic the effect of the main board and battery of the mm-wave device.
Figure 8. Quasi-isotropic spherical coverage. (a) Three SPA configurations. (b) The total beam-scanning patterns. (c) The CDF of the EIRP distribution.
To evaluate the effect of the SPA on spherical beamforming coverage, the total beam-scanning patterns and CDF of the effective isotropic radiated power (EIRP) distribution with three array topologies are extracted and presented in Figure 8(b) and (c), respectively. First, in configurations 1 and 2, a single SPA module is placed at the corner of the mobile device. The y-oriented 1 × 4 SPA array is activated in configuration 1, while the x- and y-oriented 1 × 4 SPA arrays are activated in configuration 2. To enhance the spherical beamforming coverage, an additional SPA module is placed diagonally, as shown in configuration 3. Configuration 3 can achieve an isotropic spherical coverage by controlling the magnitude (on/off) and phase of each port in a switched diversity manner. Twenty-one beams are used per each SPA module to calculate the total beam-scanning patterns and CDF of the EIRP distribution. The total beam-scanning patterns of each configuration are simulated and select the highest realized gain at every spherical angular point ${(}{\varphi},{\theta}{)}$. Then EIRP ${(}{\varphi},{\theta}{)}$ is calculated as follows: \[{\text{EIRP}}{\left({\varphi},{\theta}\right)} = {\text{Realized gain}}{\left({\varphi},{\theta}\right)} + {\text{P}}{\_}{\text{out}} \tag{1} \]
P_out is the total radio-frequency integrated circuit (RFIC) output power to each antenna port, and then the EIRP ${(}{\varphi},{\theta}{)}$ is normalized to be 22.4 dBm as specified by 3GPP regarding the UE minimum peak EIRP for power class 3. Therefore, the peak EIRP values are aligned through three curves. These simulated CDF curves are shown graphically in Figure 8(c). The difference in the CDF curve between configurations 1 and 2 results from the absence or presence of radiation in the +y-axis. This can be further examined in Figure 8(b) to quantify the difference in the beam coverage region between configurations 1 and 2. When comparing CDF curves of configurations 2 and 3, configuration 3 is clearly improved, which is attributed to the existence of beam coverage in the –x- and –y-directions, also illustrated in Figure 8(b). This ascertains the fact that higher spherical coverage is realized through configuration 3 compared to configuration 2.
From the perspective of the 3GPP specification, configuration 1 requires at least 4.9 dBm of accepted output power (P_out) to meet the 11.2-dBm value at CDF = 50% [see Figure 8(c)]. Given the fact that the recently released output power of the 5G RFIC specification is from 10 to 16 dBm, configuration 1 satisfies the requirement of the EIRP at both CDF = 100% and 50%, which are specified in 3GPP. Therefore, by arranging the SPA packages as in configuration 3, it is possible to meet both quasi-isotropic spherical coverage and the 3GPP specification.
A prototype of the SPA package is fabricated using an LTCC processing technique as shown in Figure 9(a). A metal frame made of aluminum is shown in Figure 9(b). A series of plastic pillars are used to fix the SPA packages to the fabricated metal frame. Simulations indicate that these pillars have a minimal effect on the antenna topology.
Figure 9. Photographs of the (a) fabricated SPA packages and (b) front and backside views of the SPA with a metal frame.
The S-parameters are measured using an N5247A Keysight PNA-X. The input reflection coefficients of the eight ports are illustrated in Figure 10. GGB Industries picoprobe ground–signal–ground (GSG) wafer probe tips featuring ${500}{-}\,{\mu}{\text{m}}$ pitch are used to contact GSG pads [see Figure 5(a)]. Simulation results confirm the interelement isolation of more than 16 dB at 26.5–29.5 GHz. The discrepancy between the simulated and measured results is due to the fabrication error of the LTCC process and the deviation in substrate material properties.
Figure 10. The simulated and measured reflection coefficients.
The far-field radiation patterns of the fabricated SPA packages with metal frames are measured using an anechoic far-field chamber at Pohang University of Science and Technology, Pohang, Republic of Korea. The measurement setups for both the broadside and end-fire modes are shown in Figure 11. The beamforming capabilities of the SPA are accessed using a mm-wave RF reference board consisting of 16 RF channels. Two eight-port multi-coaxial cables are utilized to excite the 16 ports of the SPA packages. However, the magnitude and phase can be scaled in 0.5-dB and 5° steps, respectively, by controlling the GUI through the Ethernet interface. Therefore, it is possible to calibrate the power and phase before measuring the radiation patterns. All measurements were made in the environment where antennas are integrated into the fabricated metal frame.
Figure 11. The measurement setup of the proposed SPA with a metal frame: the (a) broadside mode and (b) out-of-phase mode. Rx: receiver; Tx; transmitter.
The measured and simulated radiation patterns of SPA 1 and SPA 2 [see Figure 5(b)] in the E-plane at 28 GHz are illustrated in Figures 12(a) and (b) and 13(a) and (b), respectively. They were measured from –90° to 90° due to the limitation of the anechoic far-field chamber environment. The radiation patterns of broadside mode on the opposite side are not presented due to the symmetric configuration. The radiation pattern of the broadside mode of SPA 1 is slightly distorted due to blockage and scattering caused by adjacent cables. It is ascertained that a good correlation between the simulated and measured radiation patterns is achieved.
Figure 12. The measured and simulated normalized far-field radiation patterns of SPA 1 in the E-plane in the (a) out-of-phase mode and (b) top broadside mode as well as the measured normalized far-field beamforming pattern in the H-plane in the (c) broadside mode and (d) out-of-phase mode.
The beam-steering performances of SPA 1 are illustrated in Figure 11(c)–(e). The measured peak gains of the antenna array in broadside mode and out-of-phase mode are 8.4 and 4.98 dBi, respectively. Similarly, the beam-steering performances of SPA 2 are illustrated in Figure 13(c)–(e). The measured peak gains in broadside mode and out-of-phase mode are 8.3 and 5.1 dBi, respectively. It is worth mentioning that the simulated antenna efficiency of the SPA features more than 70%, including the insertion loss of the RF connector at 28 GHz. Table 1 summarizes the performance comparison with recently reported 5G mobile antennas, taking into account the metal frame. It should be noted that an SPA can be integrated into mobile devices in a planar manner without distorting the metal frame and achieve near-equal spherical coverage as well.
Figure 13. The measured and simulated normalized far-field radiation patterns of SPA 2 in the E-plane in the (a) out-of-phase mode and (b) top broadside mode as well as the measured normalized far-field beamforming pattern in the H-plane in the (c) broadside mode and (d) out-of-phase mode.
Table 1. A comparison with the mm-wave antennas integrated into 5G metal frame mobile devices.
Two indoor propagation scenarios are verified through a 3D ray shoot-and-bound (SBR) simulation at a center frequency of 28 GHz to estimate the beam coverage of the large-scale AiP featuring an SPA, as shown in Figure 14. The length and width of the hallway are 600 and 150 m, respectively, and the ceiling height is 4.7 m. The entire hallway is constructed of concrete material with a thickness of 30 cm. The four-element phased array with 8-dBi gain is used as a receiver above ground to collect data.
Figure 14. The mm-wave indoor link deployment. (a) The 4 × 4 AiPs are deployed at an interval of 100 m on the ceiling. (a) The 4 × 4 AiPs featuring an SPA are deployed at an interval of 150 m on the ceiling.
In the reference scenario [Figure 14(a)], five APs consisting of 4 × 4 phased arrays are deployed at 100-m intervals on the ceiling. In the proposed scenario, three APs consisting of 4 × 4 phased arrays with SPA technology are deployed at intervals of 150 m on the ceiling. Each AP operates as a transmitter to which a total transmit power of 30 dBm is applied. The UE, which steers the directional beam toward the ceiling direction, moves from the start point to the endpoint at a linear interval of 5 m to accurately collect the received power data. Since interferences from neighboring APs are not considered, the results from each AP and UE are extracted separately and overlapped. It is confirmed from Figures 15 and 16 that, despite using only three ceiling-type APs, an AiP with an SPA improves the received power CDF corresponding to the lower percentile range by more than 3 dB compared to the reference scenario including 5 APs.
Figure 15. The received power versus UE distance from the starting point in two indoor link deployments.
Figure 16. The CDF of the received power in two indoor link deployments.
A concept of an SPA has been devised for mm-wave mobile and stationary applications for the first time. It has been verified that the proposed structure has the advantage of relieving the recently arising main beam blockage issue without requiring deformation of the metal frame. The isotropic coverage has been accomplished by placing two SPA packages diagonally at two corners of a mobile device as an example. This concept is exemplified at 28 GHz for a proof of concept and is ascertained to satisfy the CDF benchmark in accordance with the 3GPP standards. The measured maximum realized gains in the broadside and end-fire directions are 8.4 and 5.1 dBi, respectively. The measured beam-steering performances in the broadside and end-fire directions are ±43° and ±33°, respectively. The system-level SBR simulation demonstrates that AiP modules with the SPA significantly enhance the spherical coverage CDF of fixed ceiling-type APs by more than 3 dB in lower percentile ranges with fewer transceivers. These results prove that the proposed SPA technology efficiently improves the spherical beamforming coverage in both mobile and stationary scenarios.
Junho Park and Ahmed Abdelmottaleb Omar equally contributed to this work. This work was supported in part by Institute of Information and Communications Technology Planning and Evaluation grants funded by the Korea government (MSIT) (2020-0-00858, 2018-0-00823) and in part by the Material Component Technology Development Program funded by the Ministry of Trade, Industry, and Energy (Korea) (20010540).
Junho Park (jaypark@kreemo.io) is the head of Research and Development Team at KREEMO Inc., Seoul 06682, South Korea. His research interests include millimeter-wave phased-array antennas and radio-frequency front-end designs for 5G/beyond 5G communication systems. He is a Member of IEEE.
Ahmed Abdelmottaleb Omar (ahmed.omar.1@kfupm.edu.sa) is an assistant professor at the King Fahd University of Petroleum and Minerals, Dhahran 34464, Saudi Arabia. His research interests include the analysis and design of frequency selective surface; microwave absorbers; and compact, wideband, and millimeter-wave 5G antennas. He is a Member of IEEE.
Jonghyun Kim (jonghyunkim@postech.ac.kr) is an engineer with the System Large-Scale Integration Division at Samsung Electronics, Hwaseong 18448, South Korea. His research interests include radio-frequency and analog circuits.
Jaehyun Choi (jaehyunchoi@postech.ac.kr) is a professional engineer with the Antenna-in-Package Department at LG Innotek Co. Ltd., Seoul 07796, South Korea. His research interests include multiphysics-based antennas, the circuit design of microwave/millimeter-wave, and radar in autonomous vehicles.
Beakjun Seong (jaypark@kreemo.io) is an engineer with the Research and Development Team at KREEMO Inc., Seoul 06682, South Korea. His research interests include design and verification for millimeter-wave antennas and radio-frequency modules.
Jongwoo Lee (jwlee30@kreemo.io) was a vice president of the Sales and Marketing Team at KREEMO Inc., Seoul 06682, South Korea. His research interests include millimeter-wave antennas and Internet of Things systems.
Wonbin Hong (whong@postech.ac.kr) is the Muenjae Distinguished Professor with the Department of Electrical Engineering, Pohang University of Science and Technology, Pohang 37673, South Korea. He is also with Kreemo Inc., Seoul 06682, South Korea. His research interests include applied electromagnetics and wireless circuits. He is a Senior Member of IEEE.
[1] W. Roh et al., “Millimeter-wave beamforming as an enabling technology for 5G cellular communications: Theoretical feasibility and prototype results,” IEEE Commun. Mag., vol. 52, no. 2, pp. 106–113, Feb. 2014, doi: 10.1109/MCOM.2014.6736750.
[2] W. Hong, K. Baek, Y. Lee, Y. Kim, and S. Ko, “Study and prototyping of practically large-scale mmWave antenna systems for 5G cellular devices,” IEEE Commun. Mag., vol. 52, no. 9, pp. 63–69, Sep. 2014, doi: 10.1109/MCOM.2014.6894454.
[3] Y.-W. Hsu, T.-C. Huang, H.-S. Lin, and Y.-C. Lin, “Dual-polarized quasi Yagi–Uda antennas with Endfire radiation for millimeter-wave MIMO terminals,” IEEE Trans. Antennas Propag., vol. 65, no. 12, pp. 6282–6289, Dec. 2017, doi: 10.1109/TAP.2017.2734238.
[4] H. Li, Y. Li, L. Chang, W. Sun, X. Qin, and H. Wang, “Wideband dual-polarized Endfire antenna array with overlapped apertures and small clearance for 5G millimeter-wave applications,” IEEE Trans. Antennas Propag., vol. 69, no. 2, pp. 815–824, Feb. 2021, doi: 10.1109/TAP.2020.3016512.
[5] O. Jo, J.-J. Kim, J. Yoon, D. Choi, and W. Hong, “Exploitation of dual-polarization diversity for 5G millimeter-wave MIMO beamforming systems,” IEEE Trans. Antennas Propag., vol. 65, no. 12, pp. 6646–6655, Dec. 2017, doi: 10.1109/TAP.2017.2761979.
[6] A. Li, K.-M. Luk, and Y. Li, “A dual linearly polarized end-fire antenna array for the 5G applications,” IEEE Access, vol. 6, pp. 78,276–78,285, Dec. 2018, doi: 10.1109/ACCESS.2018.2884946.
[7] J. Zhang, K. Zhao, L. Wang, S. Zhang, and G. F. Pedersen, “Dual-polarized phased array with end-fire radiation for 5G handset applications,” IEEE Trans. Antennas Propag., vol. 68, no. 4, pp. 3277–3282, Apr. 2020, doi: 10.1109/TAP.2019.2937584.
[8] J. Park, H. Seong, Y. N. Whang, and W. Hong, “Energy-efficient 5G phased arrays incorporating vertically polarized Endfire planar folded slot antenna for mmWave mobile terminals,” IEEE Trans. Antennas Propag., vol. 68, no. 1, pp. 230–241, Jan. 2020, doi: 10.1109/TAP.2019.2930100.
[9] A. Li and K. Luk, “Single-layer wideband end-fire dual-polarized antenna array for device-to-device communication in 5G wireless systems,” IEEE Trans. Veh. Technol., vol. 69, no. 5, pp. 5142–5150, May 2020, doi: 10.1109/TVT.2020.2979636.
[10] H.-T. Chou, S.-J. Chou, J. D. S. Deng, C.-H. Chang, and Z.-D. Yan, “LTCC-based antenna-in-package array for 5G user equipment with dual-polarized Endfire radiations at millimeter-wave frequencies,” IEEE Trans. Antennas Propag., vol. 70, no. 4, pp. 3076–3081, Apr. 2022, doi: 10.1109/TAP.2021.3118819.
[11] W. El-Halwagy, R. Mirzavand, J. Melzer, M. Hossain, and P. Mousavi, “Investigation of wideband substrate-integrated vertically-polarized electric dipole antenna and arrays for mm-Wave 5G mobile devices,” IEEE Access, vol. 6, pp. 2145–2157, 2018, doi: 10.1109/ACCESS.2017.2782083.
[12] S. Shakib, J. Dunworth, V. Aparin, and K. Entesari, “mmWave CMOS power amplifiers for 5G cellular communication,” IEEE Commun. Mag., vol. 57, no. 1, pp. 98–105, Jan. 2019, doi: 10.1109/MCOM.2018.1701378.
[13] W. Li, W. Chung, and K. Wong, “Highly-integrated dual-band mmWave antenna array for 5G mobile phone application,” in Proc. 14th Eur. Conf. Antennas Propag. (EUCAP), 2020, pp. 1–5, doi: 10.23919/EuCAP48036.2020.9135212.
[14] R. Rodríguez-Cano, S. Zhang, and G. F. Pedersen, “Beam-steerable multi-band mm-wave bow-tie antenna array for mobile terminals,” in Proc. 14th Eur. Conf. Antennas Propag. (EUCAP), 2018, pp. 1–4, doi: 10.1049/cp.2018.0418.
[15] S. Zhang, I. Syrytsin, and G. F. Pedersen, “Compact beam-steerable antenna array with two passive parasitic elements for 5G mobile terminals at 28 GHz,” IEEE Trans. Antennas Propag., vol. 66, no. 10, pp. 5193–5203, Oct. 2018, doi: 10.1109/TAP.2018.2854167.
[16] C. A. Balanis, Antenna Theory and Design, 3rd ed. New York, NY, USA: Wiley, 2005.
[17] S. Zhang, X. Chen, I. Syrytsin, and G. F. Pedersen, “A planar switchable 3-D-coverage phased array antenna and its user effects for 28-GHz mobile terminal applications,” IEEE Trans. Antennas Propag., vol. 65, no. 12, pp. 6413–6421, Dec. 2017, doi: 10.1109/TAP.2017.2681463.
[18] I. Syrytsin, S. Zhang, G. F. Pedersen, and A. S. Morris, “Compact quad-mode planar phased array with wideband for 5G mobile terminals,” IEEE Trans. Antennas Propag., vol. 66, no. 9, pp. 4648–4657, Sep. 2018, doi: 10.1109/TAP.2018.2842303.
[19] I. Syrytsin, S. Zhang, G. F. Pedersen, K. Zhao, T. Bolin, and Z. Ying, “Statistical investigation of the user effects on mobile terminal antennas for 5G applications,” IEEE Trans. Antennas Propag., vol. 65, no. 12, pp. 6596–6605, Dec. 2017, doi: 10.1109/TAP.2017.2681701.
[20] W. Hong, K.-H. Baek, and S. Ko, “Millimeter-wave 5G antennas for smartphones: Overview and experimental demonstration,” IEEE Trans. Antennas Propag., vol. 65, no. 12, pp. 6250–6261, Dec. 2017, doi: 10.1109/TAP.2017.2740963.
[21] J. Bang and J. Choi, “A SAR reduced mm-Wave beam-steerable array antenna with dual-mode operation for fully metal-covered 5G cellular handsets,” IEEE Antennas Wireless Propag. Lett., vol. 17, no. 6, pp. 1118–1122, Jun. 2018, doi: 10.1109/LAWP.2018.2836196.
[22] B. Yu, K. Yang, C. Sim, and G. Yang, “A novel 28 GHz beam steering array for 5G mobile device with metallic casing application,” IEEE Trans. Antennas Propag., vol. 66, no. 1, pp. 462–466, Jan. 2018, doi: 10.1109/TAP.2017.2772084.
[23] R. Montoya Moreno, J. Ala-Laurinaho, A. Khripkov, J. Ilvonen, and V. Viikari, “Dual-polarized mm-Wave Endfire antenna for mobile devices,” IEEE Trans. Antennas Propag., vol. 68, no. 8, pp. 5924–5934, Aug. 2020, doi: 10.1109/TAP.2020.2989556.
[24] J. Kurvinen, H. Kähkönen, A. Lehtovuori, J. Ala-Laurinaho, and V. Viikari, “Co-designed mm-Wave and LTE handset antennas,” IEEE Trans. Antennas Propag., vol. 67, no. 3, pp. 1545–1553, Mar. 2019, doi: 10.1109/TAP.2018.2888823.
[25] R. Rodriguez-Cano, S. Zhang, K. Zhao, and G. F. Pedersen, “mm-Wave beam-steerable Endfire array embedded in a slotted metal-frame LTE antenna,” IEEE Trans. Antennas Propag., vol. 68, no. 5, pp. 3685–3694, May 2020, doi: 10.1109/TAP.2020.2963915.
[26] R. Rodriguez-Cano, K. Zhao, S. Zhang, and G. F. Pedersen, “Handset frame blockage reduction of 5G mm-Wave phased arrays using hard surface inspired structure,” IEEE Trans. Veh. Technol., vol. 69, no. 8, pp. 8132–8139, Aug. 2020, doi: 10.1109/TVT.2020.2996360.
[27] J. Kurvinen et al., “Capacitively-loaded feed line to improve mm-Wave and Sub-6 GHz antenna co-existence,” IEEE Access, vol. 8, pp. 139,680–139,690, Jul. 2020, doi: 10.1109/ACCESS.2020.3012773.
[28] Z. Qu, S. Qu, Z. Zhang, S. Yang, and C. H. Chan, “Wide-angle scanning lens fed by small-scale antenna array for 5G in millimeter-wave band,” IEEE Trans. Antennas Propag., vol. 68, no. 5, pp. 3635–3643, May 2020, doi: 10.1109/TAP.2020.2967086.
[29] C. Mao, M. Khalily, P. Xiao, L. Zhang, and R. Tafazolli, “High-gain phased array antenna with Endfire radiation for 26 GHz wide-beam-scanning applications,” IEEE Trans. Antennas Propag., vol. 69, no. 5, pp. 3015–3020, May 2021, doi: 10.1109/TAP.2020.3028181.
[30] J. Kim, J. Park, A. A. Omar, and W. Hong, “A symmetrically stacked planar antenna concept exhibiting quasi-isotropic radiation coverage,” IEEE Antennas Wireless Propag. Lett., vol. 19, no. 8, pp. 1390–1394, Aug. 2020, doi: 10.1109/LAWP.2020.3003011.
Digital Object Identifier 10.1109/MAP.2022.3208797