The normalized H-plane radiation patterns are displayed in Figure 9. The patterns indicate a high front-to-back ratio, greater than 10 dB, throughout the operating bandwidth. The beamwidth and gain are adequate to meet the specifications of commercial devices. The proposed antenna would be vertically integrated within the commercial 5G device.
Figure 9 H-plane patterns of the proposed printed dipole integrated with ZIM.
Figure 10 Forward gain of the proposed printed dipole.
Forward gain with and without the MTM unit cells is illustrated in Figure 10. Gain enhancement of close to 2 dB is observed across the operating band. The achieved gain is relatively high for the available aperture of the antenna. It must also be noted that the high dielectric loss tangent of the low-cost substrate does not affect the on-axis gain.
BROADSIDE GAIN ENHANCEMENT
Broadside antennas with unidirectional patterns are equally important for 5G portable devices. These broadside antennas are compact and they can be panel-mounted on a smartphone or wireless dongle. Techniques to enhance gain in broadside antennas include increasing the number of array elements in the phased array, integrating a parasitic superstrate and integrating a dielectric lens with the antenna. Most of these techniques succeed in enhancing the gain of the broadside radiator but they sacrifice the electrical footprint and impedance bandwidth as the compromise. This section explores using MTM unit cells as partially reflecting surfaces (PRS).
A generic layout of the MTM-loaded broadside antenna is illustrated in Figure 11. Like the earlier case, only planar antennas are considered. The antenna is a microstrip-fed element on an electrically-thin substrate. The feeding mechanism is not very important in the broadside radiation case. However, the antenna must have a ground that is electrically large, which means that the standalone antenna will have a low gain unidirectional pattern in the absence of MTM loading. The concept of PRS will not work with bidirectional or omnidirectional antenna elements. The broadside radiator would invariably be a resonant structure; this is because a traveling wave radiator would require an electrically large radiating aperture along the direction of propagation.
Figure 11 Generic layout of a broadside antenna with MTM unit cells.
A quasi-PRS integrated with a broadside antenna operating in the 28 GHz band is shown in Figure 12a. It is a corporate-fed array, composed of four elements of inset-fed microstrip patch antennas with half-wavelength spacing. The antenna array is realized on Rogers 5880 substrate, which offers minimal dielectric loss. The feed network is optimized for wide bandwidth centered at 28 GHz.
The radiating aperture is bent, as illustrated in Figure 12b, to align the radiators with the panel of the target use case device. This method of corner bending prevents radiation toward the user. To enhance the boresight gain of this antenna array, a quasi-PRS structure is integrated with the array, as illustrated in Figure 12b.
Figure 12 (a) 28 GHz broadside antenna. (b) Realization of the quasi-PRS integrated broadside antenna.
The boresight gain characteristic of the antenna, with and without the integrated PRS from 25 to 31 GHz is shown in Figure 13. The quantum of gain enhancement is quite low due to the available aperture. Higher gain enhancement could be achieved by extending the PRS cells in the lateral dimension. However, this approach would make the design unsuitable for commercial devices.
Figure 13 Boresight gain of the quasi-PRS integrated broadside antenna.
Figure 14 Normalized radiation patterns of the broadside antenna with MTM cells.
The unidirectional narrow beam patterns of the proposed antenna are illustrated in Figure 14. Even though this topology offers natural isolation with the back-end electronics, the effective radiating volume is higher compared to conventional end-fire antennas. Therefore, this approach might not be a viable option for commercial deployment.
REDUCING MUTUAL COUPLING
Mutual coupling is an important phenomenon in multiple radiator antenna systems designed for 5G portable devices. As mentioned earlier, one of the common techniques for gain enhancement in the mmWave domain is to implement phased arrays. Typically, the radiating elements within the phased array would be placed and operated electrically close together. For instance, for maximum gain enhancement, the antennas might be placed at a half-wavelength distance at the resonant frequency. Ideally, the antennas should contribute to beamforming alone, but due to the electrical proximity of the elements, part of the input power will be wasted in mutual coupling with other ports. Two primary consequences result from this topology: the overall forward gain of the antenna system decreases due to the partial loss of energy within the ports and there may be detuning and deterioration of the input impedance characteristics of the individual ports, potentially leading to an impedance mismatch with the back-end electronics.
The other consequence of the mutual coupling effect is most impactful for MIMO antenna systems. More mutual coupling means less diversity gain and this affects the overall throughput of the antenna system. Reducing mutual coupling by even 5 dB among the ports might improve the forward gain by up to 1 dB. A generic layout of a two-port antenna system integrated with MTM for mutual coupling reduction is illustrated in Figure 15.
Figure 15 Layout of an MTM-based antenna to reduce mutual coupling.
In Figure 15, both antennas are identical and similar to most commercial implementations. The radiators can be either broadside or end-fire. Both the elements would be typically spaced at a fraction of a wavelength, computed at the center of the operating frequency. The sub-wavelength MTM unit cells must not be the reflective type because this will severely detune both constituent antennas. The performance or characteristics of the unit cell cannot be used to predict the post-integrated mutual coupling reduction phenomenon.
There are some simple and effective ways to reduce mutual coupling. These include placing the antennas electrically far away from each other or placing the radiators in an orthogonal orientation. Unfortunately, neither of these techniques would enable mmWave 5G antenna communications because distant antenna placement prevents beamforming and orthogonal antenna placement will severely degrade the beam integrity.
CONCLUSION
This article discusses the feasibility of metamaterials in the design of high gain compact antennas for mmWave 5G devices. Gain enhancement using MTM unit cells for both end-fire and broadside radiators has been explained with case studies. Finally, the concept of mutual coupling reduction using MTM unit cells for mmWave bands has been illustrated.
References
- T. S. Rappaport, S. Sun, R. Mayzus, H. Zhao, Y. Azar, K. Wang and F. Gutierrez, “Millimeter wave Mobile Communications for 5G Cellular: It Will Work,” IEEE Access, 1, 2013, pp. 335–349.
- S.K. Koul and G. S. Karthikeya, “Antenna Architectures for Future Wireless Devices,” Springer Nature, 2022.
- Y. Liu, Z. Yue, Y. Jia, Y. Xu and Q. Xue, “Dual-Band Dual-Circularly Polarized Antenna Array With Printed Ridge Gap Waveguide,” IEEE Transactions on Antennas and Propagation, Vol. 69, No. 8, Aug. 2021, pp. 5118–5123.
- S. X. Ta, H. Choo and I. Park, “Broadband Printed-Dipole Antenna and Its Arrays for 5G Applications,” IEEE Antennas and Wireless Propagation Letters, Vol. 16, pp. 2183–2186, 2017
- P. Kumar, T. Ali and M. M. M. Pai, “Electromagnetic Metamaterials: A New Paradigm of Antenna Design,” IEEE Access, Vol. 9, 2021, pp. 18722–18751.
- P. B. Saha, D. Ghoshal and R. K. Dash, “A Miniaturized Frequency Reconfigurable Antenna with Half-mode CRLH-embedded Metamaterial Arm,” Journal of Electromagnetic Waves and Applications, 35(3), 2021, pp. 277–290.
