A Low-Coupling Broadband MIMO Array Antenna Design for Ku-Band Based on Metamaterials
Article information
Abstract
In response to the demand for broadband antennas in 5G mobile communications, this paper proposes a compact broadband multiple-input multiple-output (MIMO) antenna array. By chamfering a radiation patch and adding parasitic patches, the antenna is miniaturized while also expanding its bandwidth. The mutual coupling between the two antenna elements is reduced by loading a double-layer metamaterial decoupling structure above the elements which consists of an open resonating ring and a square ring. The antenna is simulated and measured using the three-dimensional electromagnetic simulation software HFSS and a vector network analyzer. The results show that the proposed antenna has a good radiation performance of S11 < −10 dB in the operating band of 12.11–13.99 GHz (relative bandwidth of 14.41%). Furthermore, the isolation of the proposed MIMO antenna array in the operating band is improved by more than −18.8 dB on loading the metameterial structure, indicating that the array antenna is suitable for use in the Ku-band.
I. Introduction
With the rapid development of China’s mobile communication industry, the original low-frequency bandwidth can no longer meet demands, with frequency utilization gradually moving toward the Ku-band and other millimeter wave bandwidths. In addition, previously used single antenna units are unable to meet the developmental needs of the communication spectrum. To address these requirements, MIMO antennas were created. Notably, MIMO is a novel multiple-input multiple-output technology that can greatly improve network capacity while ensuring that the network bandwidth remains unchanged [1].
Although the MIMO antenna has several advantages over a single antenna, its biggest disadvantage is that it generates strong electromagnetic mutual coupling phenomena between antenna elements. Scholars have proposed various methods to reduce the degree of coupling, such as adding decoupled branches or incorporating artificial materials. In [2], the authors proposed adding two branches at the ground level and placing the MIMO antennas vertically to improve the isolation of miniaturized ultra-wideband MIMO antennas by at least −10 dB.
Metamaterials are novel artificial materials that can be embedded between radiation units to improve the isolation without increasing the size of the antenna system. The combination of a metamaterial MIMO antenna [3] with a defected ground structure can effectively enhance the isolation of the center frequency point. For instance, the application of metamaterial structures to a millimeter-band MIMO antenna, as proposed in [4], resulted in low mutual coupling of −36.2 dB at 30 GHz. In addition to the method introduced in this paper, the design of microstrip devices based on defected microstrip structures is also very common [5, 6]. Considering this context, this paper intends to present the design, simulation, and testing of a small-size broadband high-isolation MIMO antenna. In the proposed design, isolation is achieved by loading a two-layer metamaterial structure, in which open resonant circular and square rings are located on the upper and lower surfaces of the dielectric substrate, respectively. The bandwidth of the radiating patch was found to apply to the Ku-band through the optimization analysis. The simulation test results showed that the isolation of the antenna after loading the metamaterial in the working band reached more than −18.8 dB. Moreover, this study focused on optimizing and improving the radiation performance of the antenna.
II. Wideband MIMO Antenna Design
The MIMO antenna designed in this paper was printed on an FR4 dielectric substrate (having a dielectric constant of 4.4, and a dielectric loss angle tangent of 0.02). The front side of the antenna consists of two radiating patches fed by a microstrip line feed, with a rectangular corner cut out of the upper left corner of the patch, along with six small parasitic patch cells for bandwidth expansion. The design steps are shown in Fig. 1. Fig. 2 traces the bandwidth of the antenna from its initial design to the final structure, showing an expansion from the original 13.19–14.03 GHz (relative bandwidth of 6.17%) to 12.11–13.99 GHz (relative bandwidth of 14.41%). The final MIMO antenna structure is shown in Fig. 3 and the optimized parameters are shown in Table 1.
III. Metamaterial Unit Design and Analysis
Metamaterials are composite structural materials that can be artificially designed. Since the electromagnetic properties of artificial metamaterials mainly depend on their structure and dimensions, accurately designing these two elements can help adjust both metallic ohmic loss and dielectric loss near its resonance frequency to realize the absorption of incident electromagnetic waves. Furthermore, antenna performance can be improved by overlapping the absorbing frequency band of the absorbing metamaterial with the operating frequency band of the microstrip antenna.
1. Metamaterial Cell Design
As shown in Fig. 4, a two-layer metamaterial unit structure that can generate electromagnetic resonance near 13 GHz was designed. The structure was modeled in accordance with the literature [7], comprising an open resonant ring and a square ring. The outermost square side of the square ring had a length of Ws = 6 mm, the width of the metal wire was v = 0.8 mm, the radii of the outermost and innermost ring of the open resonant ring were r1 = 2.5 mm, and r2 = 1 mm, respectively. This highlights that the overall size of the designed metamaterial structure was very small. The simulation-optimized parameters were g = 0.4 mm, d = 0.4 mm, s = 0.4 mm, s2 = 0.4 mm.
To verify the electromagnetic characteristics of the designed metamaterial unit, the electromagnetic simulation software HFSS (Ansys Inc., Canonsburg, PA, USA) was employed to conduct a detailed analysis. The open resonant circular and square rings were placed on the upper and lower surfaces of the FR4 dielectric substrate, which had a thickness of 1.6 mm. The entire setup was then placed inside an airbox. In the simulation, the left and right surfaces perpendicular to the y-axis were set up for wave-port excitation, the top and bottom surfaces perpendicular to the z-axis were set as ideal magnetic conductors (Perfect H), and the front and back surfaces perpendicular to the x-axis were set as ideal electric conductors (Perfect E). Effectively, the magnetic field was situated perpendicular to the surface of the metamaterial cell to simulate the generation of the magnetic resonance when the magnetic field passed through the cell.
2. Simulation Analysis
The S11 and S21 parameters of the designed metamaterial cell were obtained using HFSS simulation, as shown in Fig. 5(a). A transmission resistance band was observed near 13 GHz due to magnetic resonance. To conduct an in-depth study of the electromagnetic properties of the material, the S-parameter inversion method [8] was implemented to extract the equivalent parameters of an equivalent homogeneous medium. The results obtained using the S-parameter value theory are presented in Fig. 5(b). The theoretical calculation process involved in the S-parameter extraction technology is as follows:
S11 denotes the reflection coefficient. The relationship between S21 and transmission coefficient T can be formulated as follows:
where k0 represents the incident wave number in free space, and d refers to the thickness of the uniform medium plate. Furthermore, the relationship of the S-parameter is related to the refractive index n and impedance z can be expressed as follows:
where R01=(z − 1)/(z + 1). The inversion of Eqs. (2) and (3) leads to the following:
where
Therefore, the value of the refractive index n can be obtained from Eq. (5) as follows:
where m is an integer related to Re(n). Finally, the two parameters that determine metamaterial properties can be obtained from the two Eqs. (9) and (10), which are:
The results showed that the value of the equivalent magnetic permeability μ at 13 GHz was negative, while that of the equivalent permittivity ɛ is positive, indicating that the metamaterial was an electronegative material.
3. Metamaterial Array Structure
To reduce the coupling current between the two closely aligned radiation patches, the array structure illustrated in Fig. 6 was loaded at some distance above the patches. The number of unit structures was obtained by conducing simulation optimization experiments using HFSS software.
IV. Overall MIMO Antenna Structure Design Process
The overall antenna structure is displayed in Fig. 7, in which h1 = 2 mm. To make the measurement results comparable, the structural dimensions of the original antenna and the loaded metamaterial antenna were kept identical.
1. S-Parameter Simulation Analysis
The simulation results, obtained using HFSS software, showed that the center operating frequency of the MIMO antenna loaded with the metamaterials shifted slightly toward a high frequency, although the overall deviation of the operating band was not large, The return loss S11 at the center frequency of the antenna declined from −26.53 dB to −30.28 dB, implying that the introduction of metamaterials failed to destroy the antenna’s impedance-matching characteristics, instead improving them slightly. Fig. 8 shows that the coupling coefficient S21 of the MIMO antenna at the central frequency point decreases from −18.68 dB to −26.50 dB, pointing to a coupling reduction of 7.82 dB compared to the original antenna. Furthermore, the coupling coefficient when considering the entire working frequency range was below −18.8 dB, with the effect of the coupling reduction found to be significant.
2. Antenna Radiation Direction Diagram
Fig. 9 shows radiation direction diagrams of the MIMO antenna with and without metamaterials at 13 GHz in the main radiation direction. Fig. 9(a) clarifies that the introduction of metamaterials narrowed the size of the antennas para-flap to a certain extent. In other words, while the MIMO antenna produces a large para-flap radiation ranging between 300° and 360° when the metamaterials are not loaded, it produces strong main-flap radiation in the direction of 60° after metamaterial loading. Moreover, the para-flap radiation in the second case is very small, further emphasizing that the metamaterial unit structure proposed in this paper exhibits favorable characteristics in terms of improving the coupling degree of MIMO antennas.
3. Antenna Surface Current Distribution
To understand the decoupling principle of the loaded metamaterial antenna more intuitively, the surface current distributions of the original antenna and the loaded metamaterial antenna at 13 GHz were investigated, the results of which are presented in Fig. 10. It is observed that the current between the two neighboring radiating patches of the loaded metamaterial MIMO antenna is confined within the metamaterial structure, effectively reducing the mutual interference between the two radiating patches.
4. Gain and Efficiency
Fig. 11 demonstrates the gain and efficiency attained by the MIMO antenna after metamaterial loading. It is evident that the peak gain ranges from 2.2 to 6.56 dBi, and the radiation efficiency range is 38%–60% in the operating frequency band. The maximum gain is 6.56 dBi at 13.95 GHz, while the minimum gain is 2.2 dBi at 12.82 GHz. Similarly, the antenna’s efficiency remained stable at 60% in the 13.23–13.35 GHz range, showing a minimum efficiency of 38% at 12.11 GHz. While the measured and simulation results for efficiency are largely in agreement, those pertaining to gain show comparatively more deviations, possibly due to the influence of the intermediate air in the dielectric layer.
5. Diversity Characteristics of the MIMO Antenna
In addition to the various basic parameters of a MIMO antenna array that affect antenna performance, some of its diversity characteristic parameters are also significant indicators of whether a designed antenna meets the diversity performance standards of a MIMO system.
Among these, the envelope correlation coefficient (ECC) can be used to verify the diversity performance of a MIMO antenna [9]. The isolation and correlation of a communication channel can be clearly observed from ECC values. A lower ECC indicates lower mutual coupling, which can be identified by a safety threshold of 1. Notably, ECC values range between 1 and 0 [10]. ECC can be calculated using the following equation:
where
Diversity gain (DG) is another basic parameter used to measure the spatial diversity of MIMO antennas. The greater the DG, the better the improvement effect of the MIMO antenna technology score set. DG can be calculated using Eq. (12), as follows:
Fig. 12 depicts the simulated ECC and DG plots of the proposed MIMO antenna. The ECC value stays near the ideal value of 13 GHz, with the overall ECC being less than 0.004 in the operating band. This confirms that the designed MIMO antenna exhibits good isolation and optimal performance in a multipath fading environment. Furthermore, in the working frequency band, the proposed MIMO antenna achieves a DG of 10 dBi, indicating good diversity characteristics.
6. Comparative Analysis of Antenna Array Elements
The MIMO antenna arrays presented in Table 2 use the same decoupling technology, as the one implemented in the present study. Compared to [7], the design bandwidth of the antenna proposed in this paper is relatively narrow, but the degree of isolation is relatively high. Although the proposed design is slightly less isolated compared to [11–14], it is relatively small and has a wider bandwidth. Furthermore, compared to [15–17], the proposed design exhibits certain advantages in bandwidth and in the degree of isolation degree. Notably, to make the overall profile of the proposed antenna thinner, the gap between the two layers of dielectric substrate can be further compressed. Therefore, compared to the antennas mentioned in the literature, the ECC value of the proposed design is closer to zero, making it suitable for use in a MIMO system. The design meets the basic requirements of a MIMO antenna, including high isolation, small volume, and broad_band, with the entire bandwidth also meeting the minimum requirements of the Ku-band.
V. Antenna Field Measurement and Results
To validate the proposed design, an antenna prototype was created and measured. The antenna was simulated using HFSS software. It was constructed according to the parameters shown in Fig. 13. The S-parameter of the antenna was measured using Agilent vector network analyzer. The measured and simulated values of the S-parameters are compared in Fig. 14, exhibiting a small frequency shift between the two. This may be attributed to factors such as SMA connector loss, milling machine limitations, and radiation boundaries during measurement. Furthermore, the measured results show that the measured frequency band of the proposed MIMO antenna is within the range of 12.11–13.99 GHz range with S11 < −10 dB and S21 < −18 dB.
Fig. 15 shows the simulation and measured radiation direction diagrams of the main plane azimuth (E-plane) and pitch (H-plane) at 13 GHz, highlighting that the introduction of the metamaterial had only a slight deviating effect on the radiation direction map.
VI. Conclusion
In this paper, a wideband MIMO antenna with a high isolation of loaded metamaterial is designed for Ku-band. The cutting angle and parasitic patches served to increase the bandwidth of the MIMO antenna, while the double-layer metamaterial design make it suitable for application in the high frequency band. By placing the designed metamaterial unit structure in a 3 × 3 arrangement at a distance of 2 mm from the radiating patch antenna, the coupling between the antenna units was effectively reduced. Simulation comparison results confirmed that the introduction of metamaterials did not deteriorate the impedance matching characteristics and far-field radiation performance of the antenna. Furthermore, the measured results of the antenna were largely consistent with the simulation results. Therefore, the proposed MIMO antenna structure is considered suitable for the field of small broadband antennas with high degree of isolation.
Acknowledgments
This research was funded by the doctoral research startup fund of Northeast Electric Power University (Project No. BSIXM-2021207) and by the National Natural Science Foundation of China (No. 61803356).
References
Biography
Xuemei Zheng, https://orcid.org/0000-0002-4332-8386 is an associate professor and master’s supervisor in Information and Communication Engineering at Northeast Electric Power University. She graduated with a Ph.D. in information and communication engineering from Harbin Engineering University in 2021. Her main research interests include MIMO antennas, microstrip device design, array antennas, high-isolation microstrip antennas, antenna beam-forming technology, and electromagnetic compatibility analysis.
Ziwei Zhao, https://orcid.org/0009-0000-4325-4308 was born in 1999. She received her engineering degree in communications engineering from the Inner Mongolia University of Technology, Hohhot, Inner Mongolia in 2022. She is currently pursuing her M.S. degree in information and communication engineering at Northeast Electric Power University. Her research interests include multi-input multi-output antennas, microstrip antennas, array antennas, and high-isolation microstrip antennas.
Yunan Zhang, https://orcid.org/0009-0009-9660-0249 was born in 2001. In June 2023, she received her bachelor’s degree in electronic information engineering from Sias University, Zhengzhou, Henan Province, China. In August 2023, she was admitted to Northeast Electric Power University, where she is currently pursuing a master’s degree in communication engineering. Her research areas include high-isolation micro-band antennas, ultra-broadband antennas, array antennas and MIMO antennas.
Tongchao Zhang, https://orcid.org/0009-0001-4891-0244 was born in 2000. He received his B.E. degree from the Shenyang Institute of Engineering in 2023. He is currently pursuing his M.S. degree in information and communication engineering at Northeast Power University. His main research areas include MIMO antennas, microstrip antennas, array antennas, high-isolation microstrip antennas, and antenna decoupling technology.
Ao Gui, https://orcid.org/0009-0003-5257-5623 received his bachelor’s degree in electronic information science and technology from Northeast Electric Power University, Jilin City, Jilin Province, in 2021. He is currently pursuing a master’s degree in information and communication engineering at Northeast Electric Power University. His research interests are the characteristic modal analysis of metamaterial antennas and circularly polarized antennas.
Han Wu, https://orcid.org/0009-0005-6051-7196 was born in 1992. He received his B.E. degree in communication engineering from Xi’an Technological University in 2013. He is currently pursuing his M.S. degree in information and communication engineering in Northeast Electric Power University, Jilin. His research interests include metamaterials, multi-band antennas, ultra-wide band antennas and MIMO antenna.