Metasurface Superstrate Beam Steering Antenna with AMC for 5G/WiMAX/WLAN Applications

A beam-steering antenna based on non-uniform metasurface superstrate and artificial magnetic conductor (AMC), operating at 3.5 GHz, is proposed. The antenna can steer the beam along θ = −18° and 18° with the superstrate and along θ = 0° in the absence of the superstrate with almost zero scan loss. The proposed antenna structure consists of a top layer of non-uniform metasurface superstrate made of a 20 × 20 grid of electrically-small square-shaped metallic pixels while the bottom part consists of AMC with a grid of 5 × 5 pixels. The radiating element, CPW-fed monopole antenna, is placed between AMC and superstrate. The fabricated prototype shows desired beam steering in directions of θ = −18°, 0°, and 18° while maintaining uniform realized gain of 5.5 dBi and matching well with simulations.


Introduction
Fifth generation (5G) is the recent wireless technology that is the next step towards the evolution of cellular communication systems [1,2]. In 5G, the beam steering capability of the antenna plays the major role for efficient coverage to all users [3]. Beam steering is the solution to today's network problems such as call drops, multipath fading in crowded areas [4,5]. Optical beam scanning for short range communication is presented in [6], which achieved 10° beam scanning. Beam scanning is achieved with the help of waffle and waffle iron waveguides. Many beam steering methods have been discussed and reported in [7], which shows one of the methods of beam steering is Birefringent Prisms. However, 1 3 this method is limited as it requires prism thickness of more than 8 cm which may create the problem of beam walkof [8,9] when the beam needs to be steered by 20°. The precise beam steering is obtained using 1D optical phased array on silicon-on-insulator which achieves 2.3° at wavelength of 1550 nm and 14.1° steering. The design is, however, limited due to its high cost and complication in designing with low beam scanning [10]. The reconfigurable partial reflecting surface (PRS) structure was deployed to achieve beam steering from −15° to 15° as reported in [11]. Besides this, the microstrip leaky wave antenna has achieved frequency scanning capability with two tapered slots in the leaky section and reduction of side lobe level is obtained by Yagi elements [12]. Also, 0° to 40° beam scanning is reported in [13] with the help of an electronically controlled microstrip antenna by changing the capacitance value. The tilted beam at ± 15° is observed in [14], by combining parasitic elements with inserted capacitances. The parasitic phased array antenna with varactors and tuning capacitance which provides continuous scanning of ± 15° is reported in [15]. A reconfigurable microstrip Yagi antenna is presented in [16] and beam scanning from −40° to + 40° in H-plane is achieved by two hexagonal slotted tunable parasitic elements with the gain of 3.36 dBi. The beam steering of −30°, 0°, + 30° is reported in [17], which used a grid of 5 ×5 electrically-small rectangular-shaped metallic pixels on the upper layer of the parasitic layer which are connected by switches. However, 360° beam steering is presented in [18] by combining an electronically steerable parasitic array radiator (ESPAR) antenna and the water antenna, but it makes the design complex and cannot be easily designed. In [19], phase varying metamaterial is used for compact steerable directive antenna which achieves a ± 20° steered beam. Furthermore, different beam steering methods have been reported in [20].
In next sections, we provide a step-by-step approach to design the proposed antenna. Sect. 2 starts with the design of a radiating element, CPW-fed monopole antenna. Once, radiating element is designed, the design of AMC and superstrate is discussed. Next, in Sect. 3, various parameters of composite antenna with AMC and superstrate are studied so that change in antenna behavior due to change in these parameters can be quantified. This section also discusses desired radiation behaviors of the proposed antenna. Finally, Sect. 4 concludes the work with concluding remarks.

Design of CPW Fed Monopole Antenna
The radiating element, CPW fed monopole antenna, geometry is shown in Fig. 1. Due to simple design and single side design on substrate, this radiating element is selected for this research. For design, the length of the monopole antenna is denoted as l, and width is denoted by W. The monopole antenna is fed through 50 Ω SMA connector. The antenna is fabricated on FR4 (ε r = 4.4 and tan = 0.02) substrate with a thickness of 1.6 mm. The other dimensions of the antenna are given as: L p = 35, W p = 22, l = 16.7, m = 9.1, g = 0.3 and l 1 = 100 (all dimensions are in mm).

Design of AMC for CPW Fed Monopole Antenna
The center frequency of AMC operation is chosen to be 3.5 GHz. A unit cell of periodicity 10 mm × 10 mm is used for AMC design. Square-shaped metallic patches of side 'a' are printed at the center of the unit cell on FR4 substrate as shown in Fig. 2a. The substrate has thickness, h 1 = 3.2 mm with permittivity of 4.4, and loss tangent of 0.02. For analysis of AMC, the periodic boundary condition (PBC) [19][20][21][22][23] is used. It facilitates quick results instead of simulating the large structure of AMC. By using PBC, shown in Fig. 2b, the reflection phase is calculated by simulation with the help of Floquet port excitation.
In order to ascertain that the AMC is indeed working at desired frequency of 3.5 GHz, unit cell simulation with variable side of the square pixel, a, is run and an attempt is made so that the reflection phase is 0° at desired frequency of 3.5 GHz. From Fig. 2c, we can see that when the side of the pixel, a, is 9.6 mm, the reflection phase is 0°. Henceforth, we will use AMC made of unit cells with a = 9.6 mm to use with the monopole antenna discussed earlier.

Design of Composite CPW Fed Monopole Antenna with AMC and Superstrate
In the previous section, we designed the AMC. Next, we place AMC underneath the antenna substrate as shown in Fig. 3 so that the AMC enhances the radiation in forward direction, i.e., along vertically up direction. It can be seen from Fig. 3 that the grid of 5 × 5 AMC unit cells is placed under the monopole antenna. It is well known that the superstrate can be used to increase the gain of the antenna [11,[24][25][26]. Mostly, the superstrate used for enhancing the design is either in the form of uniform pixels or pixels whose size changes radially out from the center of the superstrate. In such applications, the primary role of superstrate is to work in Fabry-Perot configuration and enhance the gain of the antenna element. In this research, we have a dual purpose of the superstrate. On one hand we want to modify the gain of the antenna and on the other, we want to steer the beam emanating from the antenna in the desired direction. This type of beam steering is very important for upcoming 5G systems. Figure 4a shows the geometry of the proposed superstrate which comprises of metallic patches of different size from a 1 to a 20 . The dimensions of the proposed superstrate are given in Table 1. In order to steer the beam in direction away from the broadside, there should be phase variation for waves emanating from different locations of the superstrate. This phasevarying behavior is provided by a nonuniform metasurface printed on the dielectric substrate acting as a superstrate. For this purpose, the size of the unit cells of the superstrate increases monotonically from one edge to another of the superstrate as shown in Fig. 4a. This way, the superstrate will help to steer the beam in an off-broadside direction. If this superstrate sends the beam in + θ, rotating the superstrate by 180°, taking the z-axis for rotation, will help to send the beam in the −θ direction. In order to send the beam in  broadside direction, the superstrate is removed. To ease the rotation, the interchange of the superstrate may be accomplished by a mechanical solution. In this way, this design can be said to be reconfigurable. However, the speed of beam steering will depend upon how quickly the mechanical motor can help to rotate or remove the superstrate. Figure 4b shows the layered structure of the proposed CPW-fed monopole antenna with AMC and superstrate with different views. The fabricated prototype of the proposed antenna is shown in Fig. 5.

Results and Discussion
All the simulations in this work are conducted using Ansys HFSS, a full wave simulator. The simulated S-parameter comparison of the monopole antenna, monopole antenna with AMC, monopole antenna with AMC and superstrate is shown in Fig. 6.
The comparison of the realized gain for monopole antenna with AMC and superstrate is shown in Fig. 7. It can be observed that there is 3.84 dB increment in gain after applying AMC on the monopole antenna. Furthermore, when the superstrate is applied, the gain of the antenna is 5.5 dBi at 3.5 GHz.

Parametric Analysis of the Antenna
The proposed antenna is fed by CPW as shown in Fig. 1. The parametric studies of the dimensions of the monopole antenna, gap between antenna and superstrate has been carried out to achieve good impedance matching and desired frequency band.

Length of Monopole Antenna, l
The length of the monopole antenna, l, is varied from 20 to 30 mm with the step size of 5 mm. As shown in Fig. 8a the reflection coefficient S 11 is lower than −10 dB for desired frequency bands. It can also be observed that the S 11 resonance moves to lower frequency as the length of the monopole increases.

Gap Between the Antenna and Superstrate
The gap between antenna and superstrate is varied from 4.2 mm to 10.2 mm with the step size of 3 mm. It can be seen from Fig. 8b that there is hardly any effect of varying the gap on the antenna performance. In this way, the antenna will give almost the same results even if the gap changes due to slight mishandling of the antenna.

Near Field Analysis
In order to get insights of working of the superstrate, near field behavior is very helpful [25]. Far field pattern being the Fourier transform of the near-field pattern, can give better understanding of how the superstrate is changing the near field that will eventually change the far field characteristics. To accomplish it, a vertical plane in the center of the antenna is placed on which the dominant component of the near field is studied. In our case, E x is the dominant electric field as the monopole is aligned along the x-axis. In Figs. 9 and 10, E x magnitude and phase behavior in absence and presence of the superstrate are shown. It is evident from Figs. 9a and 10a that the placement of the superstrate confines the fields and hence shows slightly higher field magnitude in case of the superstrate. It is also evident from Figs. 9b and 10b that due to placement of the superstrate, the direction of equi-phase front changes. The phase front is changed towards left, signifying that the outward radiation is expected towards left side when the superstrate is placed. From these figures, it is easy to understand that the absence or presence of the superstrate decides the beam steering direction.
After getting the desired radiation properties, the prototype of the antenna is fabricated as shown in Fig. 5. The proposed antenna is tested using R & S ZNB40 vector network analyzer and anechoic chamber for its properties.

S-Parameters Measurement
The simulated and measured S-parameters of the antenna are shown in Fig. 11. For acceptable performance, the proposed antenna needs to provide good reflection coefficients, |S 11 |, in the desired frequency band. From Fig. 11, it is evident that the fabricated antenna has |S 11 |< −10 dB over the frequency band of 3.1-3.7 GHz. The simulated and measured S-parameters are in good agreement. The fabricated antenna has wide impedance bandwidth for 5G/WiMAX/WLAN applications. Figure 12 shows the radiation pattern of the proposed antenna for φ = 0° cut. The nonuniform superstrate was designed to steer the beam in θ = ± 18°. The superstrate needs to be mechanically rotated or removed depending upon the desired steering beam direction. While, currently, we manually did the operation of mechanically placing and removing the superstrate but the superstrate can be moved by mechanical motor to automate this operation. This type of mechanical movement of a superstrate is simple and can be easily accomplished by mechanical motors.

3
It can be seen from Fig. 12 that the beam scans in ± 18° and 0° in elevation directions in presence of superstrate and in absence respectively. It is also interesting to note that the antenna does not show sidelobes in this plane. The realized gains at these steered angles are 5.5 dB signifying that the antenna shows almost zero scan loss for ± 18° scanning angles. The proposed antenna radiation patterns (co-and cross-polarization) are measured in an anechoic chamber at 3.5 GHz as shown in Figs. 13 and 14. In these figures, copolarization is shown by red lines and cross-polarization by black lines. It is evident from Figs. 13 and 14, that the antenna has very low cross-polarization components. The simulated and measured results are also in good agreement.  Figure 15 shows the simulated and measured gain of the antenna with and without superstrate. It is observed that the gain of the antenna in both cases is 5.5 dBi at 3.5 GHz. The simulated and measured values are also in good agreement.   Table   In Table 2, all existing researched designs are compared with the current work. It is evident from Table 2 that the proposed antenna gives a good beam steering with compact design.

Conclusions
A beam-steering antenna based on non-uniform metasurface superstrate and AMC with CPW-fed monopole antenna as radiating element operating at 3.5 GHz (3.1-3.7 GHz) was presented. The antenna is able to steer beams in directions of −18°, 0° and -18° in elevation while maintaining uniform gain of 5.5 dBi. The antenna is compact, has a  good front to back ratio and low sideline levels. The antenna due to these qualities is an ideal candidate for 5G, WiMAX and WLAN applications.

Authors' Contribution
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by AV, RKA and SNR. The first draft of the manuscript was written by AV and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding
The authors did not receive support from any organization for the submitted work.
Data Availability Data sharing not applicable -no new data generated, or the article describes entirely theoretical research data sharing not applicable to this article as no datasets were generated or analysed during the current study.
Code Availability No codes are used in this manuscript.

Conflict of interest
The Authors declares that there is no conflict of interest'.