Abstract

A novel wideband bandpass frequency-selective surface functioning at X, Ku, and Ka bands is proposed in this article. The designed FSS has a metallic square loop and a circular ring, and they are printed on both sides of the FR4 substrate. The proposed design FR4-based single-layer FSS is operating from 11.075 GHz to 22.075 GHz with a fractional bandwidth of 66.36%. The parameters of the square loop and circular ring regulate the characteristics of the passband. The optimum dimension of these parameters is obtained with parametric analysis. The proposed structure is measured and fabricated. However, the measured results strongly agree with the simulated results, which authenticate the proposed design performance.

1. Introduction

In most RF circuits, the performance of the circuit totally depends on the characteristics of the filter. Such a filtering operation is performed by the frequency selective surface (FSS) [1] in a microwave wireless communication environment. FSS entails 2D or 3D structures with a definite period. The structure may be a slot or patch designed on a substrate made up of dielectric. Researchers widely use wideband FSS in many applications [25], including the radomes, Cassegrain antenna, shielding, absorbers, stealth applications, and so on, since most of the applications now operate over a wide operating range. Patch-based FSS achieves band stop performance, while bandpass performance is realized using slot-based FSS [1]. The major pullback limitation for the bandpass FSS is the narrow bandwidth [68]. The design and development of wideband bandpass FSS in the radomes and in the system with frequency reuse configuration, however, have had a significant research gap in recent years, according to the researchers. Few bandpass FSS with wideband are reported in [9]. There is great demand for wideband bandpass FSS in various applications. In the literature, few bandstop FSS [10, 11] and bandpass FSS [12, 13] are reported for the application, as mentioned earlier.

To improve the bandpass FSS bandwidth, the researchers intended to develop techniques like 2, 2.5, and 3D. The latter two [1416] strongly attract investigators to the FSS design due to an additional degree of freedom compared to the 2D [1720] topology. But the major limitations are the design and fabrication complexity. In [17], planar multilayer FSS is designed, but the high-quality factor results in narrow fractional bandwidth, and in [18], multilayer FSS with a stable Q factor consisting of the metal patch with wireframe is reported; it also results in a narrow band. A multilayer FSS can attain a maximum fractional bandwidth (FBW) of 20% by using nonresonant elements and adding more conducting layers to increase the bandwidth. Most of the 2D-based FSS results in narrow bandwidth due to the resonating behaviour. This can be overcome with nonresonant cascaded structures [21], and mushroom-shaped FSS with 15% FBW is reported in [22]. The hybrid unit cell is used to achieve FBW of 11.35%, as reported in [22]. These two different structures are printed on adjacent sides of the substrates and have a high degree of design complexity. In [23], FSS based on a patch with VIA is designed, which offers FBW of 110%, and this structure has very poor band rejection characteristics, multilayer FSS with narrowband responses is reported in [24]. Another method of improving the bandwidth of the FSS is numerous steering conducting layers [25]. Even though the 63% FBW is achieved, the fabrication complexity is higher. The convoluting structures [26] are another approach, which can improve the bandwidth but has an asymmetric response due to different metallic topologies. In [27], metamaterial structure-based FSS is reported, but the gain is not stable in the operating X band. Triple band ultra-wideband FSS designed with symmetrical structures on either side of the substrate, wherein they are not identical structures [28]. It is possible to enhance the properties of frequency selectivity and miniaturisation by creating patches or etching slots. By incorporating slots and slits in the designed structure, it may increase the possibilities of wider range of applications.

Thereby, the proposed design performs as wideband (bandpass FSS). All the above methods in the literature can improve the bandwidth, but they also suffer from larger profiles, fabrication difficulties, and larger insertion loss.

In this article, a single-layer FSS with a wideband response is proposed for the X, Ku, and Ka band. The proposed FSS resonates at 16.5 GHz, and it has good stability under various incident angles. The unit cell has an overall dimension of 0.46 λ0 × 0.46 λ0. The structural parameters directly impact the resonance behaviour, and hence the optimum dimensions are obtained using the parametric analysis. The significance of the proposed FSS structure is mentioned below:(1)The motivation of the proposed work is to reduce the reflections in the aircraft at a low cost, to design an FSS with minimum thickness for stealth applications, and to achieve wideband frequency. The proposed single-layer double sided FSS is compact with a minimum thickness and covers X, Ku, and Ka band with a fractional bandwidth of 66.36% as a wideband bandpass filter.(2)The proposed prototype structure is fabricated and measured.(3)The structure exhibits good angular and polarization stability in both TE and TM modes.(4)On either side of the passband, the reflection coefficient is better than 20 dB.

Further, the structure of this manuscript is as follows: Section 2 describes the proposed unit cell design. Section 3 discusses parametric analysis with fuzzy verification. The fabricated proposed structure is compared with the measured and simulated results in Section 4. Finally, concludes the work by providing a summary in Section 5.

2. FSS Unit Cell Design and Analysis

On the FR4 substrate, a symmetrical square loop and circular patch are designed to produce a wide bandpass filter. Figure 1 depicts the proposed FSS unit cell geometry. On either side of the substrate, the unit cell consists of a symmetrical square loop with a circular conducting ring. The white area in the front view of Figure 1 is the FR4 substrate region, and the metallic region is represented in the black colour area. On both sides of the FR4 dielectric substrate, the identical structure is printed. The L1 = 11.9 mm is the length of the metallic square loop’s inner side; L2 = 12.5 mm is the length of the loop’s outer side; and R1 = 0.6 mm is the radius of the circular conductive ring, which is employed as the identical structure of the unit cell. FR4 substrate of thickness T1 (0.4 mm) with loss tangent and 0.02 dielectric constant 4.4 and copper patch thickness T2 (0.035 mm) have been used. The optimum dimension of the critical parametric is finalized using parametric analysis. In Figure 1(a) the proposed FSS is shown in front and back views. In Figure 1(b), side view is shown, followed by the FSS design in the CST EM studio environment in Figures 1(c) and 1(d).

Fundamentally, frequency-selective surfaces are periodic structures distributed evenly in one or two dimensions to perform as filters, mostly the single unit cell dimension would be half of a wavelength [1]. By considering this, for the required frequency, the overall dimension is obtained. Meanwhile, by introducing additional square patches and rings the desirable frequency band is achieved. The evolution of each element and the improved results are plotted in Figure 1(e).

The proposed unit cell has a simple design using a simple circular ring and a symmetrical square loop. The symmetrical structure is chosen to obtain polarization stability at the unit cell level. The resonant frequency varied by changing its elements perimeter [29]. The Q factor is the reciprocal of the fractional bandwidth. Q = fcentre/(f2f1). The Q factor for the proposed design is Q = 1.4977.

The full-wave simulation of the proposed FSS is performed using the CST software, and the unit cell boundary condition is used to replicate the infinite structure. Floquet ports are used for exciting the FSS. The transmission and reflection coefficients at various angles of incidence (AoI) for both polarization of TE and TM modes are presented in Figure 2. It has been noticed that the proposed FSS shows a stable resonance up to 45°. Beyond 45°, the shift in frequency is very low. The stable resonance stability at various AoI attributes to the unit cell’s reduced overall size of 0.46 λ0 × 0.46 λ0. (λ0–wavelength of free space @ lowest frequency). The angular stability of TM mode nearly achieves a stable response comparing to the minimum variation in TE mode. Since the incident wave of the E-field polarizes in Y-direction and H-field polarizes in X-direction this could cause a minor deviation in the output observed. Typically, the free space impedance of incidence for TE mode is the ratio of impedance and cosθ. Similarly, the E-field wave polarizes in X-direction, and the impedance of incidence in TM mode is the product of impedance and cosθ; hence, the H-field of the AoI is comparatively similar. It manifests a better response in TM mode compared to the TE mode polarizations [30]. With excellent stability in the resonant behaviour, the proposed FSS parameters such as L1, L2, T1, and R1 have excellent control over the passband. Further, in Section 3, the parameters of the simulated reflection coefficient of the proposed FSS are discussed.

3. Parametric Effect, Fuzzy Logic FBW Validation, and Equivalent Circuit of the Proposed FSS

Figure 3 shows the reflection coefficient of the proposed FSS for various L1 values. It is observed that the resonance frequency decreases with increasing patch length. Patch length reduction will result in a larger dielectric gap between the patches, decreasing the capacitance that increases the resonant frequency and providing a good correlation between the circumference of the ring and the effective wavelength at the resonance frequency. Hence, L1 = 11.9 mm is chosen for the final fabrication. In Figure 4, the reflection coefficient for various values of L2 is plotted, it is observed from the figure. As we increase the unit cell size, the transmission pole at 15.9 GHz is reduced, and achieving good impedance bandwidth from 11.075 GHz to 22.075 GHz. Hence, L2 = 12.5 mm is chosen for the final fabrication. Figure 5 depicts the S11 of the proposed FSS for various R1 values. The figure shows that R1 = 0.6 mm has good impedance bandwidth since the large ring will provide a larger current path, increasing the wavelength and reducing the frequency. From Figure 6, it is observed that T1 = 0.4 mm features effective impedance matching, and hence it is chosen for the final fabrication. By varying the values of the parameters, we have achieved a better fractional bandwidth of 66.36% for both the TE and TM modes of reflection coefficient (S11).

The optimum dimensions for the unit cell structure are represented in Table 1. Figure 7 displays the results of the simulation of the reflection characteristics. The proposed design manifests the reflection characteristics for frequencies in a wide range from 11.075 GHz to 22.075 GHz by providing 11 GHz and a fractional bandwidth of 66.36%. The proposed FSS simulations S11 and S21 are presesnted in Figure 7.

3.1. Bandwidth Validation Using Fuzzy Logic Empirical Formula

The correctness of the fractional bandwidth is validated with the help of fuzzy logic. This method uses mathematical interpretations by varying the variables R1, L1, and L2 one at a time. The calculated fuzzy bandwidth is shown in Table 2. In Figure 8, empirical relations between fractional bandwidth and fuzzy bandwidth are presented.

3.2. Equivalent Circuit of the Proposed FSS

In the proposed FSS, the top-layer metallic square loop is modelled by a parallel RLC circuit with resistance R1, capacitance C1, and inductance L1. The unit cell’s central square loop and circular conductor are separated by a gap, which is represented by C3. L3 represents the centre circular metallic ring. The FR4 substrate is modelled using the transmission line component based on the telegrapher model. The substrate has series inductance L5 and L6 and shunt capacitance C5 and C6. The proposed FSS equivalent circuit is presented in Figure 9(a). Similarly, R2, C2, and L2 represent the bottom-layer square loop. C4 represents the gap, and L4 represents the centre circular conductor. The equivalent circuit model’s simulation results in the ADS tool are compared with the simulation results in CST and found small variation because of the lossy factor [31] is represented in Figure 9(b). The lumped element values are represented in Table 3.

Figure 10 depicts the proposed FSS’s surface current and E-field distribution at the resonant frequency. The field distribution is taken under normal prevalence. The figure shows that more field is accumulated around the square loop and the centre circular conductor ring. This clearly shows that the area of the square loop and the circular ring is the reason for the passband characteristics.

4. Experimental and Measured Results

The simulated results of the proposed FSS are validated, and the prototype of the designed FSS is fabricated using the PCB photolithography thickness of various designs are compared using the wet etching method. The fabricated FSS is 300 mm × 300 mm and fabricated using FR4 substrate with 24 × 24-unit cells.

The total number of 564-unit cells fabricated on either side of the FR4 substrate and the proposed prototype is represented in Figure 11. In the mid of the RF absorbing wall, the fabricated FSS prototype is placed. It is excited by a transmitter horn antenna situated 2 metres apart and is received by a receiver horn antenna, respectively, as shown in Figures 12(a) and 12(b).

The absorbers in the RF absorbing wall avoid the measurement error by decreasing the EM diffraction at a large incident angle. In order to measure the range of frequency between 11.075 GHz and 22.075 GHz, two different kinds of horn antennas are used. The horn antenna used has an operating range between 1∼12 GHz and 12∼18 GHz. Figures 12(a) and 12(b) depict their corresponding measurement setup. The S11—reflection coefficient is measured as plotted in the figure, and the measured FBW is 64.15%. The fabrication error is the cause of the small discrepancy between the measured and simulated findings. The comparison of the suggested FSS’s simulated and measured reflection coefficients is shown in Figure 13. The measured AoI is represented in Figure 14, lower shift and a higher shift in the frequency are noted due to the fabrication impact.

Table 4 demonstrates the comparison of the proposed work with existing work in the literature. The thickness of various designs is compared and reported as a 45% thickness reduction from the existing work. Table 5 is the proposed design structure which is associated with square loop and circular ring hence results in obtaining wideband frequency and is found to be novel compared to the literature.

The wavelength considered for computation in the comparison table concerns a lower cut-off frequency. Hence, it is more suitable for stealth applications as it demands reduction in thickness, compact design, minimum reflection, and reduced weight. The overall unit cell size is 0.46 λ0 × 0.46 λ0. The proposed work successfully achieves wide bandwidth and minimal thickness, which is the biggest hurdle for stealth applications. The designed FSS could be used for stealth, radomes, and EM shelter applications. As a future enhancement, conformal substrate can be used instead of FR4 substrate.

5. Conclusion

The unit cell size for the proposed single-layer FSS is 0.46 λ0 × 0.46 λ0. The proposed FSS structure well maintains stability in resonance due to its miniaturized unit cell. Apart from that, the metallic square loop and circular conductor exert considerable influence over the passband properties. Hence, the square loop’s area is altered to achieve working frequency. On a FR4 substrate, the designed FSS is fabricated and measured. The proposed prototype is fabricated and measured; the measured results agree with the findings of simulations, validating the proposed structure. The proposed FSS resonates at 16.5 GHz, and it has good stability under various incident angles. The proposed work helps to reduce the reflections in the aircraft at a low cost, designed an FSS with minimum thickness for stealth applications and achieved wideband frequency. The proposed single-layer double Sided FSS is compact with a minimum thickness and covers X, Ku, and Ka bands with a fractional bandwidth of 66.36% as a wideband bandpass filter. The wide bandpass filter resonates in the X, Ku, and K band in the simulation since it is measured for 1–18 GHz, and the experimental result lies in the X and Ku band. The overall thickness is reduced. Furthermore, the FSS prototype shows a stable response for both polarizations. The designed and developed single-layer FSS includes the following features, which have been shown to be useful in applications for radars, EM shelters, and stealth.

Data Availability

No data were used to support this study.

Conflicts of Interest

The authors declare that they have no conflicts of interest.