Absorption mode

The multifunctional circuit model of the absorber/antenna is shown in Fig. 1. When switches “S ₁”, “S ₂”, and “B” are in contact, the circuit model transforms into the absorber's typical equivalent circuit model. The equivalent circuit theory is employed to analyze the characteristic of absorbing structures [Reference Costa, Monorchio and Manara52, Reference Li, Xiao, Bai and Wang53]. The equivalent input impedance of a typical single-port absorber can be defined as:

(1)$$Zin_{ab.} = R + jX$$

if the structure exhibits lumped components, i.e. resistors, capacitors, or inductors then (1) can be re-written as:

(2)$$Zin_{ab.} = R + jX = Z_L + R_L + jX$$

It should be noted that in equation (2), Z _L represents the lumped components, and R _L and jX represent the equivalent resistance and reactance caused by the absorbing structure and absorber's material. Therefore, the operation of an absorber is described in Figs 1(a) and 1(c), where the excitation is the incident-free space plane wave.

Radiation mode

The working principle and operation of the radiator are shown in Figs 1(a) and 1(d). To describe the multi-functional circuit model as the radiator's equivalent circuit model as shown in Fig. 1(a), the switches S ₁ and S ₂ have to be in-contact with “A” or the switches S ₁ and S ₂ are in-contact with some source signal. The input impedance of the antenna of a typical antenna can be written as:

(3)$$Zin_{ant.} = R + jX = R_L + R_r + jX$$

In equation (3), R _r represents radiation resistance, which shows the radiated power into the free space, R _L is the loss resistance caused by diverse losses, X is the reactance of the antenna.

It should be noted that by selecting suitable Z _L, the matching between free space and physical structure has to be preserved in both cases. Switch S ₂ is required in the physical structure because operation as an absorber/antenna is only determined by switch S ₁. In the physical structure, the switch S ₂ is realized by embedding two p-i-n diodes. When the p-i-n diodes are switched ON, the disc shape patches are in contact with some input source signal for EM radiation.

Design and analysis of the absorber

The schematic view of the designed unit cell is shown in Fig. 2. The unit cell is composed of multi-layers having two F4B substrates with a dielectric constant of 2.65. The resonator consists of a split-ring and disk-shaped patch. To obtain the favored operating frequency and optimized unit cell, a straightforward design of a ring patch has been designed and studied in [Reference Mat, Misran, Islam and Mansor54]. The optimized geometry of the unit cell as shown in Fig. 2(a) is determined as the periodicity p = 20.5 mm, split-ring width s = 0.5 mm, split ring radius R1 = 9.7 mm, and disk-shaped patch radius Rp = 5.7 mm, where the thickness of the upper substrate is h1 = 3 mm and lower substrate h2 = 1 mm. The lumped resistors with resistances equal to 470 Ω are embedded in the split ring to enhance the absorption rate. The simulations were performed by CST Microwave Studio using floquet port excitation. The schematic view of the experimental setup is displayed in Fig. 1(b). In Fig. 1(b), one horn operates as a transmitting horn and the other horn as a receiving horn antenna. The sample is set in front of them. A sheet of copper having the same dimensions as the designed structure was fixed in front of horns and measured for calibration. The fabricated sample was set at the same place to measure the reflection. To design an absorber with a high absorption rate, the coupled incident power should be comparable to the incident EM wave. Different configurations and their simulated results of the proposed units are displayed in Fig. 3. From Fig. 3(b), it is clear that the unit with lumped resistor has high absorptivity. The physical parameters of the unit cell as shown in Fig. 3, the resistance of the lumped resistors is equal to 470Ω. The absorptivity of the absorber is calculated with A = 1 − |S ₁₁|² − |S ₂₁|², where |S ₂₁|² ≃ 0 due to the shielding of the sandwiched copper's plate. Hence, the absorptivity can be rewritten as A = 1 − |S ₁₁|². In Fig. 4(a), the designed EM absorber's absorptivity increases in the frequency band from 6.2 to 8.2 GHz. The simulated and measured results are in good agreement as depicted in Fig. 4(a). The designed surface is polarization-insensitive for incident EM waves. From Fig. 4(b), it can be found that for different oblique incident angles (from 0° to 30°), the absorption characteristics are unchanged. Also, there is a spurious resonance under TM polarization when the oblique incidence increases. This is because the high order harmonic resonance will shift toward the lower frequencies at oblique incidence [Reference Munk50, Reference Luebbers and Munk51]. To emphasize the absorption phenomenon of the absorption, we plotted the E-field and the surface current distribution in Figs 5(a)–(d). From Figs 5(a) and 5(b), it is observed that the E-field is largely concentrated around the vertical resistors which shows that the lumped resistors played an important role in getting the high absorptivity. The surface current distribution is illustrated in Figs 5(c) and 5(d). From Figs 5(c) and 5(d), it is observed that the direction of the surface current on the top layer and ground plane is anti-parallel. In Figs 6(a) and 6(b), the parametric study has been carried out to show the impact on absorptivity due to lumped resistors and height of the substrate. It can be found that the height of the substrate contributes to achieving wide bandwidth where the high value of the lumped resistor contributes to improving the absorptivity of the absorber.

Fig. 2.

Configuration of the proposed unit cell: (a) top view, (b) side view, (c) fabricated prototype front side, and (d) backside.

Fig. 3.

(a) Different configurations of the presented units and (b) their simulated results.

Fig. 4.

(a) The simulated and measured absorptivity at normal incidence and (b) at different oblique incident angles for TE and TM modes.

Fig. 5.

The electric field distributions (a) at the top and (b) bottom surfaces and the surface current distributions (c) at the top and (d) bottom surfaces of the proposed unit at the frequency of absorption (7.5 GHz).

Fig. 6.

The impact on bandwidth and absorptivity when (a) the height of substrate and (b) value of the lumped resistors are varying.

Design and analysis of the low RCS antenna array

The FSS-based basic element of the planar antenna array for the proposed structure is demonstrated in Fig. 1(a). Initially, 2 × 2 array is designed as shown in Fig. 7(a).

Fig. 7.

Geometry of the four-element planar array, (a) front view, (b) bottom view and it simulated results, (c) S11 and gain, (d) directivity and efficiency, (e) 2-D radiation pattern, and (f) 3-D radiation pattern.

An additional F4B substrate is added below the ground to design a feeding network that incorporates T-junctions that excite individual elements as shown in Fig. 7(b). All circular patches are connected to the feeding network through metalized via of a diameter of 0.7 mm. The feeding network is connected to a single 50 Ω feed as shown in Fig. 7(b). The simulation of the planar antenna array is performed by CST MWS. In the simulation, all boundary conditions are set to be the open boundary. The simulated reflection coefficient (S11) realized gain, directivity, and efficiency of the basic four planar antenna array elements are given in Figs 7(c) and 7(d). It can be concluded from Fig. 7(c) that a good impedance matching bandwidth <− 10 dB, i.e. from 7.6 to 8.1 GHz has been achieved. From the realized gain, it is concluded that the basic four-element planar array exhibits a maximum peak of about 8 dB in the propagation direction. The 2-D and 3-D radiation patterns are given in Figs 7(e) and 7(f). The polar pattern for both phi = 0° and 90° is nearly the same and also the main lobe of the polar and 3-D radiation pattern points towards the normal direction.

In order to design a high gain antenna, an array of 4 × 4 elements array is proposed and designed as shown in Figs 2(c) and 8(a). The total dimension of the surface is 2.1λ × 2.1λ at the resonance frequency of 7.8 GHz. In the circuit model of antenna and absorber, the switch S2 is required in the physical structure when the structure works as an antenna/radiator. Therefore, two p-i-n switches from SKYWORKS SMP1320 − 079 LF in the SC − 79 packages [49] are incorporated in the feeding network. The structure is regarded as an antenna/radiator when the p-i-n switches are switched ON, which means that the circular patches and feeding network areconnected to some input source signal as shown in Figs 1(a) and 1(d). The DC bias circuits of the p-i-n diodes are also illustrated in Fig. 8(b). As shown in Fig. 8(b), diode D1 and D2 are powered by the DC source (Vcc). The capacitor and inductor are embedded to isolate the RF and DC sections. The diodes D1 and D2 need to be operated with some DC source by applying Vcc where D1 and D2 are switched OFF when no source is applied.

Fig. 8.

The schematic view of the (a) multi-layered structure, (b) its feeding network, (c) return loss S11 and gain, (d) directivity and efficiency of the 4 × 4 elements array, and (e) states of diodes (D1 and D2). (Note: R _OFF = 1 KΩ, C _OFF = 0.35 pF, L _CHIP = 1.5 nH, and R _ON = 0.9 Ω, L = 50 nH, C = 100 pF and V _cc = 3 V).

To evaluate the radiation performance, we measured and simulated S-parameter S11 and realized gain as presented in Fig. 8(c). It is shown that the S-parameter S11 is <− 10 dB and gain is increased linearly up to 14.5 dB over a frequency band from 7.5 to 8.0 GHz. The directivity of the antenna array is calculated as follow:

(4)$$D_{dir.} = 10 \log\left({k\times4\pi\times A\over \lambda_{o}^{2}}\right)$$

where A is the physical area of the antenna ground and k(0 ~ 1) is the effective coefficient. If k = 1, the theoretical value of the maximum directivity is given as:

(5)$$D_{dir.} = 10 \log\left({4\pi \times A\over \lambda_{o}^{2}}\right)$$

Here the area of the aperture A = 82 × 82 mm² and λ_o = 38.46 mm at f = 7.8 GHz. Using equation (5), the theoretical value of D _dir. = 17.5 dB. Where simulated directivity D _dir. = 16.12 dB as shown in Fig. 7(d). A little degredation between theoritical and simulated values may be due to lumped resistors in the outer ring or due to some other losses.

To maximize the performance of the antenna array, an optimal design of the array is carried out to maximize the array aperture efficiency. The aperture efficiency ɛ _ap of the array can be written as:

(6)$$\varepsilon_{ap} = {A_{em}\over A_{\,physical}}$$

where A _em is the antenna array's effective maximum area and A _physical is the physical area of the array aperture. The effective maximum area A _em can be written as:

(7)$$A_{em} = {\lambda ^2\over 4 \pi} \times G$$

where λ is the wavelength at f = 7.8 GHz and G is the antenna gain. The aperture illumination efficiency is calculated at working frequency 7.8 GHz. Maximum aperture efficiency of $52.9\%$ is observed at 7.8 GHz as shown in Fig. 8(d). Where the radiation efficiency is defined as ɛ _rad = G/D, where D is the directivity. The radiation efficiency is shown in Fig. 8(d), maximum radiation efficiency of $76\%$ is observed at resonance frequency 7.8 GHz.

As the lumped resistors were fixed in the outer split-ring that has little impact on the radiation performance. In Table 1, simulated results are listed to demonstrate the impact of lumped resistors. As can be seen from Table 1, a little degradation is noted in gain, directivity, and efficiency when the value of lumped resistors is increased. Where in Refs. [Reference Li, Yi, Luo, Muneer and Zhu34, Reference Rahman, Cao, Gil, Sajjad and Wang35], the antenna elements exhibitlumped resistors, these lumped resistors have a great impact on the radiation performance. To reduce the effect of the lumped resistor, in this research, the disc-shaped radiating elements without lumped resistor are kept inside the absorbing circular element.

Table 1.

Impact of lumped resistors on gain, directivity, and radiation efficiency

In Figs 9(a) and 9(b), the surface current is distributed uniformly over the circular-shaped patches and feeding network. The current is stronger in one direction of the circular-shaped patches and weaker in the other direction since each element was excited 1.2 mm away from the center of circular-shaped patches. Figure 9(b) shows that the current is equally distributed over the feeding network, where the high and low intensity of the current denotes the order of harmonics. To obtain identical radiation patterns, the disk-shaped patches were fed 1.2 mm away from the center but still at some angles the discrepancy between E- and H-plane radiation patterns is observed because the feed points impact both impedance matching as well as radiation pattern. However, the main lobe is directed along propagation while the (3 dB) angular width is about 23.50° as given in Figs 9(a) and 9(b).

Fig. 9.

The simulated surface current distribution: (a) top layer and (b) bottom layer. (c) The 2-D polar radiation pattern and (d) 3-D radiation pattern of 4 × 4 elements array.

Since EM absorbers are primarily used to convert EM energy into other forms of energy such as heat or weaken them through interference cancellation that leads to RCS reduction. To evaluate the RCS reduction performance, we simulated the proposed structure as discussed in the above sections using a time-domain solver in CST. For the x-polarized incident EM wave, the mono-static RCS of the proposed structure is compared to the equal-sized copper sheet as displayed in Fig. 10(a). From Fig. 10(a) it can be seen that the RCS of the proposed structure drops over a frequency band from 6.2 to 8.2 GHz, which ensures that the proposed structure has low RCS in both stealth and radiating bands. The structure has a maximum RCS reduction of 39.18 dB within the stealth band at 6.2 GHz and 26.12 dB RCS reduction within radiating band at 7.8 GHz. The Cartesian bistatic scattering pattern at ϕ = 0° and ϕ = 90° is also depicted in Figs 10(b) and 10(c). It can be seen that the RCS is significantly reduced in a wide angular region. When ϕ = 0° the RCS reduces in the angular region 80° ≤ θ ≤ 80° while it remains same in − 80° ≤ θ ≤ −180° and 80° ≤ θ ≤ 80°, similarly when ϕ = 0° the RCS reduces in very wide angular region − 140° ≤ θ ≤ 140° while it remains same in the small angular region, i.e. in − 140° ≤ θ ≤ −180° and 140° ≤ θ ≤ 180°. The maximum reduction is observed around about 0°. Furthermore, the 3-D RCS scattering pattern is also shown in Fig. 10(d). It can be seen that mostly incident energy is deflected toward four directions.

Fig. 10.

The scattering characteristics of the designed surface: (a) mono-static RCS, (b) bistatic RCS (ϕ = 0°), (c) bistatic RCS (ϕ = 90°), and (d) 3 -D radiation scattering pattern at 7.8 GHz.

In Table 2, we have made a comparison of some significant performance parameters of the designed structure to those reported performance parameters in previously published research works. The designed MS structure in this research work exhibits high gain, low RCS, and high radiation efficiency. To be more persuasive and eradicate the factor introduced by the larger size of the designed surface, additionally we have calculated the aperture efficiencies. In this research work, the aperture efficiency is about $45\%$, which is higher than other studies as listed in Table 2.

Table 2.

Performance comparison of the proposed structure with previously designed regular patch antennas

Note: f _r, resonance frequency; G, gain; ɛ _rad, radiation efficiency; ɛ _ap, aperture efficiency.

We further compare the novelty of this research work in Table 3. In Table 3, the previously designed structures have multiple functions to manipulate free space plane wave [Reference Sheng, Ge, Han and Zhu25–Reference Kiani, Weily and Esselle27, Reference Bakshi, Member and Mitra29, Reference Li, Cao, Liu and Wang30, Reference Li, Li, Wang, Wang and Cao33]. For example, in [Reference Sheng, Ge, Han and Zhu25], the designed structure has two alternate functions such as transmission in X-band and reflection in Ku-band. In this research study, two different switchable functions namely absorption and radiation are integrated. Compared with [Reference Li, Yi, Luo, Muneer and Zhu34], the proposed structure has a switchable function; however, the lumped resistors were embedded in radiating patch that degrades the radiation performance. The wideband absorption was achieved by embedding multiple resistors in multiple resonating rings with using the thick substrate.

Table 3.

Comparision with previously designed multifunctional EM structures

Note: Reconf., reconfigurability.