Design of Plasmon Absorbing Structure Suitable for Super High Frequency

Abstract

This paper proposed a plasmonic absorbing structure suitable for super high frequency. The plasmon absorbing structure is a periodic square cavity structure; the bottom of the cavity is a metal plate, the wall of the cavity is a polyester plate, and metal frames loaded with lumped resistors are printed on both sides of the polyester plate. The transmission characteristics of the absorbing structure are studied using the finite difference time domain method. The results show that the plasmon resonance characteristics can be effectively improved by loading the lumped resistance reasonably. The absorption rate of the absorbing structure is over 80% in the 4.3–21.5 GHz frequency band, and in the 4.3–7.7 GHz and 14.2–21.5 GHz frequency bands, the absorption rate is around 90%. Under different polarization modes, it is less sensitive to the incident angle. The transmission response of the absorbing structure is measured in a microwave anechoic chamber, and the measurement results agree well with the simulation results. After replacing the metal bottom plate of the absorber with a metal cavity, the radar cross-section of the cavity is reduced by 99.45% at 10 GHz. It proves that the designed and fabricated absorbing structure has the broadband absorbing ability, high angular stability, and broad application prospects.

1. Introduction

Radar stealth technology has always been an important research direction in the military field, with target detection changing from single-station radar to multi-station radar detection. However, by reducing the single-station radar cross section (RCS), it becomes useless in comparison to multi-station radar. To meet target stealth design requirements of the new era and to further reduce the probability of a target being detected by the enemy’s radar, it is necessary to convert detection waves into another form of energy absorption and consumption and avoid reflection in all directions. As a new type of stealth surface, a frequency selective surface (FSS) [1,2] is widely used in the design of target stealth. The frequency selective surface absorber (FSSA) [3,4] has been developed on the basis that it has shown its excellent application prospect and has outstanding broadband absorption ability and angular stability.

Common absorbing structures have both 2-D and 3-D structures. Two-dimensional absorbing structures generally use a traditional metal-dielectric FSS as the resonant unit. When the electromagnetic wave occurs on the surface of the FSS, an induced current will be generated on the surface of the FSS. The resulting scattered field and the incident field will be superimposed to form an area capable of space absorption, thereby absorbing and dissipating the incident wave [5]. The typical absorbing structure generally comprises an FSS on the top layer, the dielectric substrate in the middle, and a second FSS on the bottom [6,7,8]. However, its absorption range is narrow and sensitive to the polarization mode and incident angle. To broaden the absorption bandwidth of the FSS and reduce polarization sensitivity, changing and optimizing the structure of the FSS unit has long been a popular research topic amongst scholars. Although the multi-mode resonant unit structure [9,10] and fractal structure [11] can improve the working bandwidth to a certain extent, the design of this method is complex, and its capability to increase the absorption bandwidth is relatively limited. Compared with the optimization of the unit structure, multi-screen coupling [12,13,14] can broaden the working bandwidth to a certain extent. It arranges the FSS metal screens at a certain distance in the vertical direction of the periodic array and uses dielectric loading between the collections. Multi-screen coupling can effectively improve the mismatch loss of the single-layer structure, thereby increasing the working bandwidth. However, as the number of layers grows, the size of the absorbing structure also increases, and its application range is limited.

Subsequently, using lumped elements can improve the quality factor of the absorbing structure, reduce transmission loss, and expand the absorption bandwidth [15,16,17,18], and this is gradually becoming a well-known practice. Loading a lumped resistor in the center of the dipole can broaden the initially narrow absorption band more widely [19]. With the complexity of the structure and the increase in the number of lumped resistors introduced, the absorption bandwidth and angular stability of the absorbing structure have been further improved by coupling the two-layer metal patch loaded with resistors to the aperture [20]. The introduction of lumped resistance based on the multi-mode resonant unit structure can further broaden the absorption bandwidth [21]. Adding lumped resistors can improve the absorption bandwidth and angle stability of the absorbing structure. More importantly, we can realize the active control of the absorption characteristics of the absorbing structure by changing the resistance value without changing the cell size. When the lumped resistors are photoresistors, the transmission characteristics of the absorber are actively controlled by the external light intensity. As the light intensity changes, its transmission bandwidth and amplitude will change [22]. If the lumped resistors are thermistors, varistors, etc., we can further expand the usage scenarios and application range of the absorbing structure.

With the increase in the number of multi-screen coupling layers and the change in the coupling method, more scholars are studying the transmission characteristics of the 3-D absorbing structure [23,24,25,26]. The 3-D absorbing structure is more complex and diversified, and it has a wider working band, a low insertion loss, and a low incident angle sensitivity. It significantly broadens the scope of application of the 3-D absorbing structure. The working principle and structure of the 3-D absorbing structure are relatively complicated. Usually, the waveguide structure and FSS are combined, and the broadband 3-D absorbing structure is designed by using the complementary resonance characteristics of the two. When the FSS is on the inner wall of the waveguide cavity, its absorption principle is to use spoof surface plasmon polaritons (SSPPs) [27,28,29] to confine the incident wave to the interface and then dissipate it. The surface plasmonic polaritons (SPPs) method [30] is mostly used to design infrared band absorbers. It can achieve ultra-broad infrared band absorption through the non-close-packed plasmonic microcavity structure [31]. Unlike surface plasmonic polaritons, artificial surface plasmons have greater electromagnetic wave confinement capabilities. The field strength decays exponentially in the vertical direction, which can reduce the interference to adjacent structures, but it is also conducive to the miniaturization design of the absorber.

In contrast, the composite unit structure and multi-screen coupling make the absorbing structure challenging to design and process. An external circuit must be introduced if active devices are loaded, and this could be more inconvenient. Therefore, in this paper, a 2-D absorbing structure was designed by loading lumped elements and applying multi-layer coupling. On this basis, a 3-D absorbing structure was designed using the SSPP method, and the absorption characteristics of the absorbing structure were analyzed, calculated, and verified by experiments.

2. Theory and Design

2.1. Presentation of the Proposed Structure

The 2-D absorbing structure designed in this paper is shown in Figure 1. It comprises a metal backplane, a poly resin (PR) medium, and a metal frame loaded with lumped resistors. The openings are set in the center of the four sides of the metal frame in order to add the lumped resistors. The thickness of the PR layer is h, and the dielectric constant is ${\epsilon }_r=4.3, \tan \delta =0.025$, where $\tan \delta$ is the loss tangent of the material. The side length of the square metal frame is P, and the metal line width is w. The lumped resistors adopt 0402 chip resistors, and their size is represented by l, their width is consistent with the width of the metal line, the resistance value is R, and the width of the absorbing structure unit is D. Figure 2 is a schematic diagram of the structural size of the 3-D absorbing structure. It is a sandwich structure composed of lumped resistors FSS-PR-lumped resistors FSS, that is perpendicular to the periodic square cavity structure formed by the cross combination of metal plates. The depth of the square cavity is D, The thickness of the wall of the square cavity is H, and the length, width, and depth dimensions are consistent. On the inner surface of the square cavity, a metal line loaded with a lumped resistor is connected in the center of the 2-D absorbing structure square metal frame and is parallel to the outer metal lines. The distance between the metal wires is a. The lumped resistors on the metal lines along the depth direction of the square cavity are removed. These parameters are as follows: R = 300 Ω, D = 10 mm, P = 8 mm, H = 1 mm, w = 0.5 mm, h = 3.5 mm, l = 1 mm, and a = 3.25 mm.

2.2. Design Principles

Compared with other unit forms, such as dipoles, Y-shaped vibrators, and circular rings, the patch-type square ring unit is of great help in expanding the working bandwidth and improving the angular stability. In this design, the absorbing structure uses a square ring metal patch to load a lumped resistor.

When the electromagnetic wave is vertically incident on the 2-D absorbing structure, the equivalent circuit is shown in Figure 3. The upper surface of the absorbing structure is air and its impedance is $Z_0 =377 \mathsf{\Omega }$. The PR medium can be equivalent to a transmission line with a length of h. $Z_t$ is the impedance of the FSS loaded with lumped resistors on the surface.

The equal impedance is

$$Z_t=R+j\omega L+\frac{1}{j\omega C}$$

(1)

where L is the equivalent inductance, C is the equivalent capacitance, $\omega$ is the angular frequency of the electromagnetic wave, and j² = −1. The lumped resistance is R, and $Z_b$ is the impedance of the metal backplane, according to the transmission line theory.

Equation (2) [19] is the transmission matrix of the absorbing structure.

$$\left[\begin{array}{cc}A& B\\ C& D\end{array}\right]=\left[\begin{array}{cc}1& 0\\ 1/Z_t& 1\end{array}\right]\left[\begin{array}{cc}\cos \theta & j{Z}_0\sin \theta \\ j\sin \theta /Z_0& \cos \theta \end{array}\right]\left[\begin{array}{cc}1& 0\\ 1/Z_b& 1\end{array}\right]$$

(2)

$$\theta =\beta \cdot h=\frac{2\pi }{\lambda }\cdot h=\frac{2\pi h\cdot f}{c}$$

(3)

Equation (3) is the expression for θ, where f is the transmission frequency, c is the propagation speed of electromagnetic waves in a vacuum, and λ is the wavelength. The reflection parameter $S_{11}$ of the absorbing structure is calculated by

$$S_{11}=\frac{A+B-C-D}{A+B+C+D}=\frac{Z_{0}M\cot \theta +j{Z}_0(Z_t-Z_b-Z_0)}{(Z_{0}M+2N)\cot \theta +j({Z_0}^2+Z_{0}M+2N)}$$

(4)

and the transmission parameter $S_{21}$ is calculated by

$$S_{21}=\frac{2}{A+B+C+D}=\frac{2Z_{0}N}{Z_0(Z_{0}M+2N)\cos \theta +j[{Z_0}^2(Z_0+M)+2Z_{0}N]\sin \theta }$$

(5)

where ${M=Z}_t{+Z}_b{, N=Z}_t{Z}_b$. Since the lower surface of the absorbing structure is a metal bottom plate, it can be regarded as a grounding treatment, so $Z_b=0$. We can obtain $S_{21}=0$ from Equation (5). The absorption rate of the absorbing structure is calculated by Equation (6). Since $S_{21}=0$, the absorption rate becomes $A\left(\omega \right)=1 - {\left|S_{11}\left(\omega \right)\right|}^2$, and $S_{11}$ is shown in Equation (7).

$$A(\omega )=1-{\left|S_{11}(\omega )\right|}^2-{\left|S_{21}(\omega )\right|}^2$$

(6)

$$S_{11}=\frac{Z_0{Z}_t\cot \theta +j{Z}_0(Z_t-Z_0)}{Z_0{Z}_t\cot \theta +j{Z}_0(Z_t+Z_0)}$$

(7)

When $\mathsf{\theta }\to \pi /2$ and $S_{11}\to 0$, we can obtain the stop band of the absorbing structure, and $h \approx \lambda /4$ can be calculated in this case. When the resonance center of the top layer of the absorbing structure is $f_0$, the resonance frequency becomes $f=f_0/\sqrt{{\epsilon }_r}$ due to the intervention of PR. When the resonance center of the absorbing structure is designed to be f = 10 GHz, we can obtain $f_0=20.7 \mathrm{GHz}$ and $h \approx 3.6 \mathrm{mm}$.

The structure of the 3-D absorbing structure designed in this paper is complex and cannot be directly analyzed by the equivalent circuit method. Still, it can be regarded as the cross-distribution of the FSS in the x and y directions. When the electromagnetic wave is vertically incident on the bottom surface of the absorbing structure along the z direction, the wave vector direction is parallel to the cavity wall. The cavity wall is equivalent to the SSPP transmission line in this case. Based on surface plasmonic polaritons (SPPs), the SSPP method etches periodic structures on the metal surface or arranges periodic metal patches on the dielectric layer. When the electromagnetic wave is incident on the interface between the metal and the medium, the free electrons in the metal conductor oscillate collectively, generating surface plasmons in the microwave frequency range. The electromagnetic field strength reaches its peak at the interface between the metal and the medium, and the energy propagates along the structure’s surface. It is wholly bound near the surface of the structure, thereby achieving the effect of absorbing waves. Compared with SPPs, SSPPs have more substantial electromagnetic wave confinement capabilities, and the field strength decays exponentially in the vertical direction.

The 3-D absorbing structure designed in this paper is composed of a 2-D FSS vertical distribution. For two FSSs perpendicular to each other, when the FSS on the xoz plane is in the transverse magnetic (TM) mode, the FSS on the yoz plane is in the transverse electric (TE) mode, and vice versa. According to the SSPP theory, the electromagnetic waves in the TE mode cannot induce polarized charges on the interface because there is no electric field component perpendicular to the interface. For certain surfaces, SSPP transmission lines can only work in the TM mode. However, the 3-D absorbing structure exists with the metal patch in the TM mode, whether in the x or y direction. Therefore, the 3-D absorbing structure can support the SSPP theory in the TE and TM modes. This paper uses CST studio software to analyze its transmission characteristics. Finally, the 3-D absorbing structure is fabricated, and its transmission characteristics are tested in a microwave anechoic chamber.

3. Analysis and Discussion

3.1. Two-Dimensional Absorbing Structure

First, calculate the established equivalent circuit with the help of ADS software. The center frequency is 10 GHz, L = 24 nH, and C = 0.047 pF can be obtained through calculation. Then the reflection coefficient can be obtained through Equation (7). Figure 4 compares reflection coefficients calculated by the equivalent circuit and full-wave methods. We can see that at 8 GHz, the reflection coefficient calculated by the equivalent circuit method is slightly smaller than that of the full-wave process; at 8 GHz, the reflection coefficient calculated by the equivalent circuit method is marginally larger than that of the full-wave simulation method. However, the absorbing bandwidth calculated by the two methods is the same, which proves the reliability of the absorbing structure design in this paper. The transfer characteristics of the absorbing structure loaded with lumped resistors and the absorbing structure without lumped resistors are compared first to study the influence of lumped resistors on the absorbing structure. Figure 5 shows the distribution of the reflection coefficient and absorption rate at 0–25 GHz when the electromagnetic wave is vertically incident. As can be seen from the figure, the absorbing structure without lumped resistors has two extremely narrow discontinuous absorption bands ($A > {0.9, S}_{11} < -10 \mathrm{dB}$), which are concentrated around 19 GHz and 22 GHz regions. The two extremely narrow frequency bands gradually move to the low frequency when the lumped resistors are added. Then, the reflection coefficient at the center frequency ($f_m$) drops to −11 dB, and the absorption bandwidth widens to 6.3–14.5 GHz, the absolute bandwidth is 8.2 GHz, and the fractional bandwidth is 78.8%, $f_m =10.4 \mathrm{GHz}$. The incident wave and absorbing structure generate a strong electromagnetic resonance in this frequency band and excite the induced current. The sense current flows through the lumped resistor and is dissipated through heat loss. The above results show that loading lumped resistors broaden the absorption frequency band and lower the resonance frequency, making the equivalent absorbing structure’s impedance match the free space impedance on a broader frequency band.

According to the design principle of the traditional patch-shaped FSS, the PR layer in the free space will cause electromagnetic waves to reflect on the interface between the air and the medium because the impedance of the medium does not match the impedance of the free space. The bottom surface of the absorbing structure is a metal plate. When the electromagnetic wave passes through the PR layer, it will be completely reflected. When $\lambda /4 < h < \lambda /2$, the reflection of the bottom metal plate will cancel the reflection of the FSS. When $h < \lambda /4$, the reflection through the PR layer is superimposed with the reflection of the FSS. Therefore, it is necessary to optimize the calculation of the thickness of the PR layer. Figure 6 shows the effect of the thickness of the PR layer on the reflection coefficient of the absorbing structure within 4–18 GHz. When h = 2.5 mm, the reflection coefficient in the whole frequency band is greater than −10 dB. As h increases, the reflection coefficient of the resonant band ($f_h$) to the right of $f_m$ gradually decreases below −10 dB. Following h ≥ 4 mm, then $S_{11}\left(f_h\right) > -10 \mathrm{dB}$. The reflection coefficient of the resonant frequency band ($f_l$) on the left side of $f_m$ drops and then increases, but they are all less than −10 dB. In general, the thickness of the PR layer has a significant influence on the boundary with the absorbing structure absorption frequency band, thus affecting the resonance depth of the absorbing structure. In this paper, h = 3.5 mm is finally selected. In this case, the absorption frequency band is continuous, and the width is the largest.

In the expression of $Z_t$, the resistance value of the lumped resistor accounts for a large proportion. Figure 7 shows the influence of the lumped resistance value on the transmission characteristics of the absorbing structure. It can be seen from the figure that when R = 110 Ω, the absorption frequency band of $S_{11 }< -10 \mathrm{dB}$ is discontinuous, and $S_{11} > -10 \mathrm{dB}$ was in the 9–11.4 GHz frequency band. As the resistance value of the lumped resistor increased, $f_h$ gradually moved to the left, and the reflection coefficient gradually increased; the reflection coefficient of $f_m$ gradually decreased; $f_l$ slowly moved to the right, and the reflection coefficient remained unchanged. Meanwhile, with the increase of R, the absorption bandwidth of $S_{11} < -10 \mathrm{dB}$ gradually decreased. When R = 130 Ω, the absolute bandwidth was 8.2 GHz, and the fractional bandwidth was 78.8%. When the resistance value increased to R = 170 Ω, the absorption bandwidth is shortened to 6.7–14 GHz, the absolute bandwidth was reduced to 7.3 GHz, and the fractional bandwidth became 70.5%. Figure 8 shows the surface electric strength distribution at different frequencies. Due to the symmetrical structure of the absorbing structure, the surface electric field was symmetrical about the xoz and yoz planes, and the electric field intensity was concentrated in the square metal ring and its vicinity. Within the absorption frequency band, the electric field strength near the lumped resistance was more robust, which means that the induced current was consumed after flowing through the lumped resistance. The abovementioned results show that when R = 130 Ω, the absorbing structure has the widest absorption bandwidth.

Figure 9 shows the influence of the incident angle on the absorption rate of the absorbing structure under different polarization modes. In the TE mode, as the electromagnetic wave incident angle increased to θ = 50°, the absorption rate of the absorbing structure in the 6.5–15.5 GHz frequency band was still above 80%. The absorption bandwidth gradually shortened as the incident angle increases in the TM mode. When the electromagnetic wave incident angle grew to θ = 50°, the absorption bandwidth of the absorbing structure shrinks to 6.9–11.6 GHz. According to the principle of LC oscillation, appropriately increasing the length of the FSS resonant unit can improve its transmission characteristics. However, increasing the size of the resonant square will also lead to a smaller spacing between FSS units, weakening the transmission characteristics of the FSS. Figure 10 shows the influence of the incident angle on the absorbing structure absorption rate when P = 8 mm. When it was vertically incident, the absorption bandwidth of the absorbing structure was 7–15.2 GHz, and the absolute bandwidth was 8.2 GHz. Compared with the model with P = 6.5 mm, the absorption bandwidth hardly changed much. However, in the TM mode, as the incident angle increased to θ = 50°, the absorption rate of the absorbing structure with P = 6.5 mm in the 12–14 GHz frequency band dropped to about 50%, and the fluctuation was significant. However, the absorbing structure with P = 8 mm had an absorption rate of more than 70% in this frequency band, and the fluctuation was minimal. The abovementioned results show that increasing the size of the resonant unit can appropriately improve the angular stability of the absorbing structure. Still, it will also reduce the absorption rate at the center frequency. Increasing the resonant unit’s size will inevitably make the absorption bandwidth of the absorbing structure discontinuous, so there is a limit to the increase in the size of the resonant unit. In this paper, the size of the resonant unit was P = 8 mm.

3.2. Three-Dimensional Absorbing Structure

To broaden the absorption bandwidth of the absorbing structure and improve the angular stability, a 3-D absorbing structure was designed based on the 2-D absorbing structure. The 3-D absorbing structure uses a resonant unit similar to the 2-D absorbing structure, but the two working principles differ. The 3-D absorbing structure uses the SSPP method to bind the electromagnetic wave on the wall of the absorbing cavity for dissipation. The 3-D absorbing structure designed in this paper loaded the metal wires with lumped resistors in the center of the 2-D FSS square metal frame and parallel to the outer metal lines, removing the lumped resistors on the metal wires along the depth direction of the square cavity, and increasing the resistance of the lumped resistors. This method can increase the binding capacity of electromagnetic waves. The medium’s thickness and the lumped resistance value were optimized through the CST simulation software to increase the coupling capability of the adjacent absorbing cavities, and finally, H = 1 mm and R = 300 Ω were obtained. Figure 11 shows the absorption rate of different models. The inner wall of the square cavity of model 1 is the same as the top surface of the 2-D absorbing structure. Model 2 has three metal wires loaded with lumped resistance connected in parallel, and model 3 has four metal wires loaded with lumped resistance connected in parallel. The results show that the parallel connection of the metal lines loaded with lumped resistors can significantly broaden the absorption bandwidth of the absorbing structure. Compared with model 2, the absorption of model 3 in the 5–20 GHz frequency band decreased, and excessively increasing the parallel branch would reduce the absorbing structure’s absorption ability; therefore, model 2’s absorption ability is better.

Figure 12 shows the energy loss distribution of the three models at different frequencies in the TE mode. When the incident frequency is low, the resonance intensity is weak. The resonance points are mainly concentrated near the lumped resistance and the metal lines along the depth of the square cavity. Due to the hindering effect of the lumped resistance of the metal wire along the depth of the square cavity of model 1, the resonance point is mainly concentrated in the upper half of the metal wire frame. When the incident frequency is increased to 15 GHz, the resonance area of model 1 is no longer limited to metal wires but is mainly concentrated on the yoz wall. In contrast, the resonance characteristics of the xoz wall are weak. Due to the increase of parallel branches, the resonance area of model 2 is concentrated near the lumped resistance and the metal wires along the depth of the square cavity. At the same time, the metal wire frame on the xoz wall also produces strong resonance. As the parallel branches continue to increase, the resonance characteristics of the yoz wall are also strong. In contrast, the resonance characteristics of the metal wire frame on the xoz wall are slightly weakened, but the change is small. When the incident frequency increases to 20 GHz, the resonance characteristics of the metal wires on the wall of the square cavity and the lumped resistors at the open and bottom ends continue to increase. In contrast, the resonance characteristics of the lumped resistors in the middle position are weakened. In summary, introducing a parallel branch loaded with lumped resistors will enhance the coupling resonance between the two perpendicular walls, thereby enhancing the wave-absorbing capability of the absorbing structure. When the frequency is high, the resonance characteristics of the parallel branch at the middle position are weakened, and continuing to increase the number of parallel branches minimally helps in increasing the absorption ability of the absorbing structure.

Figure 13 shows the reflection coefficient and absorption rate distribution of the 3-D absorbing structure and the 2-D absorbing structure in the 0–25 GHz band when the electromagnetic wave is vertically incident. The 3-D absorbing structure can achieve a more than 80% absorption rate in the 4.3–21.5 GHz band, the absolute bandwidth is 17.2 GHz, and the relative bandwidth is 133.3%; the absorption rate in the 4.3–7.7 GHz and 14.2–21.5 GHz bands is above 90%. Compared with the 2-D absorbing structure, although the absorption rate of the center frequency drops to 80%, the overall absorption bandwidth has significantly broadened.

Figure 14 shows the influence of the incident angle on the absorption coefficient of the 3-D absorbing structure under different polarization modes. The results show that in the TE mode, as the incident angle of the electromagnetic wave increased to θ = 50°, the absorption rate of the absorbing structure at the center frequency decreased to 72%. The absorption bandwidth on the left side of the center frequency point was almost unchanged, and the absorption bandwidth on the right side gradually narrowed. Still, the overall absorption rate remained above 72% in the range of 4.3–19 GHz. In the TM mode, as the incident angle increased to θ = 50°, the absorptivity fluctuated strongly. The absorption rate of the frequency band on the left side of the center frequency gradually increased, and the absorption rate of the frequency band on the right side gradually decreased but remains above 80%. In summary, with the increase of the incident angle, the absorption bandwidth of the 3-D absorbing structure gradually narrowed, but the decrease range was not extensive, and the angular stability was better.

4. Experimental Measurement

A 3-D absorbing structure was manufactured and processed using a printed circuit board technology to verify the simulation results. Its size is 220 mm × 220 mm × 10 mm, composed of 20 × 20 square cavity unit arrays. The fabricated 3-D absorbing structure is shown in Figure 15. Firstly, the metal wire frame was printed on both sides of the PR4 dielectric substrate with a 220 mm × 10 mm × 10 mm size and H = 1 mm using a PCB printed circuit board, and then the 0402 chip resistor (RC0402FR-075K1L from YAGEO China) was soldered to the metal wire frame. Finally, the fabricated strip circuit board was fixed crosswise on the metal base plate.

The test system is shown in Figure 16a, which mainly includes an anechoic chamber, a vector network analyzer, and a data processing terminal. The anechoic chamber has a rectangular shape with a size of 5 m × 10 m × 6 m. The control system adjusts the turntable in the anechoic chamber to different angles. Then, the test signal was sent out from the standard gain antenna aimed at the target, and another standard gain antenna receives the echo signal. Then, the echo signal was sent to a vector network analyzer (Agilent E8363A VNA, Agilent Technologies, Palo Alto, CA, USA), that is used to cancel the time domain. Then, we measured the calibration body that has a metal plate with a 120 mm × 120 mm size. Next, we used a sub-particle board with a thickness of 5 mm to make the supporting base of the measurement target, and we placed the fabricated absorbing structure in the center of the turntable as in Figure 16b and used the wave-absorbing wedge to shield the turntable part. Limited by the laboratory antenna specifications, the absorbing structure was tested at 1–18 GHz, the test angle range was −50~50°, and the test angle interval was 0.1°. The absorbing structure was placed vertically on the turntable. Finally, we used the echo signal in the −50~50° angle domain and the 1–18 GHz frequency domain to perform inverse synthetic aperture radar (ISAR) imaging of the measurement target.

Figure 17 compares the absorption rate between the experimental test and the numerical simulation under different incident angles. The results show that when the electromagnetic wave was vertically incident, the absorption rate of the absorbing structure in the 1–6 GHz frequency band was slightly larger than the numerical simulation results. The difference gradually decreased as the frequency increased. In the 6–18 GHz frequency band, the variation trend of the absorbing structure absorption rate with the frequency was the same. When f = 10 GHz, the absorption rate of the absorbing structure was 75%, which is slightly smaller than the simulation result, which may be caused by some tiny gaps in the assembly process of the absorbing structure that will reduce its absorption rate to a certain extent in the high-frequency band. As the incident angle increased in TE mode, the absorption rate fluctuated less in the 1–6 GHz frequency band. The frequency band most sensitive to the incident angle was the 10–18 GHz frequency band. In the 10–13 GHz frequency band, the absorption rate gradually decreased as the incident angle increased. In the 13–18 GHz frequency band, the change rule was the opposite. When the incident angle increased to 50°, the absorption rate at f = 12 GHz dropped to about 67%, which is slightly smaller than the simulation result. The fluctuation characteristics of the absorption rate in other frequency bands are consistent with the simulation. With the increase of the incident angle in the TM mode, the fluctuation range in the absorption bandwidth was small and the absorption rate remained above 85%. The test results at the same frequency were slightly larger than the simulation results, but the difference was slight, and the fluctuation law of the absorption rate was consistent. In summary, the experimental results of the 3-D absorbing structure designed in this paper are compatible with the numerical simulation results, proving that the FSR developed in this paper has excellent reliability.

To further verify the absorbing effect of the 3-D plasmonic absorbing structure, we replaced the metal base plate of the absorbing structure with a metal cavity size of 220 mm × 220 mm × 600 mm, and its RCS is measured in an anechoic chamber as in Figure 16c. Figure 18 is the ISAR imaging of the empty cavity and the cavity equipped with the absorbing structure, respectively. The absorbing structure can significantly reduce the echo signal at the bottom of the cavity. Figure 19 shows the RCS distribution of the cavity at 10 GHz in TE mode. At a detection angle of 0°, the reduction of the cavity by the absorbing structure was 29.2 dBsm. In the −40~40° angle range, the average RCS of the cavity equipped with the absorbing structure was −19.88 dBsm, and the average RCS of the empty cavity is 2.74 dBsm. The absorbing structure reduced the RCS of the cavity by 99.45%.

Table 1. Feature comparison between the designed absorbing structure and the reported absorbing structure.

	Absorbing Structure in Reference	Absorption Bandwidth 1/GHz	Maximum Angle of Incidence	Cell Size/mm
2-D	[6]	3.5–11.5	Not reported	20 × 20 × 8
	[7]	3–9	45°	Top: 36 × 36; Back: 54 × 54; h = 20
	[8]	3–9	30°	20 × 20 × 8
	This work	6.3–14.5	50°	10 × 10 × 10
3-D	[24]	1.5–5.5	30°	20 × 10 × 11.5
	[25]	1.5–10.5	45°	24 × 24 × 20
	[26]	2.5–11	40°	11 × 11 × 15
	This work	4.3–21.5	50°	10 × 10 × 10

¹ For 2-D structures, the absorption bandwidth is 90%; for 3-D structures, the absorption bandwidth is 80%.

compares the absorbing structure designed in this paper and the absorbing structure in the literature regarding the absorption bandwidth, maximum incidence angle, and the absorption unit’s size. The comparison shows that the absorbing structure designed in this paper had a wider absorption bandwidth, better band continuity, and lower sensitivity to the incident angle. Compared with the passive absorbing structure [32], due to the introduction of lumped resistance, the 3-D absorbing structure designed in this paper had fewer design parameters and could actively control the wave-absorbing characteristics, giving the absorbing structure a broader application prospect.

5. Conclusions

In this paper, a 2-D absorbing structure and 3-D absorbing structure loaded with lumped resistance were designed. Using the full-band analysis method, we studied the effects of the absorbing structure design parameters, polarization mode, and incident angle on the absorbing structure performance. The results show that the absorbing structures designed in this paper have excellent absorption rates and angular stability. The absorption rate of the 2-D absorbing structure in the 7–15.2 GHz frequency band is above 90%. When the incident angle increases by 50°, the absorption rate remains above 80% in the transverse electric mode. The optimized 3-D absorbing structure can achieve a more than 80% absorption rate in the range of 4.3–21.5 GHz, the absolute bandwidth is 17.2 GHz, and the fractional bandwidth is 133.3%. When the incident angle increases by 50°, the absorption bandwidth changes little, and the absorption rate remains above 80% in the transverse magnetic mode and above 72% in the transverse electric mode. The designed 3-D absorbing structure is fabricated, and its transmission characteristics are measured in the anechoic chamber. The experimental results show that the absorption rate of the designed 3-D absorbing structure is consistent with the numerical simulation results, which proves that the 3-D absorbing structure developed in this paper has good angular stability while realizing a broadband absorption ability. Compared with traditional absorbers, this 3-D absorbing structure has a smaller structure size and thickness and has good application prospects in the structural stealth integration design of aircraft and other targets.