Single-layer quad-band FSR with enhanced angular stability for S- and C-band applications Authors: Asal Malekara, Changiz Ghobadi, Javad Nourinia, Amin Khalilzadegan Published: 2026 ABSTRACT AbstractA single-layer, polarization-insensitive frequency-selective rasorber (FSR) with quad-band absorption performance is proposed. The presented FSR is developed by integrating a quad-band resonant absorber and a single bandpass frequency-selective surface (FSS) on opposite sides of an FR4 substrate. In the absorption bands, the top layer effectively absorbs electromagnetic waves, while the backside FSS functions as a reflector. In the transmission band, the top layer becomes nearly transparent, allowing efficient wave passage through the backside FSS. The combination of the two layers results in effective multi-functional performance. The top layer achieves over 90% absorption at 2.14, 2.88, 6.02, and 7.32 GHz. The backside FSS generates a transmission band at 4.54 GHz with low insertion loss. The design exhibits enhanced angular stability up to 50° for both TE and TM polarizations, with polarization insensitivity ensured through four-fold rotational symmetry. The physical mechanism is analyzed using an equivalent circuit model and effective-medium analysis. A 15 × 15 prototype has been fabricated and measured, showing excellent agreement with simulations. This resistor-free single-layer FSR is suitable for radar cross-section (RCS) reduction, electromagnetic interference (EMI) mitigation, and multi-functional radome applications in S- and C-bands.Keywords: Frequency selective rasorber, Single-layer FSS, Polarization insensitive, Angular stability, EMI mitigation, Radar cross-section reductionSubject terms: Engineering, Materials science, Physics Introduction The increasing demand for compact and efficient electromagnetic (EM) devices in stealth technology, electromagnetic interference (EMI) mitigation, and RF front-end systems has attracted considerable attention on frequency selective surfaces (FSSs)1. These periodic structures are designed to control the propagation of EM waves by allowing or blocking them at specific frequency bands. The filtering behavior of FSS depends on the design and configuration of its unit cells, which determine whether EM waves are transmitted or reflected2. In recent years, FSS-based absorbers have attracted significant attention due to their superior characteristics compared to traditional absorbing structures. Their compact, thin, and lightweight nature, combined with enhanced mechanical flexibility3, makes them highly suitable for integration into advanced electromagnetic systems. As a result, FSS-based absorbers are widely employed in various applications, including EMI shielding4, radar cross-section (RCS) reduction5, and stealth technology6, where efficient control of EM wave propagation is critical. Several designs have been reported in the literature for constructing FSS-based absorbers7–10. A typical configuration comprises a sandwich structure, incorporating metallic resonators on one side and a complete metallic plate (ground plane) on the other side of a dielectric substrate11. The metallic ground plane blocks EM wave transmission through the structure entirely, and absorption occurs at the resonant frequency where the reflection rate is minimized. In this scheme, the complete metallic plate introduces a limitation by disallowing out-of-band transmission, which may be undesirable in some applications. To overcome this limitation, frequency selective rasorbers (FSRs) have been introduced, offering both absorption and transmission capabilities12,13. An FSR structure can be constructed by replacing the ground plane of a typical FSS-based absorber with a bandpass FSS. By doing so, additional transmission bands are achieved alongside the absorption bands. The FSRs can be implemented using either 2-D or 3-D structures14. While 3-D FSRs provide enhanced filtering by separating absorption and transmission paths, their design is more complex and costly to fabricate. Thus, this work focuses on the research of 2-D FSRs to address the deficiencies of 3-D designs15–17. Many previously reported 2-D FSR designs rely on multi-layered structures with integrated resistive components to achieve broad or multi-band absorption and transmission18–25. While these configurations offer functional performance in certain applications, they often suffer from increased thickness, fabrication complexity, and limited angular stability. To overcome these drawbacks, recent studies have explored single-layer FSRs. For example, a flexible single-layer FSR with dual transmission and triple absorption bands was proposed in26, and another design utilized frequency selective surfaces on opposite sides of a single dielectric substrate to achieve polarization-insensitive response27. Despite these advances, achieving multi-band response, high selectivity, and enhanced angular stability by a resistor-free single-layer structure remains a challenge. Broadband absorption is undeniably useful for general applications. However, there is an increasing demand to consider narrow absorption bands to meet the requirements of spatially selective filtering. For instance, when dealing with targeted RCS reduction or EMI issues, designers typically need to eliminate specific interference frequencies without shutting down the entire frequency spectrum. This is exactly why discrete-band absorbers are developed. As highlighted in28,29, utilizing multi-band absorbers is a proven and effective strategy for specific stealth and EMI applications, especially when covering the whole band is not an option. This paper presents a single-layer, polarization-insensitive FSR that achieves four distinct absorption bands and a transmission band located between them. The design is realized on a low-cost FR4 substrate using four concentric meandered copper rings on the top side as the resonant absorber and a meandered slot on the backside as the bandpass FSS. The unit cell size was reduced through meandering resonators to improve angular stability, resulting in more consistent performance across a wider range of incident angles for both TE and TM polarizations. Four-fold rotational symmetry ensures complete polarization insensitivity. Compared to recent multi-band metamaterial absorbers30–33 and multilayer FSRs15,19,20,34,35, the proposed resistor-free single-layer design offers superior multi-functionality, compactness, and fabrication simplicity. The simulated results demonstrate over 90% absorption rates at the resonance frequencies of 2.14, 2.88, 6.02, and 7.32 GHz. In addition, a passband with minimal insertion loss of 1.1 dB is achieved at 4.54 GHz. The proposed FSR maintains stable performance under incidence angles up to 50°. The absorption frequencies of this design fall within the S- and C-bands. These frequency bands are heavily crowded by air-traffic control, weather monitoring, naval radars, and sub-6 GHz 5G networks30,35,36. In practice, the proposed FSR works as a dual-purpose radome. It suppresses those specific threat frequencies but still maintains a 4.54 GHz passband for secure data transmission. Since it can block unwanted radar signals without cutting off essential data links, the structure is well-suited for stealth radomes, RCS reduction in unmanned aerial vehicles, and EMI reduction in systems co-located with meteorological radars. The rest of the paper is organized as follows: “Design of the structure” discuses about the design methodology of the proposed structure. Some simulations were carried out to evaluate the performance of the presented structure in “Analysis and performance verification”. A prototype of the proposed FSR was fabricated, measured, and a performance comparison with previous works is given in “Experimental verification and discussion”. Finally, concluding remarks are provided in “Conclusion”. Design of the structure As previously mentioned, this paper introduces a novel single-layer FSR featuring four absorption bands with a transmission band in between. The design methodology of the proposed FSR involves three steps. In the first step, a single-layer FSS-based absorber is designed with four narrow absorption bands. The unit cell of the proposed absorber consists of four concentric ring resonators printed on a metal-backed substrate. In the second step, a single bandpass FSS layer with a unit cell of ring slot resonator is designed to generate a passband at the one frequency positioned between the absorption frequencies of the first step absorber. The critical point that should be considered in this step is that the bandpass FSS layer must effectively function as a reflector at the absorption frequencies to serve as a suitable replacement for the metal ground of the absorber to ensure the preservation of absorption characteristics. In the final step, the ground plane of the absorber is substituted with the bandpass FSS to attain a structure that exhibits simultaneous absorption and transmission characteristics. A comprehensive description of each step will be provided in the following sections. Design of FSS-based absorber The initial step of the design involves the construction of a single-layer FSS-based absorber. The unit cell of the presented absorber is illustrated in Fig. 1. A 1.6 mm thick FR4 with relative permittivity of εr = 4.4 and loss tangent of tan δ = 0.02 is used as the substrate. The copper thickness and conductivity are t = 0.035 mm and σ = 5.8 × 107 s/m, respectively. The unit cell consists of four concentric copper ring loop resonators on top of the substrate, labeled Resonator I, II, III, and IV, indexing from the outermost to the innermost ring. A complete copper ground plane is also situated on the back side of the substrate. Unit cell configuration of the proposed FSS-based absorber, (a) front view, (b) side view. To miniaturize the structure and enhance the angular stability, a meandered pattern is chosen for three of the ring resonators. This meandering reduces the unit cell size without increasing the physical dimensions. A smaller electrical size delays the onset of grating lobes, which typically degrade performance at oblique incidence in periodic structures. As a result, the proposed FSR maintains stable absorption and transmission characteristics up to 50° for both TE and TM polarizations. Additionally, the four-fold rotational symmetry of the unit cell ensures complete polarization insensitivity for normal and oblique incidence. The optimized dimensions of the unit cell are gathered in Table 1. Table 1Dimensions of the proposed FSS-based absorber.ParameterValue (mm)ParameterValue (mm)ParameterValue (mm)L20w10.85l40.1W20w20.9s10.35r19.95w30.8s20.2r29w41.2s40.5r34.9l10.5h1.6r44l20.9 Dimensions of the proposed FSS-based absorber. The metallic backing of the absorber prevents any electromagnetic wave transmission through it (|S21(ω)| = 0). Thus, the absorptivity (A(ω)) is expressed as37: Figure 2(a) illustrates the absorptivity response for each of the ring resonators individually. These simulations and analysis are carried out in the CST Microwave Studio simulator using a periodic boundary condition and Floquet port excitation. As is clear, the two outermost resonators provide absorptions at 2.21 and 2.95 GHz, with peak absorptivity values of 99.53% and 99.01%, respectively. Furthermore, two bands emerged at 5.97 and 7.20 GHz with absorptivity values of 99.91% and 99.98%, respectively, attributed to the inner ring resonators. After simulating individual resonators, all the resonators are colocated to generate the final scheme of the absorber’s unit cell. Figure 2(b) illustrates the absorptivity of the proposed absorber. It can be seen that the absorber structure exhibits four absorption peaks of 99.96%, 93.56%, 99.89%, and 99.73% at 2.14, 2.88, 6.02, and 7.32 GHz, respectively. It is worth noting that combining the four resonators results in a marginal shift in absorption peak frequencies owing to the capacitive coupling between the resonators. Figure 2(c) illustrates the absorptivity, and scattering parameters of the presented absorber. It should be noted that due to the centrosymmetric design of the proposed structure, the optical responses for both transverse-electric (TE) and transverse-magnetic (TM) polarizations are fundamentally identical at normal incidence. Therefore, only one representative curve is plotted in this case for the sake of brevity and clarity. Simulated (a) absorptivity of individual resonators, (b) absorptivity of the presented absorber, (c) reflection, transmission and absorptivity of the presented absorber. For a better understanding of the working principle of the provided FSS-based absorber, an equivalent circuit model (ECM) is accompanied in the following. The ECM of the proposed absorber can be considered as illustrated in Fig. 338. The impedance of ZR represents the equivalent impedance of the upper layer. Z0 = 120π and are the characteristic impedance of the free space and the dielectric substrate, respectively. The impedance of the ground plane is considered near zero (short circuit) in the case of considering it as a perfect metal plane. ECM of the proposed FSS-based absorber. As indicated by Eq. (1), achieving complete absorption requires a reflection coefficient close to zero. Using the reflection coefficient, the input impedance of the absorber (Zin(ω)) can be calculated relative to the impedance of the free space as follows39: Based on the above equation, S11 will approach zero when the input impedance (Zin(ω)) equals the impedance of free space (Z0). Such a condition occurs at resonance frequencies. To verify the validity of this claim at the corresponding absorption points in Fig. 2(b), real and imaginary parts of the normalized input impedance of the proposed absorber are sketched in Fig. 4. The normalized impedance values at absorption frequencies of 2.14, 2.88, 6.02, and 7.32 GHz are 0.85 + j0.07 Ω, 1.10 + j0.44 Ω, 0.87 + j0.07 Ω, and 0.93 - j0.01 Ω, respectively. As is apparent, the real parts at all four absorptivity peaks tend to unity, while the imaginary parts are nearly zero. Normalized input impedance of the proposed FSS-based absorber. Design of FSR As stated earlier, the second step in realizing an FSR is designing a suitable bandpass FSS layer and then replacing it as the ground plane of the absorber. The unit cell of the proposed single bandpass FSS is sketched in Fig. 5. This unit cell comprises a meander loop slot etched onto a metal plane. Meandering the slot is done to attain a compact structure with increased angular stability. The unit cell of the proposed bandpass FSS. r5 = 6.9 mm, w5 = 0.6 mm, l5 = 0.3 mm, s5 = 0.3 mm, L = 20 mm, W = 20 mm. As illustrated in Fig. 6, the designed bandpass FSS simulation result demonstrates a transmission band at 4.54 GHz under normal incidence, which applies identically to both TE and TM polarizations. Based on the results, the bandpass FSS effectively acts as a reflector at absorption frequencies and is a suitable substitute for the absorber’s ground plane. Simulated S-parameters of the proposed bandpass FSS. Figure 7 shows the perspective view of an array of the presented FSR unit cells. The ground plane of the designed absorber is replaced with the bandpass FSS to attain the final configuration of the proposed FSR. Perspective view of an array of the proposed FSR. Figure 8 shows the CST simulation results of the proposed FSR. As seen, a transmission band at 4.54 GHz with a minimum insertion loss of 1.1 dB is achieved. Also, there are four absorption bands at 2.14, 2.88, 6.02, and 7.32 GHz with peak absorptivity values of 92%, 91%, 97% and 97%, respectively. The quality factors for the absorption bands can be calculated using the formula Q = f₀ / Δf, where Δf represents the half-power bandwidth. The values for the four absorption bands are 31.5, 39.1, 42, and 44.24, which highlight the excellent frequency selectivity and narrow-band performance of the proposed FSR. Simulated S-parameters and absorptivity of the proposed FSR. To provide a deeper insight into the operational mechanism of the proposed FSR, the ECM of the structure under a normal incident wave is given in Fig. 9. Each metallic ring resonator is modeled as a series RLC circuit. The parallel L5C5 is included due to the presence of narrow slits on the Resonator IV. The resistances of R1-4 and rd are associated with ohmic losses in resonators and dielectric loss of the substrate layer, respectively. Also, the slot resonator is modeled as a parallel LC circuit. To determine the precise values of these lumped elements, an optimization process was conducted using Keysight Advanced Design System (ADS) software to closely match the full-wave CST simulation responses. Figure 10(a,b) compares the simulation results derived from CST and ECM. As observed, while the resonant frequencies match excellently, there is a discrepancy in the operation bandwidth (Q-value). This deviation can be attributed to the inherent limitation of simple lumped-element ECMs40,41. CST accurately models higher-order Floquet modes and edge effects that influence the Q-factor, whereas ECM uses simplified frequency-independent components to maintain its straightforward topology. Overcomplicating the ECM would diminish its primary advantage as a simple, intuitive design tool. ECM of the proposed FSR. R1 = 1.8 Ω, R2 = 3.15 Ω, R3 = 11.06 Ω, R4 = 17.6 Ω, rd = 1.5 Ω, L1 = 28.3 nH, L2 = 14.7 nH, L3 = 17.6 nH, L4 = 13.76 nH, L5 = 6.32 nH, L6 = 0.44 nH, C1 = 170.81 fF, C2 = 170.6 fF, C3 = 24.36 fF, C4 = 62.13 fF, C5 = 270 fF, C6 = 3 pF. Simulation results of the proposed FSR by using CST and ECM. (a) S-parameters, (b) absorptivity. Analysis and performance verification The effect of structural geometries This section explores the behavior of the proposed FSR when its geometries are varied. Each ring resonator is responsible for the resonance occurring at a specific frequency. The resonators’ dimension determines the length of the effective current path and the location of the corresponding resonant frequencies. As shown in Fig. 11(a-d), the location of the resonance frequencies alters by varying the dimensions of the resonators. Furthermore, Fig. 12 explicitly illustrates how variations in the geometrical parameters influence the transmission frequency. Effect of the geometries variation on absorption frequencies (a) investigation of ‘r1’ and ‘w1’, (b) investigation of ‘s2′, (c) investigation of ‘r3’ and ‘w3’, (d) investigation of ‘l4′ (all dimensions are in mm). Effect of the geometries variation on transmission frequency (all dimensions are in mm). Surface current and EM field distribution The functionality of the proposed FSR is further assessed through the analysis of surface current (Fig. 13), magnetic field (Fig. 14), and electric field (Fig. 15) distributions at the absorption and transmission frequencies. As shown in Fig. 13(d,e), strong opposing currents are observed on the top inner rings and backside slot at the higher absorption frequencies (6.02 and 7.32 GHz). This creates a strong magnetic field trapped inside the substrate, leading to magnetic resonance and high absorption in these bands (see Fig. 14(d,e)). At the lower absorption frequencies (2.14 and 2.88 GHz), the electric field is mostly concentrated on the outer meandered rings (see Fig. 15(a,b)), while the magnetic field is quite weak (see Fig. 14(a,b)). This shows that electric resonance plays the main role in absorbing energy in these bands. The surface current distribution at (a) 2.14 GHz, (b) 2.88 GHz, (c) 4.54 GHz, (d) 6.02 GHz, (e) 7.32 GHz. The magnetic field distribution at (a) 2.14 GHz, (b) 2.88 GHz, (c) 4.54 GHz, (d) 6.02 GHz, (e) 7.32 GHz. The electric field distribution at (a) 2.14 GHz, (b) 2.88 GHz, (c) 4.54 GHz, (d) 6.02 GHz, (e) 7.32 GHz. At the transmission frequency of 4.54 GHz, the top rings exhibit weak current excitation and operate in an inductive regime with minimal resonance (see Fig. 13(c)). This results in low energy dissipation and near-transparent behavior for the incident wave. The wave reaches the backside slot, where strong currents along the meandered path indicate capacitive resonance, allowing efficient passage with low insertion loss (see Fig. 15(c)). The corresponding magnetic field distribution at this frequency (see Fig. 14(c)) shows minimal magnetic field strength, further supporting the weak resonance behavior and confirming the low energy dissipation at the transmission frequency. Based on the presented discussions, the simulations confirm the multi-functional performance of the proposed FSR. The FSR exhibits strong electric and magnetic resonances at the absorption frequencies, ensuring efficient energy absorption. However, it operates in a weak inductive/capacitive non-resonant state at the transmission frequency, which minimizes energy dissipation and results in near-transparent behavior for the incident wave. Angular stability Angular stability for both the polarization angle (Φ) and the incident angle (θ) is crucial for FSRs. The presented FSR is analyzed under various incident EM wave polarization angles, as illustrated in Fig. 16a,b. Due to the centrosymmetric geometry of the structure, the responses for TE and TM polarizations are identical at normal incidence. Furthermore, the simulation results are exclusively presented for the polarization angle ranging from 0° to 45°. The structure’s response remains nearly unchanged, indicating that the designed FSR is polarization insensitive. The oblique incidence response of the proposed FSR is analyzed for TE and TM polarized incident waves at various angles. Simulated (a) S-parameters and (b) absorptivity of the presented FSR under different polarization angles. As shown in Fig. 17b,d, absorptivity remains above 90% at all four absorption frequencies for both the TE and TM modes up to an incident angle of 30°. Additionally, the bandpass region remains nearly unchanged for both modes. There is a slight reduction in the absorption rate when the incident angle rises to 50 degrees. For TE polarization, peak absorptivity values reach 94.21%, 83.24%, 89.12%, and 81.10% at the four absorption bands, respectively. For TM polarization, the values are 86.68%, 88.09%, 97.29%, and 95.98%. Notably, the additional resonance peak between 4 and 6 GHz for oblique TM polarization (Fig. 16d) is caused by the normal electric field component (Ez) exciting a parasitic mode42. This perpendicular component couples with the structure and excites a parasitic resonance mode, which is a known behavior in 3D frequency-selective structures. Besides, as illustrated in Fig. 17(a) and (c), the insertion loss at the passband changes to 1.61 dB for TE polarization and 0.95 dB for TM polarization as the incident angle increases to 50 degrees, both of which are considered acceptable. Simulated S-parameters and absorptivity of the presented FSR under different incident angles for (a)-(b) TE polarization, (c)-(d) TM polarization. Effective medium parameter extraction Permittivity (ε) and permeability (µ) are key medium parameters that determine how electromagnetic waves interact with different materials33. The effective value of these parameters (εeff and µeff) are extracted from the simulated S-parameters using the Nicolson-Ross-Weir43 method to gain deeper insight into the metamaterial-like behavior of the proposed FSR. This method treats the periodic structure as a homogeneous slab which enables the retrieval of the complex εeff and µeff over the frequency range of interest. Figure 18 shows the real and imaginary parts of εeff and µeff versus frequency. Excellent impedance matching to free space is achieved at the absorption resonant frequencies, where both the real and imaginary parts of εeff and µeff are, respectively, relatively close to each other. The exact values of the real and imaginary parts of these effective parameters at the absorption frequencies are summarized in Table 2. This condition minimizes reflection and maximizes energy transfer into the structure, resulting in the observed high absorption (> 90%) at 2.14, 2.88, 6.02, and 7.32 GHz. Frequency responses of (a) real values of εeff and µeff and (b) imaginary values of εeff and µeff. Table 2Extracted real and imaginary parts of εeff and µeff at the absorption resonant frequencies.Absorption frequency (GHz)Real(εeff)Real(µeff)Imag(εeff)Imag(µeff)2.13 (≈ 2.14)16.6817.11-18.49-19.522.94 (≈ 2.88)16.5218.21-11.94-11.486.05 (≈ 6.02)4.175.649.1610.117.34 (≈ 7.32)1.062.839.499.41 Extracted real and imaginary parts of εeff and µeff at the absorption resonant frequencies. Experimental verification and discussion To experimentally validate the proposed FSR, a 300 mm × 300 mm prototype comprising 15 × 15 unit cells has been fabricated on a lossy 1.6 mm thick FR4 substrate, as illustrated in Fig. 19a,b. Photograph of the fabricated prototype of the proposed FSR. (a) Enlarged view of the front side. (b) Enlarged view of the back side. Figure 20a,b shows the measurement setup. The fabricated prototype is measured using the free space technique44. Scattering parameters are measured using a vector network analyzer (Agilent E8363C) and two horn antennas (ETS-LINDGREN 3117) covering an operational frequency range from 1 to 18 GHz. The prototype is fixed inside an absorbent foam frame to reduce edge diffraction and direct wave coupling between the antennas. To acquire the transmission coefficient, one antenna serves as a transmitter positioned in front of the measurement setup to excite the FSR. The secondary antenna, positioned at the rear, serves as a receiver to capture signals transmitted through the FSR. Both antennas are placed in front of the structure to measure the reflection parameter. Photography and block diagram of the (a) reflection measurement setup and, (b) transmission measurement setup. To ensure accuracy in the non-anechoic environment and to mitigate the inherent phase errors associated with non-ideal plane wave illumination, environmental reflections, multipath effects, and unwanted diffractions were filtered out using the Time-Domain Gating feature of the network analyzer45. Furthermore, a rigorous background subtraction and normalization procedure was applied using Eqs. (3) and (4)46. This technique systematically eliminates static background scattering, path loss, and direct mutual coupling between the antennas. where S11,MEA and S21,MEA represent the final calibrated reflection and transmission coefficients of the proposed FSR, respectively. The subscripts “DUT”, “AIR”, and “PEC” denote the raw scattering parameters directly acquired by the vector network analyzer in three different measurement scenarios: when the FSR sample (Device Under Test) is mounted, when the setup is empty (free space), and when a metal plate of identical dimensions is utilized, respectively. As a general rule in measuring S-parameters, the distance between antennas and sample should be sufficient to allow the electromagnetic fields to transition into plane waves. The distance required for this transition, known as the Fraunhofer distance46, depends on the antenna size and the frequency of operation. Strictly satisfying this classical far-field criterion for the entire 300 × 300 mm2 physical dimension of the sample at higher frequencies would require a significantly larger measurement distance. However, increasing the distance to fully meet this criterion is impractical as it leads to severe Signal to Noise Ratio (SNR) degradation and beam spillover, where the antenna beam illuminates beyond the sample boundaries, causing spurious edge reflections. Therefore, the distance of 1.2 m was selected as an optimal engineering trade-off. Additionally, the narrow beam-width of the horn antennas at higher frequencies significantly reduces the effective illuminated area on the prototype. Most importantly, the potential systematic phase errors caused by the spherical wavefront incidence at this distance are mathematically compensated for by the rigorous PEC normalization (Eqs. 3 and 4) and Time-Domain Gating mentioned earlier. Consequently, the actual active region of the structure interacting with the incident waves properly satisfies the measurement requirement at the 1.2 m distance. This specific distance provides an optimal balance between maintaining a quasi-monostatic47 incidence angle, minimizing edge diffraction, and achieving a high SNR. This inevitably results in some minor discrepancies at higher frequencies48. A comparison between simulation and measurement results is presented in Fig. 21a-d, revealing a good agreement between them. The slight discrepancies in the results can be attributed to various factors, including geometry tolerance, deformations due to fabrication limitations, variations in the dielectric constant of the substrate, measurement errors, etc. Simulated and measured results of the proposed FSR. (a) S-parameters and (b) absorptivity under the normal and oblique incidence for TE polarization. (c) S-parameters and (d) absorptivity under the normal and oblique incidence for TM polarization. Table 3 presents a comprehensive comparison between the proposed FSR and several recent designs reported in the literature. As the experimental setups and measurement procedures differ, the performance evaluation in this section is based on simulated results to ensure consistency and fairness. Table 3Comparisons between the presented FSR and recently reported designs (λLa free-space wavelength at the lowest operating frequency, DNGb data not given).Refs.AbsorptionTransmissionNo. of layersLumped elementsCell SizeThicknessAngular stabilityPolarization 31 Triple-band(2.4, 3.5, 6.09 GHz)-Nil-1No0.0256λ2La0.013λL60oInsensitive 32 Dual-band(2.57, 7.16 GHz)-Nil-1No0.005λ2L0.014λLTE: 45oTM: 30oInsensitive 33 Quad-band(3.46, 6.44, 7.89, 12.44 GHz)-Nil-1No0.019λ2L0.018λL60oInsensitive 34 Wideband(2.03–4.33 GHz, 21.81–26.31 GHz)Wideband(5.93–18.34 GHz)4Yes0.007λ2L0.119λL30oSingle 35 Wideband(1.5–2.75 GHz, 5.55–8 GHz)Wideband(2.75–5.55 GHz)3Yes0.16λ2L0.156λL30oInsensitive 19 Wideband(2.3–5.3 GHz, 7.8–14.6 GHz)Single-band(6.3 GHz)3Yes0.027λ2L0.111λLDNGbDual 20 Wideband(2.97–7.17 GHz, 9.35–12.8 GHz)Wideband(7.9–9 GHz)4Yes0.03λ2L0.1λL20oDual 21 Wideband(6.26–9.58 GHz, 11.46–14.06 GHz)Single-band(10.4 GHz)2Yes0.141λ2L0.125λL30oDual 22 Wideband(0.5–2.5 GHz)Wideband(4.6–13.8 GHz)2Yes0.001λ2L0.1λL60oInsensitive 25 Wideband(0.98–7.12 GHz)Wideband(7.52–12.56 GHz)3Yes0.001λ2L0.11λL45oInsensitive 26 Triple-band(6.6, 11.1, 14.7 GHz)Dual-band(9.6, 13.4 GHz)1No0.108λ2L0.005λL40oInsensitive 27 Single-band(6.2 GHz)Dual-band(3.1, 8.4 GHz)1No0.04λ2L0.008λL30oInsensitiveThisworkQuad-band(2.14, 2.88, 6.04, 7.34 GHz)Single-band(4.54 GHz)1No0.019λ2L0.01λL50oInsensitive Comparisons between the presented FSR and recently reported designs (λLa free-space wavelength at the lowest operating frequency, DNGb data not given). Triple-band (2.4, 3.5, 6.09 GHz) Dual-band (2.57, 7.16 GHz) TE: 45o TM: 30o Quad-band (3.46, 6.44, 7.89, 12.44 GHz) Wideband (2.03–4.33 GHz, 21.81–26.31 GHz) Wideband (5.93–18.34 GHz) Wideband (1.5–2.75 GHz, 5.55–8 GHz) Wideband (2.75–5.55 GHz) Wideband (2.3–5.3 GHz, 7.8–14.6 GHz) Single-band (6.3 GHz) Wideband (2.97–7.17 GHz, 9.35–12.8 GHz) Wideband (7.9–9 GHz) Wideband (6.26–9.58 GHz, 11.46–14.06 GHz) Single-band (10.4 GHz) Wideband (0.5–2.5 GHz) Wideband (4.6–13.8 GHz) Wideband (0.98–7.12 GHz) Wideband (7.52–12.56 GHz) Triple-band (6.6, 11.1, 14.7 GHz) Dual-band (9.6, 13.4 GHz) Single-band (6.2 GHz) Dual-band (3.1, 8.4 GHz) This work Quad-band (2.14, 2.88, 6.04, 7.34 GHz) Single-band (4.54 GHz) Many recent FSR designs employ multilayer configurations or resistive loading to achieve wideband or multi-band absorption. However, these designs often come at the cost of increased thickness, fabrication complexity, and limited angular stability19–21,34,35. For example, the multilayer FSRs in34] and [35 offer wideband absorption with transmission but require three or four layers and resistive loading. Likewise, the recently proposed wideband structures in22] and [25 achieve polarization-insensitive responses with good angular stability; however, they inherently suffer from structural complexity. They require two and three layers, respectively, and rely heavily on lumped elements, leading to relatively thick overall profiles. By contrast, single-layer approaches have been explored to simplify fabrication and eliminate resistors26,27,31,33. While these designs exhibit improved manufacturability, they face limitations in terms of the number of absorption and transmission bands, angular stability, and overall absorption performance. The single-layer flexible FSR in26 achieves triple absorption bands with dual transmission bands, demonstrating good conformal performance. Nonetheless, its angular stability is limited to approximately 40°, and it offers fewer absorption bands compared to the proposed design. The polarization-insensitive FSR reported in27 is implemented on a single-layer substrate with a single transmission band. Although this design offers simple fabrication, it is restricted to only two absorption bands and lower angular stability. In comparison, the proposed FSR exhibits significant improvements in multiple aspects. It realizes quad-band absorption at 2.14, 2.88, 6.04, and 7.34 GHz, along with a single, well-isolated transmission band centered at 4.54 GHz. The structure is fully passive and does not include any lumped components or multilayer stacking, and it is fabricated on a low-cost FR4 substrate. Moreover, it features a compact unit cell size (0.019λ2L) and a significantly lower profile (0.01λL) compared to the multi-layer and lumped element based designs, making compact integration possible. The design maintains polarization insensitivity and demonstrates stable performance under oblique incidence angles up to 50° without the need for complex fabrication processes. These advantages are achieved using a single-layer architecture composed of concentric meandered ring resonators and a slot-based transmission structure on the backside, yielding a high-performance yet low-complexity solution suitable for practical applications. Table 3 highlights the advantages of the proposed design in achieving quad-band absorption with a single transmission band in a single-layer, resistor-free structure, while maintaining enhanced angular stability and polarization insensitivity. Conclusion In this paper, a single-layer, polarization-insensitive FSR with quad-band absorption performance is proposed. The presented FSR is developed by integrating a quad-band resonant absorber and a single bandpass frequency-selective surface FSS on opposite sides of an FR4 substrate. In the absorption bands, the top layer effectively absorbs electromagnetic waves, while the backside FSS functions as a reflector. In the transmission band, the top layer becomes nearly transparent, allowing efficient wave passage through the backside FSS. The combination of the two layers results in effective multi-functional performance. The top layer achieves over 90% absorption at 2.14, 2.88, 6.02, and 7.32 GHz. The backside FSS generates a transmission band at 4.54 GHz with low insertion loss. The design exhibits enhanced angular stability up to 50° for both TE and TM polarizations, with polarization insensitivity ensured through four-fold rotational symmetry. The physical mechanism is analyzed in detail using an equivalent circuit model and effective-medium analysis. Various simulations have been conducted to validate the performance of the proposed design. A 15 × 15 prototype has been fabricated and measured, showing excellent agreement with simulations. This resistor-free single-layer FSR is suitable for radar cross-section reduction, EMI reduction, and multi-functional radome applications in S- and C-bands. Acknowledgements The authors would like to thank the North-west Antenna and Microwave Research Laboratory (NAMRL) at Urmia University for providing technical support and facilities. The authors also acknowledge the use of Grammarly, a generative AI-based writing assistant, for improving the language, grammar, and readability of the manuscript. Grammarly was used solely for linguistic editing and did not contribute to the scientific content, data analysis, or interpretation of results. Author contributions All authors (Asal Malekara, Changiz Ghobadi, Javad Nourinia, and Amin Khalilzadegan) contributed equally to the preparation of the paper. Data availability The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request by email (a.malekara@urmia.ac.ir). Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request by email (a.malekara@urmia.ac.ir).