Skip to content
Frontiers
Search

Original Research

Frontiers in Antennas and Propagation ·

Volume 4 - 2026 | https://doi.org/10.3389/fanpr.2026.1813732

Sensitivity analysis of the microstrip-fed circular InterDigital capacitor sensor with complete and partial backside ground conductor

  • Swati Todi1*corresponding author
  • Poonam Agarwal1
  1. MicroSystems Lab (µSL), School of Computer and Systems Sciences, Jawaharlal Nehru University (JNU), New Delhi, India
Article metrics0 viewsView details
PDFCiteMetrics

Abstract

The impact of the size of the backside ground conductor on the electromagnetic field distribution, radiation pattern, directivity, and gain of one-port devices is evident; however, its contribution to effective sensing applications needs further exploration. This paper presents an extensive experimental study to investigate the impact of the complete and partial backside ground conductors, beneath the sensing area, on the sensitivity of the proposed one-port microstrip-fed Circular InterDigital Capacitor Sensor (CIDCS). It consists of a circular InterDigital Capacitor (IDC) on one side and a ground conductor on the other side of the substrate. Two distinct prototypes of the CIDCS are presented: (i) Prototype-I, configured with a complete backside ground conductor, and (ii) Prototype-II, configured with a partial backside ground conductor, while keeping other design parameters constant. The CIDCS prototypes have been fabricated, and a circular cavity made of polydimethylsiloxane (PDMS) polymer is implemented on the CIDC surface for sample containment. RF measurements have been performed with glucose solutions prepared in deionized water at varying concentrations of glucose. To achieve maximum sensor performance, an extensive experimental study is presented with varying cavity diameters. Three prototypes with varying cavity diameters: (a) 4 mm, (b) 6 mm, and (c) 8 mm have been implemented. It is revealed that the difference in the sensitivity of CIDCS prototypes I and II with a cavity diameter of 6 mm is the highest. The simulation analysis of the CIDCS exhibited that the radiation efficiency reduced from 26 % in Prototype-I to 5 % in Prototype-II. However, the experimental findings confirm that the sensitivity of Prototype-II with a 6 m m cavity increases by 20.3 % and 49.41 % in terms of S 11 frequency and return loss, respectively, with improvements in the linear coefficient of 2.26 % and 0.324 %, respectively. From the experimental study, it is observed that the prototype with a partial backside ground conductor offers a significant improvement in sensing applications, where high responsiveness towards minor variations in the permittivity of the medium is more crucial than radiation efficiency.

Keywords

1 Introduction

Microwave sensors have gained significant attention due to their numerous advantages, including mediator-free, real-time, planar design with non-destructive sensing capabilities and high penetrability through biological materials. Due to these inherent characteristics, microwave sensors find applications in diverse fields such as dielectric characterization of liquid samples (Han et al., 2024; Palandoken and Gocen, 2025), environmental (Gugliandolo et al., 2022; Frau et al., 2021), and biomedical monitoring (Masi et al., 2023; Arabahmadi et al., 2025), and the industrial domain (Haq and Koziel, 2022).

The ground conductor in microwave devices plays a pivotal role in governing the electromagnetic field distribution, current distribution, and radiation characteristics (Balanis, 2016), thereby influencing key sensor metrics, including resonant frequency, bandwidth, and sensitivity. Extensive research is being devoted to the development of one-port microwave devices, as they inherently exhibit both antenna and sensor behavior, where a partial ground conductor is utilized to enhance performance. A quad-band antenna with a partial ground plane (Gangwar and Alam, 2019), a CRLH (composite right/left-handed) resonant antenna where the ground structure beneath the IDC structure is defected (Lee et al., 2010), a broadband omnidirectional monopole antenna incorporating a partial ground plane with a vertical slot and two L-shaped slots (Douhi et al., 2025), multiband operation using a microstrip patch antenna with partial ground plane (Samuel et al., 2025) are reported in the literature. Various antennas for ultra-wideband (UWB) applications have been reported, where the defected ground structure (Gopi et al., 2021; Alqahtani et al., 2022; Kanagasabai et al., 2021) and partial ground configurations (Chen et al., 2007; Tiwari et al., 2021; Tiwari et al., 2022; Lin et al., 2011; Parameswari and Chitra, 2021; Mobashsher and Abbosh, 2015) have been implemented. Flexible DGS-based microstrip antenna, where a circular-shaped slot is etched in the patch as well as the ground plane (Soni et al., 2024), and a split ring resonator unit cell antenna integrated with a meander line and DGS (Rahman et al., 2022) are also reported in the literature.

Many studies have reported the use of a partial ground conductor in one-port sensors to achieve better performance. A planar Yagi-Uda antenna-based sensor for glucose monitoring is reported, in which the partial ground plane is embedded with an inverted-L-shaped stub (Nella et al., 2023a; Nella et al., 2023b). Singh et al. (2023) reported a planar glucose sensor, in which two rectangular strips are arranged perpendicularly to each other with semi-circular-shaped intersections, and a square-shaped slot is etched out from the ground plane just beneath the patch. Gharbi et al. reported a circular ring-shaped (El Gharbi et al., 2021) and square ring-shaped (Gharbi et al., 2021) monopole antenna-based glucose sensor with a partial ground plane (El Gharbi et al., 2021; Gharbi et al., 2021). Harnsoongnoen and Buranrat (Harnsoongnoen and Buranrat, 2023b) reported a microstrip glucose sensor configured of four spurlines inserted into a patch antenna with a partial ground plane. Rahman et al. reported a microwave sensor configured with a circular patch and semi-circular slots (Rahman et al., 2019) and a circulated Psi-shaped patch (Rahman et al., 2018) with a partial ground plane having a slot (Rahman et al., 2019; Rahman et al., 2018). A rectangular patch antenna, with a defected ground structure (DGS) beneath the patch, for size reduction and enhanced performance, has been reported (Njokweni and Kumar, 2020). Zhang et al. (2022) reported a triple-ring microstrip patch antenna with DGS, consisting of a semicircular-shape structure, to increase capacitance and inductance and thereby sensitivity. Various two-port sensors exploring the role of a partial ground conductor have also been reported in the literature. Abdelwahab et al. (2021) reported a microfluidic sensor, where an LC resonator is placed in series with the microstrip transmission line. The ground plane below the resonator has been removed to reduce parasitic capacitance and improve sensitivity (Abdelwahab et al., 2021). Javadizadeh et al. (2023) reported a sensor comprising two InterDigital Capacitors (IDCs) placed parallel to a spiral inductor, resulting in an ELC resonator, where the circular area of the ground plane underneath the ELC resonator is removed to decrease the parasitic capacitance between the ground and the sensing area. A microstrip transmission line with single and double circular-shaped slots etched out from the ground plane is used to develop a band-stop filter (BSF) and demonstrate the BSF as a glucose sensor (Mansour et al., 2023). Our recent work reported a comparative experimental study, in which the impact of conventional and altered ground conductors on the sensitivity of a two-port LC Coplanar Sensor has been explored (Todi and Agarwal, 2025).

It is apparent that the ground conductor influences the electromagnetic field distribution, radiation characteristics, and sensor performance; however, the effect of the complete and partial backside ground conductor on sensor performance has not been adequately quantified, to the authors’ knowledge. To fill this research gap, this work presents a comprehensive comparative experimental study to quantify the impact of the presence versus absence of the ground conductor beneath the sensing area (within the same design) on the sensor performance. Hence, two prototypes of the proposed one-port sensor, one with a complete backside ground and another with a partial backside ground conductor, have been demonstrated. To provide insights into the visual differences between complete and partial backside ground, a schematic illustration of the one-port sensor with the top, 3D, and cross-sectional views of the complete backside ground conductor is presented in Figure 1a(i-iii) and partial backside ground in Figure 1b(iv-vi), respectively. In this work, the proposed one-port sensor is configured with a microstrip-fed circular interdigital capacitor-structured patch on one side and a ground conductor on the backside of the FR4 substrate. InterDigital Capacitor (IDC)-based devices are widely reported in the microwave sensing applications (Zhao et al., 2024; Sari et al., 2025; Harnsoongnoen and Buranrat, 2023a; Piekarz et al., 2024; Hosseinzadeh and Pakvarjouy, 2025; Khan, 2024; Yunos et al., 2021; Mazumder et al., 2025) due to their ability to confine strong electric fields within a compact structure. A rigorous simulation study has been presented that analyzes the electromagnetic field distribution, resonant frequency, and radiation characteristics of the proposed sensor prototypes. The sensor prototypes have been fabricated in-house, with a circular cavity implemented using the PDMS (polydimethylsiloxane) to hold the aqueous sample. To achieve the maximum performance of the sensor, cavities with three different diameters: (a) 4 mm, (b) 6 mm, and (c) 8 mm have been implemented, keeping the cavity height constant. RF measurements have been conducted on prototypes using aqueous glucose samples at varying concentrations as the sample under test (SUT). Experimental findings reveal that prototype-II with 6 m m cavity diameter demonstrated marked improvement in sensitivity. The proposed Circular InterDigital Capacitor Sensor (CIDCS) design, along with a detailed electromagnetic simulation study of complete and partial backside ground-structure prototypes, is presented under Section 2. The sensor fabrication with the proposed experimental methodology is discussed under Section 3. The experimental assessment, including the measured outcomes and the results analysis, is presented under Sections 4, 5, respectively, with conclusions under Section 6.

Schematic representation: **(a)** Complete Backside Ground (i) Top view, (ii) 3D perspective, (iii) Cross-sectional view, and **(b)** Partial Backside Ground (iv) Top view, (v) 3D perspective, (vi) Cross-sectional view.
Figure 1. Schematic representation: (a) Complete Backside Ground (i) Top view, (ii) 3D perspective, (iii) Cross-sectional view, and (b) Partial Backside Ground (iv) Top view, (v) 3D perspective, (vi) Cross-sectional view.

2 Proposed sensor design

The proposed Circular InterDigital Capacitor Sensor (CIDCS) comprises a circular-shaped InterDigital Capacitor (IDC) structure on one side and a ground conductor on the other side of the 1.5 m m thick FR4 substrate (dielectric constant = 4.3). The 3D schematic representation of the CIDCS prototype is illustrated in Figure 2A, where the width of the circular fingers w f and gap between the fingers g f are 0.5 m m and 0.3 m m, respectively. The CIDC is electromagnetically fed through a 50 Ω microstrip transmission line, having a length l strip of 13.5 m m and a width w strip of 2.9 m m, respectively. Here, the impact of complete and partial backside ground conductors on sensor performance has been explored. Therefore, based on the ground conductor length, two prototypes of the proposed CIDCS have been investigated:

  • CIDCS Prototype-I is configured with a complete metallic backside ground conductor beneath the IDC structure, as shown in Figure 2B, where the length of the ground conductor l C G is 40 m m.
  • CIDCS Prototype-II is configured with a partial backside ground conductor, exposing the bottom surface of the substrate, as shown in Figure 2C, where the ground conductor length l P G is 14.1 m m.
Schematic representation of the CIDCS prototype: **(A)** Top view, **(B)** Prototype-I with complete backside ground conductor, **(C)** Prototype-II with partial backside ground conductor.
Figure 2. Schematic representation of the CIDCS prototype: (A) Top view, (B) Prototype-I with complete backside ground conductor, (C) Prototype-II with partial backside ground conductor.

The proposed CIDCS Prototype-I (with complete backside ground) is designed to operate at a frequency of 6.25 G H z. The analytical modeling of the proposed sensor prototype is carried out in MATLAB, using the established formulations reported in the literature (Pozar, 2011; Adhikari et al., 2019; Beeresha et al., 2016; Kumar et al., 2020), which have been appropriately adapted to suit the requirements of the present study. A comprehensive description of the modeling approach and its implementation is discussed in (Todi et al., 2026). The 3D modeling and subsequent optimization are executed in the electromagnetic simulation software CST. To develop the CIDCS Prototype-II, a partial backside ground conductor of length 14.1 m m is created, where the length of the backside ground conductor is optimized through a simulation-based parametric study. In both configurations, the dimensions of the CIDC are kept constant, with variations made solely to the backside ground conductor length.

2.1 Electromagnetic simulation analysis

Extensive parametric simulations have been performed on the CIDCS prototypes I and II to optimize and quantify the electromagnetic characteristics, as listed in Table 1. The CIDCS Prototype-I demonstrates a simulated resonant frequency 6.345 G H z with return loss 45.331 d B, while the CIDCS Prototype-II exhibits two resonant frequencies 2.149 G H z and 6.984 G H z with return losses 32.141 d B and 29.163 d B, respectively, as shown in Figure 3a. In Prototype-II, simulation analysis at resonant frequency 6.984 G H z is not conducted because the FR4 substrate is constrained with an upper frequency limit. The electric field distribution for the CIDCS prototypes I and II at the corresponding resonant frequencies is illustrated in Figures 3b,c, where the maximum field intensities are 63281 V / m and 1.66 × 1 0 5 V / m, respectively. The partial removal of the backside ground conductor resulted in a significant increase in electric and magnetic field strength, sensor capacitance, surface current, power flow, and electric and magnetic energy density. The partial backside ground removal caused more of the fringing electric fields to remain concentrated between the IDC fingers, creating a stronger and more localized field (Yeo and Lee, 2019). The electric field distribution around the IDC fingers in Prototype-II is significantly higher and more concentrated than in Prototype-I, as depicted in Figures 3b,c. This tighter confinement of the field lines enhances the coupling between the fingers, resulting in higher capacitance (Lee et al., 2010) and, consequently, a lower frequency in Prototype-II. Additionally, eliminating the ground conductor beneath the IDC reduced the parasitic capacitance introduced in the IDC structure, further improving the performance (Ebrahimi et al., 2019). The proposed CIDCS prototypes are one-port designs, which exhibit antenna characteristics as well; therefore, their radiation characteristics have been evaluated. The radiation pattern for the proposed CIDCS Prototype-I is presented in Figure 3d, which shows that the main lobe is broad, with wide coverage in azimuth, pointing perpendicular to the CIDCS Prototype-I surface (in z-direction; unidirectional) with a maximum directivity 7.06 d B i, gain 1.34 d B i, and the radiation efficiency 26 %. Here, the radiation efficiency of 26 % is excellent for the 1.5 m m thick FR4 substrate (Nascimento et al., 2011). The main lobe direction is 7 °, with 3 d B angular width of 83.2 °, indicating wide coverage, and side lobes are suppressed with a value of − 10.2 d B, as depicted in Figure 3f. The back lobe is present, indicating weak radiation behind the sensor due to the ground conductor. The radiation pattern for the CIDCS Prototype-II is illustrated in Figure 3e, which reveals that the stronger lobes are distributed towards the ±z-direction, which is due to the absence of ground, omnidirectional in the xz plane, with a maximum directivity 2.83 d B i, gain − 10.2 d B i, and the radiation efficiency 5 %. The main lobe direction is 177 °, with a 3 dB beamwidth of 93.5 °, affirming wide coverage, with a side lobe level of − 2.1 d B, as shown in Figure 3g. The front-to-back ratio of the proposed CIDCS prototypes I and II is 15.91 and 2.08, respectively, revealing a substantial change in the radiation characteristics upon the removal of the partial backside ground conductor.

TABLE 1. Simulated electromagnetic characteristics of CIDCS prototypes I and II.

CIDCS with
Simulated electromagnetic characteristicsComplete backside groundPartial backside ground
Resonant Frequency ( G H z )6.3452.149
Electric Field Distribution ( V / m )632811.66 × 1 0 5
Magentic Field Distribution ( A / m )366734
Q-factor4650
Directivity ( d B i )7.062.83
Gain ( d B i )1.34−10.2
Inductance ( n H )115.96125.41
Capacitance ( p F )0.005430.0437
Surface Current ( A / m )290595
Current Density ( A / m 2 )1,4411,194
Power Flow ( V A / m 2 )1.45 × 1 0 63.39 × 1 0 6
Electric Energy Density ( J / m 3 )0.02810.153
Magnetic Energy Density ( J / m 3 )0.04510.176
Front-to-Back (F/B) Ratio ( d B )15.912.08
Power Loss Density ( W / m 3 )4.02 × 1 0 77.3 × 1 0 7
CST simulated results: **(a)** S 11 parameters of CIDCS prototypes I and II, Electric Field Distribution: **(b)** Prototype-I, **(c)** Prototype-II, 3D Polar Plot: **(d)** Prototype-I, **(e)** Prototype-II, Radiation Pattern: **(f)** Prototype-I, **(g)** Prototype-II.
Figure 3. CST simulated results: (a) S 11 parameters of CIDCS prototypes I and II, Electric Field Distribution: (b) Prototype-I, (c) Prototype-II, 3D Polar Plot: (d) Prototype-I, (e) Prototype-II, Radiation Pattern: (f) Prototype-I, (g) Prototype-II.

The simulation results exhibited that the presence versus absence of a ground conductor beneath the IDC structure has a significant impact on the electromagnetic performance of the sensor. The ground conductor beneath the substrate acts as a reflector, limiting electromagnetic coupling and field distribution between the IDC fingers. Conversely, the absence of a ground conductor beneath the IDC reduced parasitic capacitance and enhanced electromagnetic coupling and capacitance between the IDC fingers, significantly altering the radiation characteristics. While the presence of a complete backside ground conductor can enhance radiation properties such as directivity and gain, the focus of the paper is to evaluate whether a complete backside ground conductor is required for effective sensing applications and improved sensor performance.

3 Sensor fabrication and experimental setup

The proposed CIDCS prototypes are fabricated on the FR4 substrate with a two-sided copper clad thickness 35 μ m, using the conventional lithography technique; the top and bottom views of the fabricated CIDCS prototypes I and II are depicted in Figures 4a,b, respectively. The circular IDC has been defined as the sensing area using the biocompatible polydimethylsiloxane (PDMS) polymer. The cured PDMS polymer, with thickness h pdms = 3 m m, is carefully cut to the required square shape of side l pdms = 12 m m, and a through-hole of diameter d pdms = 4 m m is created at its center. The square-shaped PDMS block, with a through-hole at its center, is aligned with the circular IDC, as illustrated in Figure 4c (labeled Ⓐ), and bonded firmly in position using the PDMS polymer. The adhesion of the PDMS block, with a through-hole, resulted in a circular cavity to hold the sample under test, as shown in Figure 4c (labeled Ⓑ).

**(a)** Top view of the fabricated CIDCS prototypes, **(b)** Bottom view of the fabricated CIDCS prototypes, **(c)** Schematic representation of the realization of the CIDCS prototype with circular cavity made of PDMS polymer, **(d)** Top view of the CIDCS prototypes realized with the circular cavity.
Figure 4. (a) Top view of the fabricated CIDCS prototypes, (b) Bottom view of the fabricated CIDCS prototypes, (c) Schematic representation of the realization of the CIDCS prototype with circular cavity made of PDMS polymer, (d) Top view of the CIDCS prototypes realized with the circular cavity.

Here, the sensing area has been configured based on the electric field concentration and fringing field interaction in the interdigital capacitor. Most of the sensing occurs in the fringing fields of the adjacent fingers; therefore, to achieve maximum sensor performance, experiments with varying cavity diameters have been conducted, and the diameter that yields high sensitivity is demonstrated for further analysis. To analyze the effect of all the adjacent fingers in the IDC, a systematic experimental study has been carried out by implementing three distinct cavities made of PDMS block with diameters of 4 m m (Figure 4d), 6 m m (Figure 4d), and 8 m m, keeping the thickness of the block constant. The smallest cavity considered has a diameter of 4 m m, to cover at least two circular IDC fingers, such that the electromagnetic coupling between the fingers can be utilized for the sensing mechanism. The cavity diameter has been systematically increased from 4 m m to 12 m m in steps of 2 m m, where the diameter of the outermost circular IDC finger is 11.5 m m approximately. The diameters with odd values have been excluded because the width of the circular IDC, as well as the radius increment, is 0.5 m m. Therefore, the inclusion of odd values is deemed to have a negligible impact. Also, the experimental results for cavity diameters larger than 8 m m were found to be insignificant, therefore, are excluded from further analysis.

The experimental evaluation of the impact of complete and partial backside ground on the proposed sensor performance has been demonstrated using the R & S network analyzer to measure the RF parameters across the frequency range within 1 − 10 G H z; the experimental setup is shown in Figure 5. The frequency sweep from 1 − 10 G H z is considered, because the CIDCS prototype-I has been designed to operate around 6.25 G H z frequency. When the partial backside ground conductor is removed in prototype-II, the frequency is reduced to approximately 2.15 G H z. Therefore, a wideband sweep ensures that the resonant behavior of both prototypes is captured within a single, consistent measurement framework, enabling direct comparison, capturing the complete spectral behavior, and providing a comprehensive characterization of the proposed sensor structures. Additionally, this approach allows verification that no unintended, spurious higher-order resonances interfere with the sensing region. The performance of the CIDCS prototypes has been demonstrated using glucose samples prepared in deionized water, where the glucose concentration varies within the range 0.0 − 0.5 g / m l, with a step increment of 0.1 g / m l. The sample volumes of 35 μ L, 75 μ L, and 150 μ L have been measured carefully using a micropipette and poured into the circular cavities of diameter 4 m m, 6 m m, and 8 m m, respectively. The PDMS cavity has been filled with red-colored solution, as shown in the insets of Figure 5, to facilitate visual identification of the presence of SUT in the cavity. The cavity has been rinsed with deionized water and dried carefully to prepare the sensor for reuse. The RF measurements have been performed under a controlled environment with consistent temperature and humidity conditions, and the samples under test had been kept at ambient temperature.

Experimental setup to demonstrate sensor performance.
Figure 5. Experimental setup to demonstrate sensor performance.

4 Results and analysis

To functionally validate the sensor, the experimentally measured S-parameters of the CIDCS prototypes I and II, without a PDMS cavity on the top surface, are compared with the corresponding simulated results, as shown in Figure 6a. Minor deviations may be attributed to fabrication tolerances, inherent parasitic elements, SMA connectors, etc. CIDCS Prototype-I, before assembling the PDMS cavity, exhibits simulated and experimental resonant frequencies of 6.345 G H z and 6.342 G H z, with return losses of 45.3313 d B and 26.7351 d B, respectively. Likewise, Prototype-II, in the absence of the cavity on top, exhibits simulated resonant frequencies of 2.149 G H z and 6.984 G H z, with respective return losses of 32.1409 d B and 29.1629 d B, and experimental frequencies of 2.341 G H z and 6.777 G H z, with respective return losses of 16.3135 d B and 44.2433 d B.

**(a)** Simulated vs. experimental results for CIDCS prototypes I and II, before implementation of PDMS cavity, S 11 parameters of the CIDCS prototypes I and II, before and after implementation of the PDMS cavity of varying diameter: **(b)** 4 m m, **(c)** 6 m m, and **(d)** 8 m m.
Figure 6. (a) Simulated vs. experimental results for CIDCS prototypes I and II, before implementation of PDMS cavity, S 11 parameters of the CIDCS prototypes I and II, before and after implementation of the PDMS cavity of varying diameter: (b) 4 m m, (c) 6 m m, and (d) 8 m m.

The impact of adhering the PDMS cavities of varying diameters of 4 m m, 6 m m, and 8 m m to the circular IDC structure has also been investigated, as shown in Figures 6b–d, respectively. On realization of the CIDCS prototypes with the circular cavity of diameter 4 m m, the Prototype-I revealed the experimental resonant frequencies of 3.0227 G H z and 5.8766 G H z, with return losses of 14.0896 d B and 21.7868 d B, respectively, while the Prototype-II exhibits measured resonant frequency of 2.0624 G H z, with return loss of 26.1411 d B, respectively, as depicted in Figure 6b. When CIDCS prototypes have been implemented with a PDMS cavity of diameter 6 m m, Prototype-I demonstrates the experimental frequencies of 3.056 G H z and 6.0327 G H z, with return losses of 17.7154 d B and 16.7267 d B, respectively, and Prototype-II exhibits measured resonant frequency of 2.116 G H z, with return loss of 21.0499 d B, respectively, as illustrated in Figure 6c. On the other hand, the Prototype-I demonstrates the measured peak frequencies of 3.125 G H z and 6.2037 G H z, with return losses of 27.8627 d B and 20.36 d B, respectively, while Prototype-II exhibits measured resonant frequency of 2.1644 G H z, with return loss of 44.397 d B, respectively, as shown in Figure 6d, when implemented with 8 m m cavity.

To evaluate the sensor performance, RF measurements have been carried out using aqueous glucose solutions with various glucose concentrations as the sample under test (SUT). The response of the sensor has been characterized in terms of S 11 parameters, emphasizing the return loss and resonant frequency shift. The experimental results demonstrating the capability of CIDCS prototypes I and II to distinguish between different glucose levels are presented in Figure 7. For glucose concentrations between 0.0 and 0.5 g / m l in aqueous solutions, a relative shift has been observed in prototypes I and II, as illustrated in Figures 7a–f, respectively. The insets show an enlarged view for clear visualization, where the frequency span is chosen based on the most distinguishable shifts relative to varying glucose concentrations. As the glucose levels increase from 0.0 − 0.5 g / m l, the S 11 peak frequency shifts from 1.5582 to 1.6062 G H z, 1.4352 to 1.4652 G H z, and 1.3391 to 1.4231 G H z, with shift in return losses from 13.5238 to 13.3709 d B, 7.1928 to 17.6586 d B, and 16.3503 to 13.4861 d B for Prototype-I with circular cavity of varying diameter of 4 m m, 6 m m, and 8 m m, as shown in Figures 7a,c,e, respectively. With the increasing glucose concentration, the S 11 peak frequency shifts from 1.4772 to 1.5282 G H z, 1.3541 to 1.3901 G H z, and 1.4744 to 1.5584 G H z, with shift in return losses from 18.2934 to 7.8973 d B, 25.3808 to 10.3763 d B, and 13.5746 to 7.7366 d B for Prototype-II with circular cavity of varying diameter of 4 m m, 6 m m, and 8 m m, as shown in Figures 7b,d,f, respectively. The shifts in S 11 peak frequency and the corresponding return loss, w.r.t. varying concentrations of glucose, for prototypes I and II, each with circular cavity of diameter 4 m m, 6 m m, and 8 m m, are listed in Table 2. The experimental data have been fitted using linear regression analysis, where the fitted peak frequency and return loss, along with the straight line equations, for CIDCS prototypes I and II are presented in Figures 8a–f, respectively. The experimentally measured sensitivities, in terms of S 11 peak frequency and return loss, for CIDCS prototypes I and II with the PDMS circular cavities of varying diameters: 4 m m, 6 m m, and 8 m m, are presented in Table 3. The measured sensitivity for the CIDCS Prototype-I, in terms of S 11 peak frequency, is 95.168 M H z / g / m l, 59.157 M H z / g / m l, and 175.714 M H z / g / m l, with regression coefficient ( R 2 ) values 0.989, 0.973, and 0.939, and 2.480 d B / g / m l, 19.993 d B / g / m l, and 5.048 d B / g / m l, in terms of return loss, with R 2 values 0.246, 0.927, and 0.823, respectively. With the removal of the partial backside ground conductor in CIDCS Prototype-II, the experimental sensitivity in terms of S 11 peak frequency is 102.568 M H z / g / m l, 71.169 M H z / g / m l, and 173.714 M H z / g / m l, with R 2 values 0.902, 0.995, and 0.984, and 20.644 d B / g / m l, 29.871 d B / g / m l, and 11.547 d B / g / m l, in terms of return loss, with R 2 values 0.944, 0.930, and 0.939, respectively.

Experimental results of CIDCS with PDMS cavity of varying diameter: **(a)** Prototype-I with 4 m m cavity, **(b)** Prototype-II with 4 m m cavity, **(c)** Prototype-I with 6 m m cavity, **(d)** Prototype-II with 6 m m cavity, **(e)** Prototype-I with 8 m m cavity, **(f)** Prototype-II with 8 m m cavity.
Figure 7. Experimental results of CIDCS with PDMS cavity of varying diameter: (a) Prototype-I with 4 m m cavity, (b) Prototype-II with 4 m m cavity, (c) Prototype-I with 6 m m cavity, (d) Prototype-II with 6 m m cavity, (e) Prototype-I with 8 m m cavity, (f) Prototype-II with 8 m m cavity.

TABLE 2. Experimental S 11 parameters w.r.t. glucose concentration for complete and partial backside ground.

Glucose Concen. (g/mL)Complete backside groundPartial backside ground
4 m m cavity6 m m cavity8 m m cavity4 m m cavity6 m m cavity8 m m cavity
R.C.*Freq.R.C.*Freq.R.C.*Freq.R.C.*Freq.R.C.*Freq.R.C.*Freq.
(dB)(GHz)(dB)(GHz)(dB)(GHz)(dB)(GHz)(dB)(GHz)(dB)(GHz)
0.0− 13.52381.5582− 7.19281.4352− 16.35031.3391− 18.29341.4772− 25.38081.3541− 13.57461.4744
0.1− 18.16931.5702− 8.03691.4382− 14.82891.3541− 15.42741.4772− 21.69991.3601− 11.61931.4884
0.2− 34.27321.5792− 9.84551.4472− 14.41681.3751− 12.13851.4892− 16.09931.3691− 10.11211.5044
0.3− 23.86481.5912− 11.68561.4502− 19.94021.4081− 10.35791.4940− 13.9071.3751− 9.07651.5184
0.4− 18.49031.5972− 13.30571.4562− 13.82061.4081− 9.26291.5102− 12.58891.3811− 8.22331.5464
0.5− 13.37091.6062− 17.65861.4652− 13.48611.4231− 7.89731.5282− 10.37631.3901− 7.73661.5584

R.C. - Reflection Coefficient.

Regression analysis of the S 11 parameters for the CIDCS with PDMS cavity of varying diameter: **(a)** Prototype-I with 4 m m cavity, **(b)** Prototype-II with 4 m m cavity, **(c)** Prototype-I with 6 m m cavity, **(d)** Prototype-II with 6 m m cavity, **(e)** Prototype-I with 8 m m cavity, **(f)** Prototype-II with 8 m m cavity.
Figure 8. Regression analysis of the S 11 parameters for the CIDCS with PDMS cavity of varying diameter: (a) Prototype-I with 4 m m cavity, (b) Prototype-II with 4 m m cavity, (c) Prototype-I with 6 m m cavity, (d) Prototype-II with 6 m m cavity, (e) Prototype-I with 8 m m cavity, (f) Prototype-II with 8 m m cavity.

TABLE 3. Experimentally measured sensitivity of the CIDCS prototypes with circular PDMS cavity of diameters: 4 m m, 6 m m, and 8 m m

Sensor prototypeCavity diameterSUT volumeReturn Loss (dB/g/mL)Frequency (MHz/g/mL)
Prototype-I4 m m35μL2.480 ( R 2 = 0.246 )95.168 ( R 2 = 0.989 )
Prototype-II4 m m35μL20.644 ( R 2 = 0.944 )102.568 ( R 2 = 0.902 )
Prototype-I6 m m75μL19.993 ( R 2 = 0.927 )59.157 ( R 2 = 0.973 )
Prototype-II6 m m75μL29.871 ( R 2 = 0.930 )71.169 ( R 2 = 0.995 )
Prototype-I8 m m150μL5.048 ( R 2 = 0.823 )175.714 ( R 2 = 0.939 )
Prototype-II8 m m150μL11.547 ( R 2 = 0.939 )173.714 ( R 2 = 0.984 )

It is evident from Table 3 that the optimum results have been obtained for the CIDCS prototypes with a circular cavity of diameter 6 m m; therefore, further result analysis has been carried out for the CIDCS prototypes I and II with a 6 m m cavity. The RF measurements have been performed three times to demonstrate the repeatability of the proposed CIDCS prototypes I and II with a cavity of diameter 6 m m. The mean return loss and frequency with standard deviations are illustrated in Figures 9a,b, respectively. The standard deviation (SD) is determined to be 0.1896 d B and 2.1767 M H z and 0.9006 d B and 3.5713 M H z for Prototypes I and II, respectively, exhibiting the minimal dispersion of measurements. The mean absolute deviation (MAD) is 0.1349 d B and 1.5733 M H z and 0.6492 d B and 2.4433 M H z for prototypes I and II, respectively, further supporting the consistency of the sensor response. The coefficient of variation (CV) is calculated in terms of return loss and frequency, which is found to be 1.68% and 0.16% for prototype-I and 5.64% and 0.26% for prototype-II, respectively, demonstrating high precision with negligible relative variability. The repeatability coefficient is determined to be 0.5252 d B and 6.0294 M H z for prototype-I and 2.4945 d B and 9.8924 M H z for prototype-II, respectively, indicating good repeatability. The uncertainty associated with the measurements is calculated to be ± 0.0774 d B and ± 0.8886 M H z for prototype-I and ± 0.3677 d B and ± 1.4579 M H z for prototype-II, respectively. Moreover, the limit of detection (LOD) has been estimated to be 0.0239 g / m l and 8.93 E − 5 g / m l in terms of magnitude and frequency for Prototype-I and 0.1609 g / m l and 7.28 E − 5 g / m l in terms of magnitude and frequency for Prototype-II, respectively, confirming the sensitivity of the sensor towards low-level detection. These statistical parameters demonstrate that the proposed sensor prototypes exhibit excellent repeatability and reliability for quantitative measurements. The mean data have been utilized to determine the average sensitivity of prototypes I and II, as presented in Figures 9c,d, respectively. The average sensitivity of the Prototype-I, in terms of S 11 return loss and frequency, is 20.753 d B / g / m l and 64.571 M H z / g / m l with R 2 values 0.938 and 0.990, respectively. The average sensitivity of the Prototype-II has been enhanced to 30.179 d B / g / m l and 79.74 M H z / g / m l with R 2 values of 0.939 and 0.997, respectively.

Mean experimental results for CIDCS prototypes I and II with circular cavity of diameter 6 m m: **(a)** Mean return loss, **(b)** Mean frequency, Average sensitivity in terms of: **(c)** Return loss, **(d)** Peak frequency.
Figure 9. Mean experimental results for CIDCS prototypes I and II with circular cavity of diameter 6 m m: (a) Mean return loss, (b) Mean frequency, Average sensitivity in terms of: (c) Return loss, (d) Peak frequency.

5 Discussion

The CST-based electromagnetic simulation results, as presented in the Section 2.1, revealed that eliminating the partial backside ground conductor in CIDCS Prototype-II resulted in enhanced electric and magnetic field distribution by 160 % and 100 %, respectively, surface current by 105 %, and electric and magnetic energy density by 444 % and 290 %, respectively; however, it also led to degradation of radiation characteristics including directivity by − 59.91 %, gain by − 861 %, current density by − 17.14 %, front-to-back ratio by − 86.92 % and radiation efficiency reduced from 26 % in Prototype-I to 5 % in Prototype-II. It is evident from the experimental results for cavity optimization that the best results have been obtained for the cavity diameter 6 m m. The experimental sensitivity for Prototype-II is found to be increased by 20.3 % and 49.41 % in terms of S 11 frequency and return loss, with improvement in linear coefficient ( R 2 ) by 2.26 % and 0.324 %, respectively, as compared to Prototype-I. The mean sensitivity for Prototype-II, determined from the repeated experiments for cavity diameter 6 m m, is increased by 23.49 % and 45.42 % in terms of S 11 frequency and return loss, with improvement in R 2 value by 0.707 % and 0.107 %, respectively, as compared to Prototype-I.

The experimental outcomes reveal that the partial removal of the ground conductor in the CIDCS Prototype-II caused a substantial increase in the electromagnetic field intensity at the sensing area, resulting in significantly improved sensitivity. With the complete backside ground structure in CIDCS Prototype-I, the majority of fields are restricted between the CIDC and ground, highly localized in the dielectric substrate, leading to a smaller fringing field (Balanis, 2016; Saadat-Safa et al., 2019), which limits the interaction of the electric field intensity in the sensing area with the samples under test, thereby limiting the sensitivity of the sensor to small changes in the dielectric properties of the SUT. With the removal of partial backside ground, the constrained field becomes weaker as there is no ground conductor beneath the CIDC, causing the stronger fringing field. The increased fringing leads to the strong interaction between the samples under test and the circular IDC, thereby amplifying its ability to detect subtle variations in permittivity due to varying glucose concentrations between 0.0 and 0.5 g/mL. The sensitivity enhancement due to increased electric field intensity can also be understood through perturbation theory (Pozar, 2011; Todi and Agarwal, 2025), where the frequency shift depends on the interaction between the fields and the SUT. As the electric field becomes stronger in the sensing region, the perturbation increases, resulting in a larger frequency shift and improved sensitivity. Moreover, any defect in the ground structure induces changes in the effective capacitance and inductance of the transmission line (Khandelwal et al., 2017), which further improved the performance of the Prototype-II. Therefore, although the overall radiation metrics of the proposed CIDCS Prototype-II may have degraded compared to Prototype-I, the experimental findings affirm that Prototype-II offers significant enhancement in sensing applications, where greater sensitivity to variations in the permittivity of the medium is more important than achieving high radiation efficiency.

A comparative evaluation of the proposed CIDCS prototypes I and II with the work documented in the literature is presented in Table 4. The comparison is based on various aspects, including sensor size and design, type of substrate used, operating frequency, ground configurations (whether partial or defected), sensitivity, coefficient of linearity, and the presence of comparative evaluation. It is worth noting that most of the sensors utilized FR4 as substrate (Nella et al., 2023a; Nella et al., 2023b; Singh et al., 2023; Rahman et al., 2019; Njokweni and Kumar, 2020), making it a practical choice for sensor implementation. The high linearity in the CIDCS Prototype-II supports the reliable and predictable behavior of the sensor. Microwave designs with partial/defected ground conductors have been documented in the literature for size reduction (Lee et al., 2010; Alqahtani et al., 2022; Njokweni and Kumar, 2020; Mobashsher and Abbosh, 2015), bandwidth improvement (Tiwari et al., 2021; Tiwari et al., 2022; Parameswari and Chitra, 2021; Alqahtani et al., 2022), UWB applications (Chen et al., 2007; Gopi et al., 2021; Kanagasabai et al., 2021), better impedance matching (Douhi et al., 2025; Alqahtani et al., 2022), and sensitivity enhancement (Njokweni and Kumar, 2020; Zhang et al., 2022; Abdelwahab et al., 2021). The reported studies are focused on improved performance of the device using partial/defected ground; however, quantitative comparative evaluation demonstrating the improvement w.r.t the original design is not provided. In contrast, this work presents a systematic comparison highlighting both the CIDCS prototypes to demonstrate the improvement. Here, a quantitative analysis is presented to understand the impact of the presence/absence of the ground conductor beneath the sensing area in the same design on sensor performance. This experimental study attempts to fill the gap by conducting a comparative analysis to quantify the role of the ground conductor in enhancing sensitivity.

TABLE 4. Comparative evaluation of the proposed approach with reported designs.

Ref.Sensor designSubstrateSensor size (mm 2)Operating frequency (GHz)Glucose concen-trationGroundSensitivityR 2Quantified comparative study of partial vs. complete groundRemarks
(Nella et al., 2023a), Nella et al. (2023b)Planar Yagi-Uda AntennaFR430 × 405.50-500 mg/dLPartial with inverted L-shaped stub−−No−
Singh et al. (2023)Antenna composed of two rectangular strips placed perpendicular to each otherFR430 × 305.70-900 mg/dLDefected0.089 MHz/(mg/dL)0.9339No−
El Gharbi et al. (2021)Circular ring monopole antennaFelt Textile37 × 372.45%, 10% and 20% in waterPartial550 MHz/mL for 5% sugar concentration−No−
Rahman et al. (2019)Circular patch with semi-circular-shaped slotsFR426 × 244.320%, 40%, 60% and 80% in waterPartial and slotted−−No−
Njokweni and Kumar (2020)Rectangular microstrip antennaFR451.3 × 51.32.420%, 50% and 80% (w/V) in waterTwo rectangle-shaped slots−−NoDefected ground to improve performance and reduce size of antenna
Zhang et al. (2022)Triple-ring microstrip patch antennaTaconic TLX-825 × 252.91, 3.46, and 4.5850-500 mg/dLPartial0.25 MHz/(mg/dL)0.995NoPartial ground to improve capacitance, inductance, and sensitivity
This workCircular IDC-shaped patchFR440 × 356.3450-0.5 g/mLComplete19.993 dB/g/mL and 59.157 MHz/g/mL0.927 & 0.973YesPerforms better as an antenna
This workCircular IDC-shaped patchFR440 × 352.1490-0.5 g/mLPartial29.871 dB/g/mL and 71.169 MHz/g/mL0.930 & 0.995YesPerforms better as a sensor

6 Conclusions and future research directions

In this paper, a comparative experimental study is presented, where two prototypes of the proposed microstrip-fed Circular InterDigital Capacitor Sensor (CIDCS) have been demonstrated, one with a complete ground conductor and the other with a partial ground conductor on the backside of the substrate. The experimental findings reveal that the prototype with a complete backside ground conductor maintains radiation characteristics such as gain, directivity, efficiency, and front-to-back ratio. In contrast, incorporating a partial backside ground conductor changes the effective capacitance and inductance, enhances fringing fields, and strengthens the interaction of the sample with the sensing area, thereby improving sensitivity at the cost of reduced radiation performance. The observed trade-off validates the use of partial ground conductors as an effective design approach when high responsiveness is prioritized over radiative efficiency, particularly in sensing applications such as glucose monitoring, where even minor variations in the dielectric constant need to be captured accurately.

The proposed CIDCS prototypes I and II are contact-based, real-time, and label-free, offering a robust platform for glucose sensing. The repeatability observed in the RF experiments showed that the sensor performs consistently and reliably under identical environmental conditions, which is the key factor for any practical application. Despite these promising results, several challenges require further investigation. The RF measurements have been performed in controlled environmental conditions, which ensured the repeatability of the sensor, but the reproducibility of the experimental results warrants further analysis. Although the CIDCS prototypes demonstrated excellent sensor performance, there is scope for improving sensitivity to achieve higher precision and consistency. Future work would focus on sensor design optimization by refining the geometry of the sensing region and exploring alternative sample placement methods to increase sensitivity and enable the detection of lower glucose concentrations. In addition to performance optimization, future research would focus on multi-parameter analysis, examining shifts in inductance, capacitance, Q-factor, signal-to-noise ratio (SNR), and S 11 phase for multi-dimensional analysis. By leveraging the multidimensional data, the sensor can achieve higher specificity, improved sensitivity, and more robust detection.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Francesca Benassi, University of Bologna, Italy

Reviewed by: Merih Palandoken, Izmir Kâtip Çelebi University, Türkiye

Syah Alam, Trisakti University, Indonesia

Statements

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Funding

The author(s) declared that financial support was received for this work and/or its publication.

Data availability

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

Author contributions

ST: Conceptualization, Data curation, Methodology, Validation, Writing – original draft. PA: Conceptualization, Formal Analysis, Funding acquisition, Supervision, Writing – review and editing.

Acknowledgements

The authors would like to thank UGC-UPE II by University Grant Commission (UGC), the DST- INSPIRE (Innovation in Science Pursuit for Inspired Research) Faculty Award Research Grant (IFA12-ENG-24) (2013-18), and the DST-PURSE (Promotion of University Research and Scientific Excellence) program (2014-19) by the Department of Science and Technology (DST), Government of India, for the financial assistance. We would like to thank Prof. Shatendra Sharma and Mr. Amrish K. Gajjar from the University Science Instrumentation Centre (USIC) at Jawaharlal Nehru University (JNU) for the workshop support. Swati Todi would like to thank UGC for the NON-NET Research Fellowship (2018-23) and Council of Scientific & Industrial Research (CSIR), Human Resource Development Group (Extra Mural Research Division) for the Research Associateship (09/0263(20473)/2024-EMR-I).

References

  1. Abdelwahab H., Ebrahimi A., Tovar-Lopez F. J., Beziuk G., Ghorbani K. (2021). Extremely sensitive microwave microfluidic dielectric sensor using a transmission line loaded with shunt LC resonators. Sensors 21, 6811. doi:10.3390/s21206811
  2. Adhikari K. K., Wang C., Qiang T., Kumar A., Wu Q., Maharjan S. (2019). Real-time accurate quantification of nanoliter ethanol using performance-optimized micro-fabricated microwave resonant sensor. J. Phys. D Appl. Phys. 53, 085402. doi:10.1088/1361-6463/ab5c68
  3. Alqahtani A., Islam M. T., Talukder M. S., Samsuzzaman M., Bakouri M., Mansouri S. (2022). Slotted monopole patch antenna for microwave-based head imaging applications. Sensors 22, 7235. doi:10.3390/s22197235
  4. Arabahmadi M., Mashayekhi V., Asgharivaskasi R., Azadi-Tinat N., Koohestani M. (2025). A simple and extremely-sensitive non-invasive coupled-line microwave sensor for biomedical monitoring applications. Sensors Actuators A Phys. 395, 117026. doi:10.1016/j.sna.2025.117026
  5. Balanis C. A. (2016). Antenna Theory: Analysis and Design. John Wiley and Sons.
  6. Beeresha R., Khan A., Manjunath Reddy H. (2016). Design and optimization of interdigital capacitor. Int. J. Res. Eng. Technol. 5, 73 – 78. doi:10.15623/ijret.2016.0533016
  7. Chen Z. N., See T. S., Qing X. (2007). Small printed ultrawideband antenna with reduced ground plane effect. IEEE Trans. Antennas Propag. 55, 383 – 388. doi:10.1109/TAP.2006.889823
  8. Douhi S., Eddiai A., Cherkaoui O., Mazroui M. (2025). Design of a compact, highly flexible, high-performance wideband all-textile antenna for wearable and portable IoT devices. Appl. Phys. A 131, 1 – 16. doi:10.1007/s00339-025-08342-z
  9. Ebrahimi A., Scott J., Ghorbani K. (2019). Ultrahigh-sensitivity microwave sensor for microfluidic complex permittivity measurement. IEEE Trans. Microw. Theory Tech. 67, 4269 – 4277. doi:10.1109/TMTT.2019.2932737
  10. El Gharbi M., Martinez-Estrada M., Fernandez-Garcia R., Gil I. (2021). Determination of salinity and sugar concentration by means of a circular-ring monopole textile antenna-based sensor. IEEE Sensors J. 21, 23751 – 23760. doi:10.1109/JSEN.2021.3112777
  11. Frau I., Wylie S., Byrne P., Onnis P., Cullen J., Mason A. (2021). Microwave sensors for in situ monitoring of trace metals in polluted water. Sensors 21, 3147. doi:10.3390/s21093147
  12. Gangwar A. K., Alam M. S. (2019). A miniaturized quad-band antenna with slotted patch for WiMAX/WLAN/GSM applications. AEU-International J. Electron. Commun. 112, 152911. doi:10.1016/j.aeue.2019.152911
  13. Gharbi M. E., Fernández-García R., Gil I. (2021). Textile antenna-sensor for in vitro diagnostics of diabetes. Electronics 10, 1570. doi:10.3390/electronics10131570
  14. Gopi D., Vadaboyina A. R., Dabbakuti J. K. (2021). DGS based monopole circular-shaped patch antenna for UWB applications. SN Appl. Sci. 3, 198. doi:10.1007/s42452-020-04123-w
  15. Gugliandolo G., Naishadham K., Crupi G., Donato N. (2022). Design and characterization of a microwave transducer for gas sensing applications. Chemosensors 10, 127. doi:10.3390/chemosensors10040127
  16. Han X., Liu K., Zhang S. (2024). High-sensitivity dual-band microfluidic microwave sensor for liquid dielectric characterization. IEEE Sensors J. 24, 36689 – 36697. doi:10.1109/JSEN.2024.3469529
  17. Haq T., Koziel S. (2022). Inverse modeling and optimization of CSRR-Based microwave sensors for industrial applications. IEEE Trans. Microw. Theory Tech. 70, 4796 – 4804. doi:10.1109/TMTT.2022.3176886
  18. Harnsoongnoen S., Buranrat B. (2023a). Advances in a microwave sensor-type interdigital capacitor with a hexagonal complementary split-ring resonator for glucose level measurement. Chemosensors 11, 257. doi:10.3390/chemosensors11040257
  19. Harnsoongnoen S., Buranrat B. (2023b). Glucose concentration monitoring using microstrip spurline sensor. Meas. Sci. Rev. 23, 168 – 174. doi:10.2478/msr-2023-0022
  20. Hosseinzadeh S., Pakvarjouy M. (2025). Microwave-based single port glucose sensor arranging interdigital and complementary split ring resonator. Measurement 246, 116695. doi:10.1016/j.measurement.2025.116695
  21. Javadizadeh S., Badieirostami M., Shahabadi M. (2023). Ultrasensitive miniaturized planar microwave sensor for characterization of water–alcohol mixtures. Sci. Rep. 13, 14144. doi:10.1038/s41598-023-41035-2
  22. Kanagasabai M., Sambandam P., Alsath M. G. N., Palaniswamy S., Ravichandran A., Girinathan C. (2021). Miniaturized circularly polarized UWB antenna for body centric communication. IEEE Trans. Antennas Propag. 70, 189 – 196. doi:10.1109/TAP.2021.3098517
  23. Khan M. R. R. (2024). Development of a battery-free, chipless, and highly sensitive radio frequency glucose biosensor. Micromachines 15, 272. doi:10.3390/mi15020272
  24. Khandelwal M. K., Kanaujia B. K., Kumar S. (2017). Defected ground structure: fundamentals, analysis, and applications in modern wireless trends. Int. J. Antennas Propag. 2017, 2018527. doi:10.1155/2017/2018527
  25. Kumar A., Wang C., Meng F. Y., Zhou Z. L., Zhao M., Yan G. F. (2020). High-sensitivity, quantified, linear and mediator-free resonator-based microwave biosensor for glucose detection. Sensors 20, 4024. doi:10.3390/s20144024
  26. Lee M.-J., Pyo S., Yoon W.-S., Shin I.-C., Kim Y.-S. (2010). A size reduced CRLH resonant antenna based on interdigital capacitors with defected ground structure. Microw. Opt. Technol. Lett. 52, 2142 – 2145. doi:10.1002/mop.25362
  27. Lin C.-C., Jin P., Ziolkowski R. W. (2011). Single, dual and tri-band-notched ultrawideband (UWB) antennas using capacitively loaded loop (CLL) resonators. IEEE Trans. Antennas Propag. 60, 102 – 109. doi:10.1109/TAP.2011.2167947
  28. Mansour E., Allam A., Abdel-Rahman A. B. (2023). A novel approach to non-invasive blood glucose sensing based on a single-slot defected ground structure. Int. J. Microw. Wirel. Technol. 15, 32 – 40. doi:10.1017/S1759078722000174
  29. Masi A., Brizi D., Monorchio A. (2023). Millimetric inclusion detection through a contactless microwave spiral sensor for biomedical applications. IEEE Sensors J. 23, 12796 – 12807. doi:10.1109/JSEN.2023.3271412
  30. Mazumder A., Biswas A., Sau T. K., Bhimalapuram P., Syed A. (2025). Radio frequency based detection of specific and nonspecific protein interaction. Front. Sensors 6, 1511104. doi:10.3389/fsens.2025.1511104
  31. Mobashsher A. T., Abbosh A. (2015). Utilizing symmetry of planar ultra-wideband antennas for size reduction and enhanced performance. IEEE Antennas Propag. Mag. 57, 153 – 166. doi:10.1109/MAP.2015.2414488
  32. Nascimento D. C., da S., Lacava J. C. (2011). Design of low-cost probe-fed microstrip antennas. Microstrip Antennas (IntechOpen). 1 – 28. doi:10.5772/14523
  33. Nella A., Aldhaheri R. W., Kamili J. B., Sobahi N. M. (2023a). A non-invasive method of glucose monitoring using FR4 material based microwave antenna sensor. Sci. Eng. Compos. Mater. 30, 20220187. doi:10.1515/secm-2022-0187
  34. Nella A., Aldhaheri R. W., Kamili J. B., Sobahi N. M. K. A. (2023b). A non-invasive method of monitoring glucose in blood using a planar Yagi-Uda antenna and microstrip filter. Elektron. ir Elektrotechnika 29, 33 – 39. doi:10.5755/j02.eie.31173
  35. Njokweni S., Kumar P. (2020). Salt and sugar detection system using a compact microstrip patch antenna. Int. J. Smart Sens. Intelligent Syst. 13, 1 – 9. doi:10.21307/ijssis-2020-027
  36. Palandoken M., Gocen C. (2025). Microwave sensor designs for liquid material dielectric characterization: technological advances and applications. Sensors Actuators A Phys. 387, 116381. doi:10.1016/j.sna.2025.116381
  37. Parameswari S., Chitra C. (2021). Compact textile UWB antenna with hexagonal for biomedical communication. J. Ambient Intell. Humaniz. Comput., 1 – 8. doi:10.1007/s12652-021-03228-3
  38. Piekarz I., Skarzynski K., Piekarz B., Wincza K., Gruszczynski S., Sloma M. (2024). Additively manufactured microwave sensor for glucose level detection in saliva. Sci. Rep. 14, 28235. doi:10.1038/s41598-024-79867-1
  39. Pozar D. M. (2011). Microwave Engineering. John Wiley and Sons.
  40. Rahman M. N., Islam M. T., Samsuzzaman Sobuz M. (2018). Salinity and sugar detection system using microstrip patch antenna. Microw. Opt. Technol. Lett. 60, 1092 – 1096. doi:10.1002/mop.31108
  41. Rahman M. N., Hassan S. A., Samsuzzaman M., Singh M. S. J., Islam M. T. (2019). Determination of salinity and sugar concentration using microwave sensor. Microw. Opt. Technol. Lett. 61, 361 – 364. doi:10.1002/mop.31588
  42. Rahman M. M., Islam M. S., Islam M. T., Al-Bawri S. S., Yong W. H. (2022). Metamaterial-based compact antenna with defected ground structure for 5G and beyond. Comput. Mater. Continua 72, 2383 – 2399. doi:10.32604/cmc.2022.022150
  43. Saadat-Safa M., Nayyeri V., Ghadimi A., Soleimani M., Ramahi O. M. (2019). A pixelated microwave near-field sensor for precise characterization of dielectric materials. Sci. Reports 9, 13310. doi:10.1038/s41598-019-49767-w
  44. Samuel T. S. A., Balamurugan C., Raj M. G., Mary S. P., Geege A S. (2025). Enhancing gain and bandwidth in a multiband microstrip patch antenna through L-slot and partial ground plane integration. Sci. Rep. 15, 1 – 12. doi:10.1038/s41598-025-13947-8
  45. Sari F. K., Alam S., Surjati I., Mardian R. D., Sari L., Kusumah M. N. (2025). A single-port microwave sensor based on interdigital capacitor and electromagnetic coupled structure for permittivity detection of vegetables materials. Sens. Imaging 27, 7. doi:10.1007/s11220-025-00708-0
  46. Singh T., Mishra P. K., Pal A., Tripathi V. S. (2023). A planar microwave sensor for noninvasive detection of glucose concentration using regression analysis. Int. J. Microw. Wirel. Technol. 15, 1343 – 1353. doi:10.1017/S1759078723000545
  47. Soni G. K., Yadav D., Kumar A., Jain P., Yadav M. V. (2024). Design and optimization of flexible DGS-based microstrip antenna for wearable devices in the Sub-6 GHz range using the nelder-mead simplex algorithm. Results Eng. 24, 103470. doi:10.1016/j.rineng.2024.103470
  48. Tiwari B., Gupta S. H., Balyan V. (2021). Investigation on performance of wearable flexible on-body ultra-wideband antenna based on denim for wireless health monitoring. J. Electron. Mater. 50, 6897 – 6909. doi:10.1007/s11664-021-09222-x
  49. Tiwari B., Gupta S. H., Balyan V. (2022). Comparative exploration of diverse substrate materials on performance of ultra wide band antenna design for on body WBAN applications. Wirel. Personal. Commun. 124, 3661 – 3684. doi:10.1007/s11277-022-09532-1
  50. Todi S., Agarwal P. (2025). Sensitivity enhancement by introducing altered ground conductors around the sensing area of the coplanar meander-interdigital capacitor structure sensor. Arabian J. Sci. Eng. 51 (6), 7801 – 7820. doi:10.1007/s13369-025-10530-1
  51. Todi S., Patel S., Agarwal P. (2026). Bi-functional glucose Sensing-Transmission (Sens-Tra) sensor for IoT-based applications. Ad Hoc Netw. 181, 104086. doi:10.1016/j.adhoc.2025.104086
  52. Yeo J., Lee J.-I. (2019). High-sensitivity microwave sensor based on an interdigital-capacitor-shaped defected ground structure for permittivity characterization. Sensors 19, 498. doi:10.3390/s19030498
  53. Yunos M. F. A. M., Manczak R., Guines C., Mansor A. F. M., Mak W. C., Khan S. (2021). Rf remote blood glucose sensor and a microfluidic vascular phantom for sensor validation. Biosensors 11, 494. doi:10.3390/bios11120494
  54. Zhang M., Yang X., Ren M., Mao S., Dhakal R., Kim N.-Y. (2022). Microfluidic microwave biosensor based on biomimetic materials for the quantitative detection of glucose. Sci. Rep. 12, 15961. doi:10.1038/s41598-022-20285-6
  55. Zhao J.-M., Wang Y.-K., Shi B.-W., Wang Y.-X., Jiang Y.-F., Yang G.-L. (2024). Microwave biosensor for the detection of growth inhibition of human liver cancer cells at different concentrations of chemotherapeutic drug. Front. Bioeng. Biotechnol. 12, 1398189. doi:10.3389/fbioe.2024.1398189

Citation

Todi, S. & Agarwal, P. (2026). Sensitivity analysis of the microstrip-fed circular InterDigital capacitor sensor with complete and partial backside ground conductor. Frontiers in Antennas and Propagation, Volume 4. https://doi.org/
Received
Accepted
Published
Volume
4 - 2026

Copyright

Copyright © 2026 Todi and Agarwal.

License: https://creativecommons.org/licenses/by/4.0/

*Correspondence: Swati Todi

Article metrics

Views
0
Downloads
0
Citations
0
Shares
0

Related articles

Original ResearchDesign and fabrication of a TE 50 mode SIW-fed Yagi antenna array at 5.8 GHzRidha Omrani, Siwar Louati, Halim BoutayebTechnology and CodePARISO: pixelated antennas and reconfigurable intelligent surface optimizerSekhar Sekharan, Reza Fazel-Rezai, Sima NoghanianOriginal ResearchHybrid deep unrolling and graph neural networks for super-resolution direction-of-arrival estimation in physics-informed antenna arraysSai Vinay Thattukolla