Introduction

With the increasing demand for wireless medical solutions with implantable medical concerns, implantable antennas for wireless biomedical applications have become one of the most talked about technologies. In order to have a detailed understanding of a patient’s health condition, health condition monitoring is needed in daily life1. Examples include blood pressure and glucose monitoring2,3,4, capsule endoscopy5,6, and leadless pacemakers7,8,9. For blood pressure monitoring, a blood pressure monitoring device can be implanted near the artery in the human arm. The sensor measures the pressure changes in the artery and converts them into electrical signals, which are then amplified, filtered, and converted into digital signals for processing and transmission. The implantable antenna transmits the blood pressure data to an external device for storage and processing, ultimately providing accurate blood pressure data to help doctors monitor the patient’s blood pressure.

Implantable antennas must be implanted into the patient’s body. To ensure the patient’s comfort, the miniaturization of the antenna is crucial. Common miniaturization methods include: using substrates with a high dielectric constant10; introducing active elements11; designing the radiation patch into a serpentine structure12,13; opening U-shaped variant slots14,15,16; opening variant slots based on the inverted F structure and adding short-circuit vias to introduce the electrical load effect16. 17 by using a spiral radiator with two symmetrical arms and introducing an open-ended slot in the ground, the antenna’s size is significantly reduced, and its volume is only 17.15 mm3 (7 mm × 6.5 mm × 0.377 mm). Working in the complex electromagnetic environment of the human body, while achieving miniaturization, it is also necessary to expand the bandwidth as much as possible and enhance the anti-interference ability. Using low-loss, flexible biocompatible materials such as polyamide18 or Taconic CER-1019,20, or embedding curved slots21, zigzag slots22,23 and C-shaped slots24 in the radiation patch and ground of the antenna can significantly reduce the antenna size while obtaining good bandwidth performance. Among them,25 proposed an ultra-miniaturized antenna with a volume of only 10.66 mm3 and an ultra-wide bandwidth of 3040 MHz (0.79—3.83 GHz) for leadless pacemakers. 26introduced an antenna design for wireless capsule endoscopes, and the frequency bandwidth of this antenna ranges from 0.61 to 1.51 GHz, with a fractional bandwidth (FBW) of 84.91%. However, the above work is mainly based on linear polarization. Considering the continuous change of the position of the monitoring equipment in the human body linear polarization may not be sufficient. Therefore, circularly polarized (CP) antennas become an effective solution to this problem. The use of reactive impedance substrates (RIS) design27, symmetric L-shaped slots on the radiation patch28, curved slots29 and slit rings30, etc. among which30, the designed implantable antenna achieves circular polarization in three frequency bands. These designs have been cleverly applied to the miniaturization, circular polarization, and bandwidth expansion of patch antennas, providing ideas for our research.

This paper presents a novel miniaturized ultra-wideband CP implantable antenna that operates in the ISM band, has a very small size (17.78 mm3) and a simple structure for continuous blood pressure monitoring. Miniaturization and matching of the desired frequency are achieved by introducing a through-hole between the rectangular radiator and the ground plane, etching two asymmetric T-slots on the ground plane, etching a cross-shaped slot in the middle of the radiating unit, and etching four L-slot curves on the edges. Meanwhile, the bandwidth adjustment and AR bandwidth expansion were achieved by adjusting two asymmetric T-slots in the ground plane. The depth of the implanted three-layer model (skin-fat-muscle model) was 11 mm at 2.45 GHz operating frequency, as shown in Fig. 2d. An impedance bandwidth of 46.5% (1.85–2.99 GHz), a maximum gain of − 28.8 dBi, and an AR bandwidth of 23.6% (2.36–2.94 GHz) were achieved. In order to realistically simulate the usage environment, we also simulate it in the arm model, as shown in Fig. 1. The geometry of the antenna and parameters are shown in Fig. 2a–c, and the values are listed in Table 1. SAR and maximum input power calculations are also performed for this antenna system for safety reasons (Table 2). The designed antenna is compared with the same type of work as shown in Table 3. Compared to28 and29, our design has some advantages in terms of size, BW and ARBW. The antenna in30 is a very good job, it has three working frequency bands with high gain, but its absolute size (50.8 mm3) is large due to it covers low frequency band. Compared with31 and32, our work is slightly inferior in terms of the electrical length size at the lowest operating wavelength and gain, but has a very obvious advantage in terms of bandwidth and circular polarization. Finally, the designed antenna has been fabricated and experimentally verified in a crushed pork and the results show that the antenna has excellent impedance bandwidth and circular polarization bandwidth performance.

Fig. 1
figure 1

Antenna implanted in the right-hand arm model.

Full size image
Fig. 2
figure 2

Geometry of antenna. (a) Radiation plane. (b) Ground plane. (c) Side view. (d) Three-layer mode.

Full size image
Table 1 Geometric parameters optimized for the proposed antenna (unit: mm).
Full size table
Table 2 Relevant dielectric properties of tissue models at 2.45 GHz.
Full size table
Table 3 Proposed antenna comparison with references.
Full size table

Topology and design strategy

Geometry of the antenna

The structure of the ultra-wideband CP antenna is shown in Fig. 2a–c. Rogers 6010 with dielectric constant εr = 10.2, tangent loss tan(δ) = 0.0027, and thickness of 0.254 mm was selected for both the antenna substrate/covering layer. The use of a material with high dielectric constant as the substrate reduces the resonance frequency and thus achieves the size reduction. In order to further miniaturize the antenna, a thin layer of the substrate covering and the short-circuiting probes are added, while the resonance frequency is shifted to a more lower frequency band to obtain a highly miniaturized antenna (antenna size of 0.056 × 0.04 × 0.004 λ03). The addition of the cover layer brings some additional advantages, as it not only protects from contact with human tissue, but also stabilizes fluctuations in the effective dielectric constant around the antenna. Since a symmetrical layout is prone to circular polarization, four L-shaped slots symmetrical about the origin and one cross-shaped slot are cut diagonally in the radiating patch. Two asymmetric T-shaped slots cut in the ground plane help to extend the impedance bandwidth and AR bandwidth. In addition, the 50 Ω coaxial feed that excites this antenna is located at (x, y), and the short-circuit probe is placed at the optimal position in the upper left corner (N4). Together these designs achieve good miniaturization, impedance bandwidth and CP characteristics.

Simulation setup and environment

The design and simulation experiments are carried out in the High Frequency Structure Simulator (HFSS) computed using the finite element method, and considering the interaction with biological tissues, the implantable antenna of our design is embedded in a three-layer model (skin-fat-muscle) with the dimensions of (60 × 60 × 44) cubic millimeters for the simulation, and the relevant dielectric properties of the tissue model at 2.45 GHz are shown in Table 230,31, in which The thickness of both the skin and fat layers is 4 mm, the thickness of the muscle layer is 36 mm, and the tissue is surrounded by a 100 × 100 × 100 mm3 cubic radiation box. The device was then simulated in a three-layer model. After meeting the desired requirements, the device was implanted in ground pork to a depth of 11 mm for actual measurements. This provided a reliable performance benchmark for practical in vivo implantation applications.

Antenna design

The designed antenna achieves the design requirements in four steps. First, the original dimensions of the antenna are calculated by the following formula30:

$$ fr = \frac{c}{{\lambda g\sqrt {\varepsilon eff} }} \approx \frac{c}{{{\text{L}}g\sqrt {\frac{\varepsilon r + 1}{2}} }} $$
(1)

where fr is the resonant frequency, λg represents the guided wavelength, εeff represents the effective permittivity, εr is the relative permittivity, and Lg is the length of the antenna. The resonant frequency of the traditional patch antenna is 3.94 GHz. The design starts with a patch antenna with a complete ground plane and the introduction of a short-circuit probe. Firstly, in Step I, the patch antenna is slotted in an L-shape and cross shape to extend the current path and increase the capacitive effect, resulting in a reduction of the antenna resonance to 1.5 GHz. Meanwhile, the cross-shaped slot also excites resonance at 3.4 GHz, but this does not meet the requirements for working in the ISM band. Therefore, the patch is further slotted, and due to the symmetrical layout that tends to produce CP, the slots are symmetrically centered on the origin of the antenna. This causes the frequency to decrease to 1.4 GHz and 2.2 GHz, and a resonance is generated at 3.7 GHz, and the AR curve becomes flatter. However, as shown in Step II, the result is still unsatisfactory. For the existing resonance frequencies that need to be moved to a satisfactory band, some changes must be made to the ground plane. Therefore, in Step III, the circuit path is further expanded by slotting and the capacitive effect is increased by creating an asymmetric T-slot in the ground plane. The introduction of the asymmetric T-slot allows the antenna to combine the two adjacent modes in Step II, forming a wide bandwidth between 1.85 GHz and 2.99 GHz. At the same time, the AR curve becomes flatter, and its value is much lower than without the asymmetric T-slot (although still not less than 3 dB). In Step IV, another asymmetric T-slot is added. The structure design of the two asymmetric T-slots excites two orthogonal modes with equal amplitudes and a phase difference of 90°, achieving broadband characteristics and CP characteristics. This series of design adjustments, especially the introduction of the asymmetric T-slot, significantly improves the performance of the antenna in the desired frequency band. The |S11| and AR curves for each step demonstrate the evolution of the frequency and circular polarization response during the design process, highlighting the importance of the asymmetric T-slots introduced in Steps III and IV for the ultra-wideband characteristics and AR properties. Detailed information can be found in Fig. 3b–c.

Fig. 3
figure 3

(a) Design steps. (b) |S11| for each step of the design. (c) AR for each step of the design.

Full size image

Results and discussion

Antenna current distribution and radiation patterns

Figure 4a and b provides detailed information about the antenna at the 2.45 GHz frequency, the current distribution within the antenna, and the flow paths to help understand the formation and optimization of antenna performance. In the 2.45 GHz band, it is observed that the current on the patch is divided into two unidirectional flow paths. In the current distribution on the ground, it is observed that the strong currents are concentrated around two asymmetric T-slots, thus playing a key role in the tuning of the resonant frequency. In addition, at 2.45 GHz, the current distribution in the ground plane shows a circular path, which leads to energy loss, reducing the effective energy radiated into space by the antenna and decreasing the antenna gain. The radiation direction map under this current distribution is shown in Fig. 4c. In both the three-layer model and the arm model it has a better radiation pattern with the maximum radiation direction towards the outside of the body.

Fig. 4
figure 4

(a) Patch current distribution. (b) GND current distribution. (c) Radiation patterns for three-layer and arm models at the 2.45 GHz frequency.

Full size image

Parametric analysis

This section presents the effect of key parameters on the bandwidth and circular polarization of the designed antenna. The addition of L1 improves the impedance matching of the antenna in the ISM (2.45 GHz) operating band. the increase in g4 leads to a frequency shift towards lower frequencies and a long bandwidth of 1.85 -2.99 GHz occurs. The designed antenna achieves better performance when the parameters are optimized to L1 = 2 mm and g4 = 6.5 mm, detailed information is shown in Fig. 5a and b. Slotting not only affects the bandwidth but also the polarization; slotting introduces asymmetric current distribution and phase difference which affects the polarization of the antenna. We discuss the effect of varying the two T-slot parameters (g1, g2, g4) on the circular polarization. With increasing g1 and g4, the axial bandwidth of the antenna is shifted downwards to fully cover the operating band and the axial bandwidth increases significantly. With the increase of g2, the axial bandwidth of the antenna then covers the working band with a significant increase. The detailed information is shown in Fig. 6a–c, the circular polarization of the antenna is good when g1 = 4 mm, g2 = 4 mm and g4 = 6.5 mm.

Fig. 5
figure 5

(a) Effect of L1 change on |S11|. (b) Effect of g4 change on |S11|.

Full size image
Fig. 6
figure 6

(a) Effect of g1 change on CP. (b) Effect of g2 change on CP. (c) Effect of g4 change on CP. (d) Position of short pin.

Full size image

The change of short-pin position has a great effect on the antenna |S11| and antenna impedance matching, which affects the resonant frequency shift and the operating bandwidth. From Fig. 6d, it can be observed that for placing the short pin in the two positions of N1 and N3, it does not satisfy the required frequency band for operation. Meanwhile, for both N3 and N4, their impedance matching performance at 2450 MHz is good. However, the overall performance of N4 is slightly better than N3, which has a wider bandwidth. More details can be obtained by observing Fig. 6d, and finally N4 is chosen as the best short pin.

Mechanism of operation of circularly polarized antenna

Figure 7b presents the 3 dB AR representing the CP characteristics. The simulated AR bandwidth of the proposed antenna in the three-layer model is 23.6% (2.36–2.94 GHz) and the measured circularly polarized bandwidth is (2.37–2.87 GHz) 20.4%. By designing the radiating surface as a structure symmetric about the center of the antenna origin28 and introducing two asymmetric T-slots into the ground, this design excites two orthogonal modes with a 90° phase difference, resulting in circularly polarized characteristics (Fig. 8). The current distribution of the antenna provides an intuitive view to understand the circular polarization mechanism of the antenna. Figure 9 shows the current distribution of the antenna at an operating frequency of 2.45 GHZ for four phases from 0° to 270°, with the red arrows indicating the direction of the current vectors and sums at that phase angle. From the Fig. 9, it can be seen that the direction of the current combinations is all counterclockwise at this frequency, which indicates that the polarization is RHCP29,35. Also, the measurements in Fig. 8 show that the antenna exhibits RHCP characteristics. Good agreement is observed between the antenna’s simulated and measured radiation patterns at 2.45 GHz. The difference is due to the difference between the simulated and actual test environment and the loss of the RF connecting line.

Fig. 7
figure 7

Comparison of three-layer model, arm model simulation and real measurements in minced pork. (a) S-parameter comparison. (b) ARBW comparison.

Full size image
Fig. 8
figure 8

Radiation patterns of RHCP and LHCP at 2.45 GHz.

Full size image
Fig. 9
figure 9

Four phase current distribution of the proposed antenna.

Full size image

SAR analysis

Furthermore, since the designed antenna operates in human tissue, safety is an important issue to be considered and is usually evaluated in terms of SAR36. As shown in Fig. 10, the SAR distributions in the three-layer model and the arm model are displayed in the 2.45 GHz operating band. The simulated SAR values are 652.337 W/kg and 811.425 W/Kg, respectively, at an input power of 1 W. Based on the simulated 1 g average SAR values, the input power should not exceed 2.45 mw and 1.97 mw, respectively, in order to satisfy the limitations of the IEEE standard (IEEE C95.1-2019) for 1 g average SAR.

Fig. 10
figure 10

The simulated SAR distribution.

Full size image

Measurement results analysis

Figure 11 presents a physical picture of the antenna and the measurement environment. The |S11| in the three-layer model, right-hand arm model, and minced pork are shown in Fig. 7a. The proposed antenna resonates in the 2.45 GHz band with an impedance bandwidth of 875 MHz (1887–2762 MHz) measured in minced pork. Fabrication and soldering errors and the measured dielectric properties are not exactly the same as the simulated ones may be the reasons for the difference between the measured and simulated reflection coefficients. But the proposed antenna these broadbands adequately covers the ISM band (2400–2480 MHz). We measured the RHCP and LHCP by placing the antenna and the end of the pork in a microwave darkroom, and the results are shown in Fig. 8, where the main polarization of the antenna is the RHCP.

Fig. 11
figure 11

Physical drawing of the antenna and the measurement environment.

Full size image

Conclusions

A miniaturized ultra-wideband CP antenna for biotelemetry applications is presented. The antenna can operate in the ISM band (2.45 GHz) with circular polarization at 2.45 GHz. In addition, cross slots and multiple L-shaped slots are inserted in the radiating patch of the antenna, and two asymmetric T-shaped open slots are inserted in the ground plane, which play a key role in impedance matching, achieving circular polarization, extending AR bandwidth, and antenna size reduction. And the measurements of reflection coefficient and AR were carried out in the crushed pork. The measured results are in good agreement with the simulated results. In addition, we calculated the SAR according to the IEEE standard to ensure patient safety.