Novel Design of a Bandwidth Enhanced and Frequency Reconfigurable, Wearable Antenna for Body Centric Communication

Article information

J. Electromagn. Eng. Sci. 2024;24(3):264-275

Publication date (electronic) : 2024 May 31

doi : https://doi.org/10.26866/jees.2024.3.r.227

Devendra Kumar^1,, Dhirendra Mathur²

¹Department of Electronics & Communication Engineering, Rustamji Institute of Technology, Tekanpur, Gwalior, India

²Department of Electronics & Communication Engineering, Rajasthan Technical University, Kota, Rajasthan, India

^*Corresponding Author: Devendra Kumar (e-mail: dkraghav9@yahoo.co.in)

Received 2023 June 9; Revised 2023 September 7; Accepted 2023 December 8.

Abstract

I. Introduction

Microstrip patch antennas—widely used due to their low profile, compact size, and conformal design—are highly suitable for wearable applications, such as patient monitoring, sports activities, security, and firefighting. In patient monitoring applications, transmission and reception involving high data rates are necessary for various on-body/off-body devices. However, since microstrip antennas usually have a narrow bandwidth, they are insufficient for use in such wearable applications. Nonetheless, various techniques for bandwidth enhancement have been proposed in the literature. For instance, the use of a double U-shaped defective ground structure [1], folded defects [2], and L-shaped defects [3] have all been proposed for bandwidth improvement. Furthermore, stripline and L-shaped probes have been used to feed multilayer structures [4]. Some studies have also employed sequential phase feeding methods for bandwidth enhancement [5, 6]. In addition, using capacitive coupling with a small disc on top offered improvements in bandwidth in [7], while introducing multiple slots on a patch close to the resonant frequency also had the same effect [8, 9]. Similarly, implementing two parasitic patches in the annular area helped improve bandwidth [10], and so did a coplanar waveguide-fed antenna using modified ground [11]. Other proposed methods for bandwidth improvement include using a vertical coupling-based method with modified ground [12], and employing the dual resonance method using ethylene-vinyl acetate (EVA) foam as substrate [13]. Furthermore, the literature has attempted to improve antenna performance by experimenting with different substrate materials created using EVA foam and plastic polymers [14]. In addition, different techniques for realizing multiband operations have been explored, such as employing two stacked patches operating at two different frequencies [15], using three independent radiators that are switched using PIN diodes to operate at 1.575 GHz for global positioning systems (GPS) and at the 2.45 GHz and 5.2 GHz industrial, scientific, and medical (ISM) bands [16], and implementing a loop inspired dual/triple band operation that uses switching by PIN diodes for the 4G, 5G, and X bands [17]. In modern wireless applications, the use of reconfigurable antennas is preferred due to their ability to radiate more than one pattern at different frequencies and polarizations. For instance, [18] reported frequency and polarization reconfiguration using PIN switching diodes for L1, worldwide interoperability for microwave access (WiMAX), wireless local area network, and mobile bands. Frequency reconfiguration for multi-bands using microelectromechanical system (MEMS) switches has also been demonstrated in [19, 20]. A liquid dielectric-controlled polarization reconfigurable antenna for radio frequency identification has been proposed in [21], while stub switching by employing PIN diodes has been recommended for frequency reconfigurability in [22]. In addition, a frequency reconfigurable planar inverted-F antenna for GSM 850/900 and UMTS 2100 was presented in [23].

In this work, stacked layers and multiple slots near the resonant frequency are combined to improve the bandwidth of a wearable antenna. The upper patch bearing resonant slots provides the desired bandwidth enhancement. Moreover, low-cost flexible EVA foam material is used as the substrate and copper tape is used as the patch, while frequency reconfigurability is incorporated using a PIN diode. Finally, the antenna structure was tested and validated for wearable applications using an equivalent human body model.

II. Antenna Design

The dimensions of the proposed antenna were calculated using standard Eqs. (1)–(3) for microstrip design [24], as noted below:

where W is the width of the antenna, L is its length, ɛ_r indicates the relative permittivity of the substrate, ɛ_eff refers to the effective permittivity, h is the thickness of the substrate, f_o denotes the resonant frequency, and c is the velocity of light.

Notably, the antenna was simulated using the CST Microwave Studio. Its ground plane, created using copper material, had dimensions of 62 mm × 58 mm × 0.05 mm, while the substrate material was EVA foam having a relative permittivity of 1.2, loss tangent of 0.02, and size of 62 mm × 58 mm × 1.65 mm. The patch consisted of a copper tape of size 24.64 mm × 21.33 mm × 0.05 mm. A slot of size 8 mm × 11.66 mm was cut into the patch. Furthermore, a feed line of size 15.3 mm × 3 mm was used to excite the patch. The spacing between the feed and the patch was optimized to 2.5 mm on the left and right sides and 1.7 mm on the top of the feed. The basic antenna design is presented in Fig. 1.

Fig. 1

Geometry of a basic antenna.

III. Evolution of Antenna Design

The EVA foam substrate (thickness = 1.65 mm) was chosen for the proposed antenna because it is flexible and suitable for wearable applications, owing to its electrical and mechanical properties. First, the basic antenna with an EVA foam substrate of height 1.75 mm was designed for the 5.8 GHz frequency band. Energy was fed into the patch as coupling by the feed line. However, the bandwidth obtained at this frequency was 122 MHz, which is narrow and insufficient for high data rate transmission/reception. Therefore, for bandwidth enhancement, another patch was placed onto the first patch. As shown in Fig. 2(a), an upper patch of size 23 mm × 20 mm × 0.05 mm is placed at a height of 8.25 mm from the first patch, with air being the dielectric in the gap. Notably, since the thickness of the upper patch was 0.05 mm, it was difficult to suspend it in air and keep it parallel to the lower patch. Therefore, expanded polyurethane foam with a relative permittivity of ɛ_r = 1 was used in the air gap to support the upper patch. Consequently, the bandwidth increased to 203 MHz. Furthermore, two slots of equal size (4 mm × 4 mm) were created at the center of the upper patch, as shown in Fig. 2(b), with the spacing between them being 9 mm. The size of the slots and the upper patch were optimized to attain an enhanced bandwidth. Due to the creation of slots on the upper patch, the overall bandwidth increased to 223 MHz at a resonant frequency of 5.8 GHz. The flowchart of the entire fabrication process is shown in Fig. 2(c), and the antenna dimensions are noted in Table 1.

Fig. 2

Geometry of the antenna structure: (a) antenna with upper patch, (b) antenna with slots in the upper patch, and (c) flowchart of the design.

Table 1

Antenna dimensions (unit: mm)

IV. Results And Discussion

2. Antenna with Upper Patch

To achieve further enhancement in bandwidth, modifications were implemented on the upper patch. Two slots were added to the upper patch, and their locations were optimized to help widen the bandwidth. The parameters obtained for this antenna were S₁₁ = −29 dB at 5.8 GHz, bandwidth = 223 MHz, farfield directivity = 9.32 dBi, sidelobe level = −13.1 dB, HPBW = 58.7°, maximum gain over frequency = 5.0 dBi, and front-to-back ratio = 19.66 dB. A comparison of the simulated S₁₁, directivity, and gain for all three stages of antenna evolution (basic patch, incorporation of the upper patch, and slots) is illustrated in Fig. 3(a), 3(b), and 3(c), respectively. Notably, a minor tilt in the radiation pattern of the antenna was observed when the upper patch was introduced over the lower patch. This can be attributed to the phase difference between the waves radiated from the two patches—the path length between both patches created the phase difference. The related parameter values are summarized in Table 2. As shown in Fig. 3(d), the bandwidth of the antenna without slots on its upper patch is 203 MHz. Subsequently, upon adding slots onto the upper patch, the bandwidth increased to 223 MHz, as shown in Fig. 3(e).

Fig. 3

Antenna parameters: (a) S₁₁ (dB), (b) directivity (dBi), (c) gain (dBi), (d) bandwidth with upper patch, and (e) bandwidth with slots on the upper patch.

Table 2

Antenna parameters

Measurements were performed using N-5247A vector network analyzer (VNA) from Agilent Technology, the results of which were in good agreement with the simulation results. A comparison of the simulated and measured S₁₁ and gain on a single band is shown in Fig. 4(a) and 4(b), respectively. The simulated S₁₁ is −29 dB at the center frequency, while the polar gain pattern shows that most of the radiation occurs in the broadside direction, with very low radiation in the backward direction, showing a low level of sidelobes. This type of directional radiation pattern is best suited for body-centric communication. The curves depicted in Fig. 4(c) show simulated directivity of 9.32 dBi, gain from 5.0 dBi to 7.1 dBi, and radiation efficiency from −2.0 dB to −1.5 dB in the operating band. A photograph of the fabricated antenna is presented in Fig. 5. An equivalent electrical circuit diagram of the geometry shown in Fig. 2(b), created using ADS software, is presented in Fig. 6(a). The pass band signifies S₁₁ responses below −10 dB at the resonant frequency, while the stop bands signify responses above −10 dB.

Fig. 4

Antenna parameters: (a) simulated and measured S₁₁ on single band, (b) polar plot of the simulated and measured gains, and (c) simulated directivity (dBi), gain (dBi), and radiation efficiency (dB).

Fig. 5

The fabricated antenna: (a) without upper patch and (b) with upper patch.

Fig. 6

(a) Equivalent circuit, and (b) comparison of S₁₁ between CST and ADS.

To obtain the values of lumped components, first the quality factor was calculated using the bandwidth and resonance frequency. Subsequently, the values of the lumped components were calculated based on basic electrical theory. Notably, the calculated component values satisfied the 5.8 GHz resonant frequency. Furthermore, the S₁₁ data generated by CST software were simulated using ADS software to compare the S₁₁ values, as shown in Fig. 6(b). It is observed that the S₁₁ results produced by ADS are similar to those produced by CST, showing an error of 0.5% in the bandwidth. Therefore, using ADS software, the proposed design was successfully validated.

3. Frequency Reconfigurability

A simple and widely used technique for achieving frequency reconfigurability involves the use of PIN diodes to switch antenna geometry patterns. As shown in Fig. 7(a), a PIN diode is placed between the left edge of the feed line and the patch. The diode can be switched ON/OFF through appropriate biasing, as shown in Table 3. Therefore, when the diode is reverse-biased (Mode-1), the feed line is not directly connected to the patch, and energy is fed as a coupling. The antenna resonates at 5.8 GHz frequency to attain an enhanced bandwidth of 223 MHz. When the diode is forward-biased (Mode-2), the feed line is directly connected to the patch on the left side. The structure is then reconfigured, and the antenna carries out dual-band operations at both the 4.27 GHz and 5.8 GHz frequencies simultaneously.

Fig. 7

(a) PIN diode connection, (b) simulated and measured S₁₁ in the dual band mode, (c) radiation efficiency in the dual band mode, and (d) VNA measurement setup.

Table 3

Frequency switching

The PIN diode BAR 64-03W E6327 from Infineon Technology was used in this study. The specifications of this PIN diode are as follows: forward voltage = 0.8–1.1 V, forward current = 10–100 mA, forward resistance = 2.1Ω, reverse-biased capacitance = 0.20 pF, and reverse biased resistance = 3.4 kΩ. The simulated and measured S₁₁ in the dual band is presented in Fig. 7(b). The bandwidth attained in the single band (Diode OFF) mode was 223 MHz (5.651–5.874 GHz) at the resonant frequency of 5.8 GHz. In ON mode, the antenna exhibited dual band characteristics, attaining a bandwidth of 110 MHz (4.210–4.320 GHz) at 4.27 GHz (WBAN) and a bandwidth of 105 MHz (5.760–5.865 GHz) at 5.8 GHz (ISM). Although the antenna was primarily designed for the 5.8 GHz band for biomedical applications, the reconfiguration mode enabled an additional band at 4.27 GHz while maintaining the original band at 5.8 GHz with 50% bandwidth. Such dual-band characteristics are useful for short-range communication, since bandwidths of 20/40/80/160 MHz are also adopted for body area network applications. Fig. 7(c) shows that the radiation efficiency is −2.3 dB at 4.27 GHz and −5.0 dB at 5.8 GHz in dual band mode. Furthermore, the VNA measurement setup is shown in Fig. 7(d), while the diode equivalent diagram is presented in Fig. 8. In the forward-biased condition, a resistance of 2.1 Ω and inductance of 1.8 nH formed a series circuit. In reverse-biased conditions, a capacitance of 0.2 pF and resistance of 3.4 kΩ formed a shunt circuit. The phase curves in Fig. 9(a) show that the phase is zero at both resonant frequencies. The Z₁₁ (impedance) curve in Fig. 9(b) indicates that the input impedance of the antenna is 51.3 Ω at 4.27 GHz and 5.8 GHz resonant frequencies, which is close to the reference impedance of 50 Ω. The current distribution in the dual-band operation of both patches is shown in Fig. 9(c). The simulated and measured radiation patterns at the dual band frequencies are depicted in Fig. 9(d) and 9(e).

Fig. 8

Diode equivalent diagram: (a) ON state and (b) OFF state.

Fig. 9

Dual band parameters: (a) phase curve, (b) impedance curve, (c) current distribution over patches, (d) radiation pattern at 4.27 GHz, and (e) radiation pattern at 5.8 GHz.

4. Effects of Feed Width Variation

The effects of variations in the width of the feed line on antenna performance are depicted in Fig. 10. It is evident that an increase in the width of the feed line leads to an increase in the gain of the antenna, while the resonant frequency remains constant up to a certain width and then increases sharply. Meanwhile, the directivity increases, exhibiting only slight variations, and the front-to-back ratio remains almost constant, with very slight variations.

Fig. 10

Effects of feed width variation.

5. Effects of Ground Plane Size Variation

The results of the simulations conducted on the antenna using different ground plane sizes are shown in Fig. 11. As the ground plane size increases, the bandwidth remains constant, exhibiting only minor variations of 0.86 MHz. Furthermore, the resonant frequency increases to reach a maximum of 10 MHz from 5.8 GHz. Directivity increases at a faster rate initially, but remains almost constant as the ground plane size increases.

Fig. 11

Effects of ground plane size variation.

A comparison of the current work with previous related research is presented in Table 4, showing the values achieved by each proposed antenna in terms of antenna size, frequency, bandwidth, gain, material, and application in the wireless body area network. Among the reported works, some antenna structures are large, with complex array designs that are able to attain a low percentage of bandwidth enhancement. In contrast, the proposed work achieved a high percentage of bandwidth enhancement and substantial gain using a simple design that is small in size and involves a low-cost, flexible substrate.

Table 4

Comparative study with past research

V. SAR Calculation

A wearable antenna is usually used along with on-body devices, which are responsible for establishing communication between the on-body and off-body devices for patient data transmission. For this kind of communication, the radiation pattern must be directional and propagate away from the body. Moreover, radiation directed toward the body is hazardous and, therefore, must be minimized. The simulated 3-dimensional radiation depicted in Fig. 12(a) shows that the radiated field is in the broadside direction, while minimal radiation is present in the back direction toward the body. Furthermore, to calculate the specific absorption rate (SAR), an equivalent human body model of size 94 mm × 90 mm × 70 mm was considered, as shown in Fig. 12(b). The thickness of the skin, fat, and muscle layers was 3 mm, 7 mm, and 60 mm, respectively. The dimensions and electrical parameters of the cubic human body model are noted in Table 5 [25]. The antenna was kept 2 mm away from the body model to provide space for wearable clothes. The calculated results presented in Fig. 12(c) show that the SAR value is 0.145 W/kg for 10 gram of tissue, which is well within the limits of the SAR value (2 W/kg for 10 gram tissue) prescribed by the Federal Communications Commission and the International Commission on Non-Ionizing Radiation Protection [26, 27].

Fig. 12

A 3D radiation and SAR calculation: (a) structure of the transparent 3D radiation pattern, (b) equivalent human body model, and (c) onbody measurement.

Table 5

Dimensions/electrical parameters of the equivalent human body model

Furthermore, the back radiation was calculated in accordance with Eq. (4) as the ratio of the radiated power in the back hemisphere from the antenna to the total radiated power, which can be expressed as follows:

Here, E is the electric field, θ is a variation of the angle of the electric field in the vertical plane, and φ is a variation of the angle of the electric field in the horizontal plane.

VI. Bending Effects

An equivalent cylindrical structure that accounts for the air gap and the skin layer of the body was created, as shown in Fig. 13(a). The outer cylinder with 2 mm thickness represents the air gap, while the inner cylinder with 3 mm thickness represents the skin layer. The bending effects of the antenna were studied by placing it on the surface of the cylinders with different radii. The results of S₁₁ based on the different radii are shown in Fig. 13(b). It was observed that the resonant frequency of the antenna drifted toward the higher side, while the return loss became poorer, as the radius (R) of the cylinder increased. However, no significant effect on the radiation pattern and efficiency of the antenna was observed due to bending.

Fig. 13

Bending effects: (a) cylinder mimicking the air gap and skin layer of the body and (b) S₁₁ at different radii.

VII. Conclusion

This work proposes a simple design for a low-cost frequency reconfigurable dual-band antenna that can be used to establish wireless connectivity between on-body and off-body devices, enabling data transmission from the patient’s body to other remote monitoring places. Substantial gain, bandwidth, and directional radiation patterns are the primary requirements of high-rate data transmissions in biomedical telemetry. Exhibiting enhanced bandwidth and directional radiation, the proposed antenna met all these requirements. Furthermore, the antenna achieved low back radiation toward the body, a high front-to-back ratio, and a very low SAR value, all of which are highly desired parameters for wearable applications. A parametric study was conducted to understand the effects of variation in dimension on the parameters of an antenna. The physical properties of the EVA foam substrate and the stability of the antenna parameters, such as resonant frequency, bandwidth, directivity, and front-to-back ratio, under body loading conditions make it suitable for telemetry applications.

Article information Continued