Flexible Antennas Design and Test for Human Body … · antennas reported on the literature, organic paper and Kapton® polyimide film were selected as the candidates for the UWB

1

Flexible Antennas Design and Test for Human Body

Applications Scenarios

João Vicente Faria

Instituto Superior Técnico

Universidade de Lisboa

Lisbon, Portugal

[email protected]

Abstract— This work presents a flexible Coplanar Waveguide

(CPW)-fed UWB monopole antenna used for applications in the

proximity of the human body. The antenna structure consist of a

leaf-shape radiating patch and a symmetric ground plane both

printed on the same side of the dielectric substrate. The proposed

antenna has been designed, and subsequently fabricated on two

different substrate materials (standard commercial paper and

Kapton®) using two distinct fabrication processes. The

simulated and measured results for the return loss (|S11|),

radiation patterns and total and radiation efficiencies are in good

agreement, showing that the antenna has a very good

performance in flat at condition. The antenna achieves an

impedance bandwidth from 2.9 GHz to beyond 12 GHz,

satisfying the UWB standards (3.1-10.6 GHz).

In order to easily conform to the body surface, the

suggested UWB antenna was designed to meet the requirements

of wearable devices, such as being flexible, compact and

mechanically robust. Thus, the influence of bending and

crumpling deformations on the antenna Radio-frequency (RF)

performance has been investigated. The simulated and measured

results show that the proposed antenna functions satisfactorily

under smooth bending and crumpling conditions. However, for

more severe deformation scenarios the antenna presents a

considerable performance degradation.

Finally, the Electromagnetic (EM) behavior of this

antenna has been examined in the vicinity of a human upper arm.

Regarding the computational simulations, four human arm

models with different geometric and dielectric properties have

been presented. The antenna return loss, total efficiency, and

radiation patterns have been analyzed for different antenna-body

distances. Measurements with a real human arm have also been

carried out with the antenna placed 3 mm and 6 mm away from

the human arm. Furthermore, the antenna Specific Absorption

Rate (SAR) was evaluated, showing values considerably below

the regulatory standards.

Keywords—Flexible antennas, Coplanar Waveguide

(CPW), Ultra-Wideband (UWB), wearable applications,

return loss, impedance bandwidth, bending, crumpling,

Specific Absorption Rate (SAR).

I. INTRODUCTION AND STATE OF THE ART

Over the last years, our World has been in a paradigm

transformation related to the emergence of a new technological

trend based on new information and communication systems [1].

Therefore, telecommunications have become a significant part

of everyday societies life-style, promoting the share of

information (voice, data, photos, videos) between people

through computers and other mobile devices. More recently,

there have been an increasing demand and progress in Wireless

Communication Systems. This segment of communications

industry are currently established in many developing countries,

replacing gradually the traditional Wired Communication

Systems [1].

In these types of communication systems, there are certain

issues like channel characterization, antenna transmitted power,

and the behavior of the EM fields near the human body which

have to be well-known by the antenna designers. Once the

human body has a high dielectric permittivity and low

conductivity at microwave frequencies, body tissues are an

uninviting and often hostile environment for the wireless signal

propagation. In general, antennas placed near the human body

suffer from reduced efficiency, due to electromagnetic

absorption in human tissues, radiation pattern deformation, and

variations in impedance at the feeding point.

Flexible antennas are developed making use of substrates that

can be easily integrated in irregular surfaces without losing its

functional features. The emerging of new materials, with very

specifics chemical properties, and new manufacturing processes

has led to the development of dielectric substrates that promise

good results regarding flexibility, efficiency, weight, and

reproducibility of the antennas.

The choice of the dielectric on which the conductor is built

influences the antenna's behavior regarding impedance

bandwidth, radiation efficiency and gain. Thickness, dielectric

permittivity of the material and the loss tangent must be

measured and controlled so as to reach the specifications of the

different applications.

2

The main objective of this work is to study and analyze a

flexible antenna for applications in the close proximity of the

human body. Since the antenna is designed to be mechanically

flexible, its capability to flex and function properly under

bending and crumpling conditions must be evaluated. The

interaction between the human body and the proposed flexible

antenna, and the influence on its performance, is also an issue to

be studied along this work. The proposed antenna is based on

[2], and its structure consists on a leaf-shaped monopole antenna

designed to operate at the entire UWB (3.1 GHz to 10.6 GHz)

frequency range. This work aims to investigate the possibility of

designing the proposed antenna structure using different

substrates. The designed antenna is optimized using CST

Microwave StudioTM and then fabricated through distinct

fabrication processes.

II. DESIGN OF A PRINTED CPW-FED UWB ANTENNA

A. Substrate Material Selection

Since the antenna is supposed to be mechanically flexible,

the suggested materials must present high malleability and

robustness levels, in order to tolerate certain deformation

scenarios, such as bending, crumpling, twisting and others. In

addition, these flexible materials should exhibit favorable

electromagnetic properties (loss tangent and relative

permittivity) to provide an easy integration with RF circuits and

to achieve the performance requirements of modern flexible and

wearable electronics. After reviewing a wide range of flexible

antennas reported on the literature, organic paper and Kapton®

polyimide film were selected as the candidates for the UWB

monopole antenna dielectric substrate materials. A comparative

study of the dielectric properties of the standard office paper and

Kapton® polyimide is presented in Table 1.

Table 1 - Kapton® and paper properties.

B. Antenna Geometry

The printed CPW-fed UWB monopole antenna proposed in

this work has an overall size of 37.5 mm x 33 mm (L x W). The

antenna geometry consists of a four-step monopole radiator and

the symmetrical rectangular ground plane (13 mm x 14.5 mm),

both placed on the same side of the substrate. The radiating patch

has a leaf shape, composed by a semielliptical base with a

diameter of 30 mm added to a centered-circle and two sided-

semicircles on its top. Moreover, a 50Ω coplanar waveguide of

3.2 mm width and 12.8 mm length is used to feed the radiating

patch. After performing an optimization process with CST

Microwave Studio™ software, the final antenna design

parameters are illustrated in Figure 1.

Figure 1- Geometry and design parameters of the proposed

UWB antenna in mm.

C. Fabrication Method

The fabrication process of flexible and wearable antennas

should guarantee good agreement with the design and simulation

results. The final antenna design was fabricated by two different

fabrication methods. The first step of the process was shared by

both methods and consists on exporting the antenna layout from

CST to AutoCAD software and design the production mask with

an appropriate scale factor (1:4 mm).

In the first fabrication procedure the metal patterns of the

antenna are photolithographically printed on the Kapton® film

substrate using the technology available in Instituto de

Telecomunicações (IT) RF laboratory.

The second fabrication method consists on attaching the

radiating patterns of the antenna directly on the paper substrate.

A CAMM-1 Servo X-24 Vinyl Cutter from Roland [3] available

in (EDP) FabLAB was used to cut off the antenna metal layer on

a flexible adhesive copper tape. The copper that is not part of the

antenna circuit was removed and the resultant radiating pattern

was then manually glued on a commercial sheet of paper. Both

prototypes were connected with a 50 Ω SMA connector, which

was carefully soldered on the antenna final prototype. Figure 2

show both fabricated antenna prototypes.

Figure 2 - Prototype I - Kapton® and Prototype II - Standard

office paper.

3

D. Final Prototype Selection

After obtaining the different prototypes for the proposed UWB monopole antenna, the next step is to measure the free-space return loss for both designs.

From the measured data presented in Figure 3, the prototype I

(Kapton®) shows a -10 dB impedance bandwidth that clearly

covers the targeted 3.1-10.6 GHz UWB standard, while the

prototype II presents an upshift from 2.9 to 4.15 GHz in the

lower -10 dB frequency edge. These unexpected results may be

due to the lack of precision in the substrate material (paper)

characterization, the low mechanical resolution of the cutter, and

the poor accuracy of the fabrication process. Taking into account

the presented results, prototype I was selected as the final

prototype of the monopole UWB antenna.

Figure 3 - Measured antenna |S11| using paper and Kapton®

substrates.

E. Free-Space Performance

In this subsection, the performance of the antenna presented

in Figure 4 is evaluated in free-space scenario. The results for

the return loss, |S11|, operating band, radiation patterns and

efficiencies of the antenna are fully studied.

Figure 4 - Antenna simulation model (left) and final antenna

prototype (right).

1) Return Loss and Operation Band

After the designing stage, the performance of the adopted

antenna was evaluated in free-space scenario. From Figure 5, it

can be seen that the UWB band was fully covered for both

simulated and measured cases. The lower edge of the operation

band corresponds to the same frequency point in the simulated

and measured |S11| curves. As can be seen from measurements,

some ripples occur particularly at the lower frequencies, due to

the RF cable effect [4].

Figure 5 - Simulated and measured |S11| curves in free-space

2) Radiation Patterns

Figure 6 presents the 2D radiation patterns of the proposed

antenna in yz and xz planes. The radiation patterns presented

below have been computed at three frequency points (3.1 GHz,

6.05 GHz and 10 GHz) using the CST Microwave Studio™.

Figure 6 - 2D simulated radiation patterns at 3.1, 6.05, and

10GHz (yz-plane on the left and xz-plane on the right).

It is observed that the yz-plane patterns are almost

bidirectional with large back lobes and looking like a doughnut

or a slightly pinched doughnut at the three selected frequency

points. In its turn, the xz-plane radiation pattern is

omnidirectional at lower frequencies (3.1 GHz). Moreover, the

xz radiation pattern is close to a multi-lobe shape at the higher

frequency of the band (10 GHz).

3) Radiation and Total Efficiency

The antenna radiation and total efficiencies were simulated in free-space, Figure 7.

4

Figure 7 - Simulated radiation and total efficiencies

The proposed antenna presents a very high efficiency at the

whole UWB band. The total efficiency is at least 95%, while the

radiation efficiency is very close to 100%. At some frequency

points the radiation efficiency slightly exceeds 100%, which is a

recurrent problem when the simulations are performed using

Computational Electromagnetics (CEM) tools [5].

The total and radiation efficiency of the proposed coplanar-

fed UWB antenna (Figure 8) was measured using a cavity-based

method. The adopted measurement procedure was based on an

extension of the generalized Wheeler Cap method [6] for

evaluating the efficiency of small UWB antennas.

Figure 8 - Estimated radiation and total efficiency.

Despite some missing data points (especially at higher

frequencies) that correspond to the invalid efficiency values

caused by cavity resonances, there is a reasonable agreement

between the simulations and measurements results of the

antenna efficiencies.

III. DEFORMATION EFFECTS ON THE ANTENNA

PERFORMANCE

In case of wearable antennas it is quite difficult to maintain a flat

antenna surface since these antennas should conform their shape

to the surface on which they are placed, without hindering the

user's movements and comfort. Moreover, the analysis of

antenna performance parameters under certain deformation

scenarios is crucial to guarantee a reasonable operation in the

close proximity of the human body. In this section, the proposed

UWB antenna is tested under various bending and crumpling

scenarios in order to investigate its potential behavior in

wearable applications.

A. Bending

The bending of the proposed antenna structure was

performed in two different directions: bending along the x–axis

(horizontal) and bending along the y-axis (vertical). In order to

mimic these bending deformations in CST Microwave StudioTM,

a vacuum cylindrical structure of radius R = 10 mm and R = 6

mm was directly attached on the antenna back-side substrate.

Then, the substrate layer and the radiating elements of the

antenna were separately bent around the cylinder. Figure 9

shows the simulation models for horizontal (x-axis) and vertical

(y-axis) bending of the proposed UWB antenna.

Figure 9 – Simulation models for horizontal (left) and vertical

(right) bending.

To recreate the bending scenario performed in CST

Microwave StudioTM, a 10 mm radius manufactured foam

cylinder was used. The antenna structure was directly bent

around the cylinder with the aid of two foam rings, as shown in

Figure 10.

Figure 10 - Bending measurements setup – horizontal bending

(left) and vertical bending (right).

1) Return Loss and Operation Band

Since the performance parameters are likely to be affected

during the bending operation, the return loss and the -10 dB

operation band of the proposed UWB antenna must be

evaluated under these deformations scenarios. The following

subsections present the comparison between the simulated and

measured |S11| when the proposed UWB antenna was

5

horizontally and vertically bent. This comparative analysis was

performed using a 10 mm bending radius.

a) Horizontal Bending (x-axis)

Figure 11 - Comparison between the simulated and measured

return loss for the horizontal bending scenario (R=10mm).

As it can be observed from Figure 11, the return loss of the

CPW-fed monopole antenna is slightly affected under x–axis

bending. For both simulated and measured results, some marked

resonances become evident on the return loss curves. Moreover,

the lower edge of the -10 dB operation band is shifted towards

higher frequencies.

b) Vertical Bending (y-aixs)

Figure 12 - Comparison between the simulated and measured

return loss for the horizontal bending scenario (R=10mm).

The simulated and measured return losses of the antenna in

the y-axis bending direction are shown in Figure 12 exhibiting

good matching over the whole UWB band. In this case, only

small resonance shifts and minor bandwidth variations can be

observed. In simulations, an accentuated resonance is observed

at approximately 7.7 GHz. For both simulations and

measurements a wide -10 dB bandwidth is achieved and thus the

UWB demands are perfectly reached.

2) Total Efficiency

The simulated total efficiency of the antenna was also

evaluated when the antenna is bent around the x-axis and around

y-axis. For this, two curvature radius are considered: R = 6 mm

and R = 10 mm.

a) Horizontal Bending (x-axis)

Figure 13 illustrates the simulated total efficiency when the proposed UWB is bent around its x–axis at two different curvature radius. According to the results, it is obvious that this bending scenario has a strong influence on the simulated total efficiency of the proposed CPW-fed antenna.

For a smooth bending deformation (R = 10 mm), the total efficiency remains acceptable (between 80% and 100%) for the entire UWB frequency range. Otherwise, when the antenna is subjected to a bending radius of 6 mm, the simulated total efficiency is considerably reduced

b) Vertical Bending (y-aixs)

In Figure 14, which is presented below, the comparison of

total efficiencies is presented for a vertical bending deformation.

It can be attended that, regardless of the bending radius adopted,

the total efficiency of the proposed UWB antenna does not

change.

Figure 14 - Simulated total efficiency for the vertical bending

scenario.

Figure 13 - Simulated total efficiency for the horizontal

bending scenario.

6

B. Crumpling

After analyzing the bending effects on the antenna

performance, the next step is to test the UWB monopole antenna

under crumpling conditions. Taking into account the small size

of the proposed antenna, crumpling deformations are not

expected to occur in a realistic usage scenario. This particular

deformation was only considered for the simulations.

The simulation method used to reproduce the crumpling

deformation is very similar to the one used for bending cases.

The antenna was laid over a cosine-type surface with different

crumpling amplitudes (CA = 1 mm; CA = 2 mm and CA = 4

mm), Figure 15.

This section describes the entire simulation procedure, and

presents the simulated results for the return loss and total

efficiency of the crumpled antenna.

Figure 15 - Antenna crumpling.

1) Return Loss and Operation Band

The simulated reflection losses of the antenna under various

crumpling cases are shown in Figure 16. It can be seen that

modifications on the antenna's impedance performance are more

pronounced, as the crumpling amplitude gets higher. Thus, the

bigger the crumpling amplitude is, the narrower is the antenna's

operation band. Among all of the considered crumpling

situations, only the CA = 1 mm case covers satisfactorily the

targeted 3.1-10.6 GHz UWB standard.

Figure 16 - Simulated return loss curves for crumpling

scenarios.

2) Total Efficiency

Figure 17 depicts the effect of the crumpling deformations on

the antenna simulated total efficiency. It can be seen that when

the crumpling amplitude increases from CA= 1 mm to CA= 4

mm, the antenna becomes less efficient.

Figure 17 - Simulated total efficiency for crumpling scenarios.

IV. ANTENNA PERFORMANCE IN THE HUMAN ARM

PROXIMITY

With the recent development of body-centric applications,

wearable antennas have become an essential part of WBAN

communication systems. Their design process and performance

evaluation are quite difficult, due to the presence of the user's

body [7]. The electrical characteristics of different body tissues

have a considerable effect on the performance of antennas that

operate in close proximity of the human body. Thus, one of the

main objectives of this section is to understand the interactions

between the human body and the EM waves radiated from

flexible antennas used in a wearable scenario. Studying the

nature of dispersive human tissues and their electrical and

geometrical properties, and how the proposed UWB monopole

antenna performance parameters are changed by introducing an

antenna-body distance (d) are the general guidelines of this

section.

1) Human Arm Models

Since the human body is a non-homogenous and multi-layer

medium, it is mandatory to study the dielectric properties of each

tissue comprised in the models presented above. Accurate

modeling of a real human arm requires the dispersion

characterization of each body tissue, on the operating frequency

range of the suggested antenna. In Figure 18, three four-layered

simplified models, and a homogenous arm model are shown.

7

Figure 18 - Human arm models: elliptical, rectangular, flat,

and homogenous (from left to right).

The three human equivalent arm models have four dispersive

tissue layers which are skin with 2 mm thickness, fat with 4 mm

thickness, muscle with 35 mm thickness and bone with 35 mm

thickness. The electric properties (relative permittivity and loss

tangent) of each human arm tissue were defined in the whole

UWB frequency range.

The proposed homogenous human arm model consists of a

single dispersive material with an electrical conductivity (σ) of

1 S/m and a relative permittivity (εr) of 42. The electrical

properties of this arm model were obtained from [8].

2) Return Loss and Operation Band

In order to investigate the impedance performance of the

antenna near the human body, simulations were carried out using

the four simplified human arm models previously introduced.

Furthermore, in a laboratory environment, the proposed antenna

was tested in the presence of a real human arm.

For a comparative analysis, the simulated |S11| curves using

the four models are compared with the measured on-body return

loss curves considering a separation gap of 3 and 6 mm. The

free-space return loss curve is also plotted as a reference for

human arm results.

a) Antenna placed 3 mm off the models (d = 3 mm)

Figure 19 - Simulated and measured |S11| curves in the

proximity of the human arm at d = 3 mm.

For a distance between the antenna and the skin layer equals

to 3 mm, the elliptical model results show that the lower -10 dB

band edge is shifted downwards from 2.9 to 2.5 GHz in relation

to the free-space curve. Therefore, the simulation for this human

arm model can exceptionally cover a much wider frequency

band starting from 2.5 GHz to 10.6 GHz and beyond. For all the

other adopted arm models the lower edge of the -10 dB operation

band is slightly shifted to the right. It is also observed that, when

the antenna-model space (d) is 3 mm, some accentuated

resonances appear in the return loss curves, e.g. 5.5 GHz and 9.2

GHz. Although reasonable good agreement between the

simulated and measured results was achieved, some

discrepancies between the results were detected.

a) Antenna placed 6 mm off the models (d = 6 mm)

Figure 20 - Simulated and measured |S11| curves in the

proximity of the human arm at d = 6 mm.

By increasing the distance between the UWB antenna and the

human arm models to d = 6 mm the frequency shift on the lower

-10 dB band edge diminishes. As was expected, the antenna

impedance bandwidth starts to widen due to the less influence of

the human body tissues. It can also be seen that the magnitude

of the resonances becomes lower as the antenna-model distance

increases. Just as the d = 3mm case, the measured results are in

accordance with the simulated results.

3) Total Efficiency

The simulated total efficiency of the antenna is also presented

assuming the four human arm models kept at distance d of 3 and

6 mm from the antenna. The results were obtained in CST

Microwave StudioTM for 2-12 GHz frequency range.

a) Antenna placed 3 mm off the models (d = 3 mm)

Figure 21 - Simulated antenna total efficiency for d= 3 mm.

8

As it apparent from the simulated results, the total antenna

efficiency is significantly deteriorated in close proximity of the

human body tissues. In this case (d = 3 mm), the simulated total

efficiency remains below 70% for the entire UWB frequency

range.

b) Antenna placed 6 mm off the models (d = 6 mm)

Figure 22 - Simulated antenna total efficiency for d= 6 mm.

When the antenna-body space is 6 mm, the presented UWB

monopole antenna still achieves a good total efficiency

(ranging within 70% and 90%) for almost the whole UWB

band. Comparing the results for each human arm model, it is

observed that the shape of the model has an insignificant

influence on the antenna total efficiency results. However, the

simulated results obtained for the homogenous model are

slightly different especially for the lower frequencies.

4) SAR (Specific Absortion Rate)

The whole evaluation of the SAR was performed in CST

Microwave Studio™ with the proposed compact UWB antenna

placed 3 mm off to the four-layered flat arm model. The flat

model was chosen for its simplicity and due to its simulation

time which is much shorter when compared with the other

suggested human arm models. The total mass of the chosen

model was then defined as 0.82 kg considering the density

values assigned to each tissue that comprises the arm model.

Since the quantification of SAR depends on the operating

frequency of the antenna, three frequency points (3.1 GHz, 6.05

GHz and 10 GHz) were chosen for this simulation over the entire

UWB band.

Furthermore, for 3.1-10.6 GHz UWB applications, the FCC

limits the spectral power density to -41.3 dBm/MHz, which

roughly corresponds to 8 x 10-5 mW of total radiated power per

MHz. Considering the antenna impedance bandwidth

approximately 9 GHz (3 GHz to 12 GHz), the input power is

normalized as 1 mW for all the SAR simulations.

Assuming the ICNIRP limit for the antenna SAR on human

limbs (4 W/kg) as a reference, only the maximum SAR results

for 10 g of tissue are considered for this analysis. The computed

values (Table 2) show that the maximum SAR of the proposed

UWB remains below this standard, for every case. When the

antenna is located 3 mm far from the model, the highest

maximum SAR value (10 g of tissue) occurs at 3.1 GHz

(0.0250931 W/kg). Thus, it can be noticed that, for a fixed

distance d, the maximum SAR of the antenna decreases as the

frequency increases.

V. CONCLUSIONS AND FUTURE WORK

A. Main Conclusions

In this paper a CPW-fed UWB antenna was totally designed

and developed. The antenna structure was simulated with

standard commercial paper and Kapton® as dielectric substrates

showing, in both cases, a -10 dB simulated impedance

bandwidth that totally covers the UWB frequency range. After

designing the optimized antenna geometry, two prototypes were

produced, through two distinct fabrication processes. After

testing these two prototypes in free-space, it was found that the

lower limit of the operation band was slightly shifted and thus

the measured return loss of prototype II (paper-based antenna)

does not comply |S11| < -10dB criterion for the entire UWB

operation band.

The free-space performance of the proposed UWB antenna

was analyzed and investigated by using numerical simulations

and measurements. The simulated results were divided in four

categories: the return loss curves (|S11|), operation bands,

efficiency (radiation and total) and radiation patterns (gain).

Otherwise, measurements concerned only the return loss,

operation bands and efficiency. Generally, the measured results

were in agreement with the simulated results in terms of return

loss and the UWB operation band was fully covered for both

simulated and measured |S11| curves.

Since one of the main goals of this work is to design a

flexible antenna to be integrated within flexible electronic

devices, bending deformations were also considered. Simulated

and measured results have shown good RF performance under

bending scenarios. However, for more sever bending conditions

the antenna parameters were strongly affected. It was noticed

from the results that vertical bending deformations have no

influence in the obtained results.

In order to evaluate the performance of the antenna under

crumpling scenarios, a cosine-type surface was built in CST

Microwave StudioTM. Three different crumpling amplitudes

were considered for the simulations (CA=1/2/4mm). In all

crumpling cases, the reflection coefficient and hence impedance

Table 2 - Simulated results for the Maximum SAR, Total SAR, and

Average Power for 1g and 10g of tissue considering a

distance between the antenna and the at arm model of 3

mm.

9

bandwidth of the antenna has changed. As the crumpling

amplitude increases, the more noticeable are the effects on the

antenna's performance. The antenna total efficiency under

crumpling was reduced with the increasing of the crumpling

amplitude.

Finally, the proposed UWB antenna design was measured

and simulated near the human body. Four human arm models

were fully characterized for this purpose. In order to understand

how the body tissues influence the antenna parameters,

simulations were carried out using these human arm models. In

a real usage scenario, the proposed antenna was tested in the

close proximity of a real arm. It was found that the human arm

proximity strongly influences the antenna impedance matching.

As was expected, the antenna total efficiency was also reduced

due to the absorption by the human arm tissues.

B. Future Work

Although the antenna design process was carefully studied

during this work, there are still some aspects about the substrate

material selection and fabrication methods that need to be

discussed in more detail. There are a large amount of new

available substrate materials, such as silicone based polymers

(PDMS and LCP), e-Textiles and microfluidics (fluid metal

alloy injected into microfluidic channels comprising a silicone

polymers) [9] that can be used as dielectric substrates for the

proposed antenna design.

Moreover, some new fabrication processes should be

adopted for future work. The 3D printing is a relatively recent

and high technological method, which allows to create complex

objects and cavities of various shapes and sizes from a digital

model [10]. This layer-by-layer approach to produce 3D printed

antennas can be proposed in a future study as a promising

solution to fabricate the proposed UWB antenna structure.

Furthermore, the inkjet printing process is also a good candidate

for the antenna's fabrication method due to its capability of

printing flexible circuits quickly and with a minimal cost. The

method used for the fabrication of the prototype II is still a

poorly developed technique. This proceeding allows the antenna

designers to cut out copper circuits from adhesive backed copper

foil using a vinyl cutter. The desired metallic patterns can be

attach into different substrate materials. Therefore, the research

on this area should also be carried out for future work.

As future work, the antenna design presented during this

work can be tested in other relevant scenarios, such as new

deformations (twisting), under the movement of the human body

and under different environmental conditions (humidity, heat).

REFERENCES

[1] M. Fransman, “Evolution of the telecommunications industry into the internet age," Jets Paper - University of Edinburgh Institute for Japanese Eurpean

Technology Studies, 2000.

[2] M. Koohestani, N. Pires, A. Skrivervik, and A. Moreira, “Performance study

of a uwb antenna in proximity to a human arm," Antennas and Wireless

Propagation Letters, IEEE, vol. 12, pp. 555-558, 2013.

[3] Roland, “Camm-1 servo gx-24 vinyl cutter." [Online]. Available:

http://www.rolanddga.com/products/cutters/gx24/.

[4] T. Hertel, “Cable-current effects of miniature uwb antennas," in Antennas

and Propagation Society International Symposium, 2005 IEEE, vol. 3A, July 2005, pp. 524-527 vol. 3A.

[5] N. Pires, “Small antennas as wireless system components," Ph.D. dissertation, L’Ecole Polytechnique Féderale de Lausanne and Instituto Superior

Técnico, August 2014.

[6] N. Pires, C. Mendes, M. Koohestani, A. Skrivervik, and A. Moreira, “Novel

approach to the measurement of ultrawideband antenna efficiency," Antennas

and Wireless Propagation Letters, IEEE, vol. 12, pp. 1512-1515, 2013.

[7] P. Hall and Y. Hao, “Antennas and Propagation for Body-centric Wireless Communications”, ser. Artech House antennas and propagation library. Artech

House, 2006.

[8] M. Rutschlin, “Body wearable antennas," CST EUROPEAN USER

Conference, April 2013.

[9] G. Hayes, J.-H. So, A. Qusba, M. Dickey, and G. Lazzi, “Flexible liquid metal

alloy (egain) microstrip patch antenna," Antennas and Propagation, IEEE

Transactions on, vol. 60, no. 5, pp. 2151-2156, May 2012.

[10] S. S. Bukhari and W. G. Whittow, “Flexible 3-d printed substrates for

antenna applications," Progress in Electromagnetic Research Symposium (PIERS), p. 1-5, 2013.