Available online at www.sciencedirect.

com

Sensors and Actuators A 141 (2008) 396–403

A wearable yarn-based piezo-resistive sensor

Ching-Tang Huang a,b , Chien-Lung Shen b,c , Chien-Fa Tang b , Shuo-Hung Chang a,d,∗
a Department of Mechanical Engineering, National Taiwan University, Taiwan
b Department of Product Development, TTRI, Taipei, Taiwan
c Institute of Biomedical Engineering National Yang-Ming University, Taiwan
d Mechanical and Systems Lab., ITRI, Hsin Chu, Taiwan

Received 28 June 2007; received in revised form 27 September 2007; accepted 14 October 2007
Available online 1 November 2007

Abstract
Smart textiles using fabric-based sensors to monitor gesture, posture or respiration have been exploited in many applications. Most of fabric-
based sensors were fabricated by either coating piezo-resistive materials on a fabric or directly knitting conductive fibers into fabrics. Obviously,
structures of textiles, including yarn structure and fabric structure, will affect the performances of sensors. However, researches on the effects of
the structures have not been explored yet. In this paper, yarn-based sensors were fabricated by using piezo-resistive fibers, elastic, and regular
polyester fibers. Single and double wrapping methods were employed to fabricate the yarn-based sensors. Performances of the designed yarn-based
sensors were evaluated by measuring their resistance changes under variable loading. It is shown that slippage occurs between the piezo-resistive
fibers and the core fibers. The relationship of the resistance versus the strain cannot be described as a linear function and should be modeled as a
second order equation. Due to the symmetric structure, the double wrapping yarn could resist the slippage and higher linearity in the resistance
curve can be provided. Thus it can be served as a better sensing element. The study also investigates the issue of the twist per meter (TPM) and
finds that there are no significant effects for using different TPM. Finally, experiments were conducted on a respiration monitoring system to prove
the feasibility of the yarn-based sensors and the results demonstrate that the yarn-based sensor can track the respiratory signals precisely.
© 2007 Elsevier B.V. All rights reserved.

Keywords: Smart textile; Yarn-based sensor; Piezoresistive

1. Introduction the physiological conditions of the wearer. Thus they could be

Smart textiles are becoming very popular in the past decade
[1]. One advanced study was to develop a wearable-computing
technique which integrated smart textures with semiconductors.
An intelligent textile was proposed to fabricate silicon flexible
skins with regular textiles [2]. Another application in semicon-
ductors was to form flexible transistors on textile fibers [3,4].
Although inspiring results have been reported in the past, prob-
lems such as complicated process, mass production, washability,
and wearing comforts are still under investigation.
The other study was to develop textiles which can detect
environmental conditions, and then react and adapt to environ-
mental changes. These smart textiles can measure and monitor
∗ Corresponding author at: Department of Mechanical Engineering, National
Taiwan University, Taiwan.
E-mail addresses: cthuang.0528@ttri.org.tw (C.-T. Huang),
Clshen.0865@ttri.org.tw (C.-L. Shen), cftang.0277@ttri.org.tw (C.-F. Tang),
shuochang@itri.org.tw, shchang@ntu.edu.tw (S.-H. Chang).
applied in healthcare systems. One type of the smart sensing
textures was developed by using fabrics having piezo-resistive
properties. The approach for fabricating the fabrics was to
coat a thin layer
 of piezo-resistive materials, such as polypyr-
role (PPy, a -electron conjugated conducting polymer) or
a mixture of rubbers and carbons, on conventional fabrics
to form fabric-based sensors [5–7,9,12]. The function of the
developed sensors is similar to that of flexible strain gauges
which can measure strains when they are subjected to a ten-
sile stress. Many applications based on this kind of sensing
fabrics were developed. One typical application was to cap-
ture posture or motion [6,11,13,14]. The others were related to
measuring biomechanical signals for healthcare, especially for
respiration detections [15–19]. The coated fabrics were highly
dependent upon knitting or weaving topology. Performances
can be quite limited if structures of fabrics and yarns were not
properly designed or optimized. Therefore, the piezo-resistive
sensing fabrics might have some shortcomings such as low
dynamic range, poor repeatability, performance deterioration

0924-4247/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.sna.2007.10.069
 C.-T. Huang et al. / Sensors and Actuators A 141 (2008) 396–403
after washing or repeated folding, and complicated manufac-
turing process.
Besides the development of the piezo-resistive sensing fab-
rics, another approach was to knit conductive fibers with
non-conductive base fibers [8]. The knitted fabrics sensors can
be regarded as equivalent circuits with a network of resistances,
capacitances, and inductances. As the deformation of the fab-
rics sensor occurs, the electrical properties of the elements will
be changed. The changes were measured to calculate the defor-
mation [8–10]. Although the integration of conductive fibers
improve the sensing performance even after washing, the sen-
sor can only be made by using knitting process and a large
knitting cloth is needed in order to obtain satisfactory results.
This requirement might limit the freedom in design of mod-
ern clothes. Furthermore, the fabric-based sensor is considered
only as two-dimensions because it is made by plane knitting.
Space resolution is not high and the obtained information from
the average area change could be quite limited. Moreover, the
research issues such as yarn topology and structural deformation
of fabrics-based sensors have not been investigated yet.
In this paper, yarn-based sensors were developed to improve
sensing characteristics [20]. Instead of using fabrics as base
elements, we first fabricated the yarns by using piezo-resistive
fibers, elastic, and regular polyester fibers. Then the yarn was
used as raw materials to make cloth, dress and sensing tex-
tiles. As compared to fabric-based sensors, the yarn itself is
a sensing element and thus it is easier to be used in smart tex-
tiles by conventional knitting or weaving processes. Dependent
upon different applications, several sensing segments can be
embedded into textiles such that distributed strains can be mea-
sured with the sensors. This approach has the advantages such
as higher space resolutions, more comforts, better functionality
and easier in style design. For fabrication process, single and
double wrapping methods were employed. Experiments were
performed to measure the resistance changes of the yarn under
variable loading. The linearity of the single and double wrap-
ping yarn sensors was evaluated. It is found that the double
wrapping method can achieve higher linearity than the single
wrapping approach. Physical interpretations are given to illus-
trate this phenomenon. Furthermore, different twists per meter
(TPM) of carbon coated fibers (CCF) wrapping on the core yarn
was also investigated. No significant effects on different TPM
were found in the experiments. Finally, a respiration monitoring
system was used as a test bed to prove the feasibility of the yarn-
based sensors and the results demonstrate that the yarn-based
sensor can track the respiratory signals precisely.
2. Materials and methods of the yarn-based sensors
Yarns are normally considered as the basic elements of form-
ing fabrics and textiles. But fibers are truly the raw elements
of yarns. Yarns are made by combining different types of fibers
into a skein. Fiber materials and forming process can have dra-
matically effects on the characteristics of the yarn. The process
of forming a yarn can be categorized into three main methods.
One simplest method is the doubling which the fibers are put
in parallel to form the yarn. The fibers are bonded together by
397
Fig. 1. The yarn structure of yarn-based sensors: (a) single wrapping and (b)
double wrapping.
mutual friction forces. As the yarn is subjected to tensile stress,
the fibers can overcome friction force such that mutual slip-
page can occur. Another method is the entangling approach.
This approach is to use pressured air to entangle fibers in fabri-
cating the yarn. The structure of the yarn by this method is not
regular and the uniformity of the yarn might be a problem when
used as a sensing device. The other method is the wrapping pro-
cess. This method is to helically wrap one or two fibers on the
core yarn. The approach of wrapping one fiber on a core yarn
is called the single wrapping. Another approach is to wrap one
fiber in clockwise direction and the other fiber in counterclock-
wise direction. This approach is called the double wrapping.
Normally, the twist per meter (TPM) is used to describe the
twist density of the wrapping.
In this paper, the wrapping process was adopted to fabri-
cate the yarn-based sensor as shown in Fig. 1. Three different
kinds of fibers which include piezo-resistive fibers, elastic fibers,
and polyester were used to form the yarn. The piezo-resistive
fiber was chosen as the carbon-coated fiber (CCF) (resistivity
3 × 105 /cm, RESISTAT F901). The fabrication process was
to integrate the polyester fibers with elastic fibers into a compos-
ite core yarn at first. The length of the elastic fiber was stretched
three times in the process. Then the CCF was wrapped on the
core yarn by using the single and double wrapping approaches.
The motivations of integrating three kinds of fibers were given
as follows. In forming the fabrics, the yarns should have the simi-
lar linear density and the strength capability must be high enough
for weaving or knitting process. Since the linear density of the
CCF is only 24 dtex, where dtex equals to gram per 10,000 m, the
CCF should be combined with other fibers to increase linear den-
sity and force capability. In this paper, a 56 dtex/48 f polyester
filament was chosen to integrate with the CCF. The 48 f means
that 48 filaments are contained in a 56 dtex skein. The adding of
the elastic fibers (Lycra, 22 dtex) could improve the elasticity of
the sensing yarn. The linear densities of the yarns using the sin-
gle and double wrapping approaches are approximately within
90–110 dtex which are close to regular yarns.
To evaluate the effects of the twist, samples were fabricated
with TPM equal to 150, 275 and 450. The 450 TPM means that
the CCF wraps the core yarn 450 times per meter. Six differ-
ent types of samples were fabricated and listed in Table 1. In
the next section, experiments were performed to evaluate the
performances of the samples.
3. Experimental results of the yarn-based sensors
To evaluate the sensing behaviors of the yarn, experiments on
resistance changes under different loadings were conducted in
398 C.-T. Huang et al. / Sensors and Actuators A 141 (2008) 396–403

this section. The Mini44 INSTRON was utilized for measuring

6.949x2 + 2.564x + 0.7

tensile forces and Fluke189 multi-meter was used to measure

4.186x + 0.6415
the resistances. The samples were selected to be 6 cm long with
the preload equal to 20 g. The force and the resistance were mea-

Double
sured as the length of the sample was stretched per mm. The total

0.9882

0.9993
D450

450
stretched length was 14 mm and 15 points were taken. To evalu-
ate the uniformity of the yarn sensor, five samples were prepared
for each design listed in Table 1. The plots on the resistances

6.425x2 + 3.445x + 0.639

and forces versus the strains (L/L) were shown in Figs. 2–4.
4.944x + 0.5855 The curves were fitted by the first and second order regression
functions, and the coefficients of determination (R square) were
calculated. The statistical data were listed in Table 1.
Fig. 2 shows the curves and the fitting equations for the CCF
Double

0.9917

0.9985
D275

fiber alone. For the resistance curve, the R squares for the first
275

and second regression models are equal to 0.983 and 0.985,

respectively. Linear relation is adequate in describing the rela-
9.152x2 + 3.592x + 0.632

tionship between the resistance and the strain. The R squares

for the two models are very close. However, the second order
regression model provides higher accurate fitting for the force
5.7271x + 0.555

curve.
Figs. 3 and 4 show the resistance and force curves for the
Double

0.9878

0.9981

single and double wrapping yarns, respectively. As compared to

D150

150

the force curves in Figs. 2–4, the forces that the single and double
wrapping yarns can sustain under L/L equal to 23% (14 mm
stretched) were about 0.3, and 0.4 kg, respectively. Under the
55.628x2 + 4.263x + 1.719

same strain, the force that the CCF alone can sustain was only
0.09 kg. The wrapping of the CCF with the elastic and polyester
17.242x + 1.249

fibers increases the force capability.

There are two interesting phenomenon observed in Fig. 3.
The first is that the resistance curve for the single wrapping yarn
0.9586

0.9993
Single
S450

should be modeled as a second order upward curve equation.

450

The second is that the curves for different twists (TPM equal to
150, 275 and 450) have no significant effects on the character-
43.908x2 + 4.060x + 1.821

istics of the yarns. The phenomenon can be illustrated by the

CCD pictures of the yarns under different elongations (Fig. 5).
As the elongation equals to 0, the CCF was wrapping around
14.304x + 1.450

the core yarn. As the elongation increases to 5 mm, the struc-

ture changes and one cannot exactly define which one is the
0.9622

0.9992

core yarn. The resistance of the yarn under the strain between
Single
S275

275

0 and 7% increases slowly as shown in Fig. 3. It is observed

that the geometric shape of the CCF changes from the curve
to the straight to avoid elongation when the applied strain is
56.218x2 + 3.425x + 1.686

small. At this stage, the stress is mostly taken by the core yarn.
As the elongation increases further, the CCF becomes straight
and it cannot resist the elongation by structural change. The
16.541x +1.212

CCF takes the strain directly and the resistance changes become
large at this stage. The observation indicates that the friction
0.9547

0.9997
Single

(constraint) force between the CCF and the core yarn in the sin-
S150
Type

150

gle wrapping approach cannot prevent the relative motion. The

The statistical data of the yarns

slip motion caused the non-linear behavior as shown in Fig. 3

and this explains why the data is better fitted by a second order
Second order regression
First order regression

regression. Furthermore, the slip phenomenon occurs for the

samples with different TPM, and thus the effects of the TPM are
Wrap structure

not obvious as shown in Fig. 3.

Fig. 6 shows the CCD pictures of the double wrapping yarns
Table 1

under different elongations. Due to the symmetric structure, the

TPM

R2

R2

lateral forces of the two CCFs are balanced with each other. The
 C.-T. Huang et al. / Sensors and Actuators A 141 (2008) 396–403 399

Fig. 2. The elongation graph of 24 dtex CCF.

core yarn remains in the center. There is no obvious structural
change as compared to the single wrapping yarn. Without the
slippage, the double wrapping approach can provide higher lin-
earity as shown in Fig. 4. Reflecting to the experimental results,
the R square of the linear regression model of the double wrap-
ping yarn is higher than that of the single wrapping yarn. To
further illustrate the merits of the double wrapping approach,
the coefficient of variations (CV) are listed in Table 2. The CV
was calculated by the standard deviation divided by the mean
value for every five samples at a specific elongation. It is shown
that the variations of the samples fabricated by the double wrap-
ping approach are smaller than those by the single wrapping
approach.
Although the double wrapping yarn can achieve higher lin-
earity, its sensitivity is lower than that of the CCF alone. Because
two CCFs were wrapped around the core yarn, the parallel con-
figuration of two conductive fibers makes the resistance to be
approximate half of the CCF alone. The sensitivity (slope) of

Fig. 3. The elongation graph of single wrapping yarns.

400 C.-T. Huang et al. / Sensors and Actuators A 141 (2008) 396–403

Fig. 4. The elongation graph of double wrapping yarns.

respectively. To illustrate this phenomenon, the shape changes

Fig. 5. The pictures of the single wrapping yarns under different elongations.
the CCF alone is equal to 18.601 as shown in Fig. 2. Theoreti-
cally, the sensitivity of the double wrapping approach is about
9.3. However, taking the D450 as an example, the slope of the
first regression model is 4.186 at 14 mm elongation which is
much less than 9.3. The slopes of the samples for TPM equal
to 275, and 150 at 14 mm elongation are 4.9496, and 5.7271,
shown in Fig. 6 are examined. As the tensile stress is applied on
the double wrapping yarn, both the longitudinal force and the
lateral force should be considered. The longitudinal force causes
the elongation, but the lateral force compresses the core yarn to
change the shape. The shape change alleviates the deformation
of the CCFs. Therefore, the sensitivity of the double wrapping
yarn is less than 9.3. Higher TPM results in higher lateral forces
and thus the shape change under the same applied strain was
larger. It explains why the sample with higher TPM has lower
slope. Note that the shape change should be also dependent upon
the mechanical properties of the core yarn.
In next section, the sample D450, is used as the base ele-
ments to build a monitoring respiration system. It was shown
that the breathing signal can be measured even utilizing the low
sensitivity design of D450.
4. Experimental test on a respiration monitoring system
In order to verify the feasibility of the yarn-based sensor,
an experimental system for monitoring respiration signal was

Table 2
The coefficient of variations (CV) at some specific elongations
Sample Elongation

2 mm 5 mm 8 mm 11 mm 14 mm

S150 (%) 3.60 3.92 3.18 2.12 0.84

S275 (%) 2.96 3.79 5.16 2.29 1.98
S450 (%) 2.79 4.09 3.93 3.02 3.68
D150 (%) 1.59 2.40 1.49 0.97 0.99
D275 (%) 1.55 0.71 1.48 2.41 0.90
D450 (%) 1.55 1.09 0.82 1.18 1.67
Fig. 6. The pictures of the double wrapping yarns under different elongations.
 C.-T. Huang et al. / Sensors and Actuators A 141 (2008) 396–403 401

Fig. 7. The tested sample.

developed in this section. For simplicity, a Crochet knitting

machine was used to make elastic bands, which were combined
with several commercial polyester yarns (333 dtex), rubber yarns
( 0.5 mm) and two sensing yarns. The function of the rubber
yarns is to provide the flexibility. The length of the elastic band
is about 5 cm. Two elastic bands were sewed on a garment to
form a sample. Two sizes of samples were designed for kids and
adults. Although the sample A for kids is shorter than the sam-
ple B for adults, the lengths of the elastic bands for two samples
are the same. The two sensing yarns were parallel connected,
and a bridge amplifier was used to acquire the measurements
as shown in Fig. 7. In fact, the four sensing yarns can be used
separately such that the distributed deformation of the textiles
can be measured.
Fig. 8. The experimental system:(a) bridge sensing circuit and (b) the block
The experimental system is shown in Fig. 8, where a diagram of experimental system.
NPB7200 ventilator was used to supply air to an artificial test
lung periodically to simulate the breathing. The baby garment
with the yarn-based sensor was clad on the artificial lung. The The breath per minute (BPM) for adults is about 10–20 BPM
change of air pressure and volume in the artificial lung causes in normal condition, in case of tachypnea, the BPM of adults
the deformation of the sensors and the resistance changes were might be higher. The BPM for children can be higher than 30
measured by the bridge sensing circuit, as shown in Fig. 8(a). and lower BPM or apnea are dangerous for children. It means
After bridge circuit, a TI instrumentation amplifier, INA326, the detection of 10–30 BPM is important for both adults and
was connected to amplify the raw signal, which has high CMRR children. Due to the reasons, the BPM used for experiments
(114 dB) to reject the common mode interference and high PSR were chosen as 10 and 30. The experimental results were shown
(␮V/V = 3) to reject the power supply interference. And then, in Fig. 9. It is clear that both samples A and B can measure
the NI PCI-6024 A/D card installed in the PC was utilized to the number of the breath accurately. The BPM can be calcu-
sample two analog signals simultaneously with a 250 Hz sam- lated by measuring the time at the first notch and the time at the
pling frequency. The first is the original air pressure monitored final notch. The BPM was then calculated by the time difference
by Beacon GM350 and the pressure is transformed into an ana- divided by the number of the pulse. The errors for each measure-
log signal, which is denoted as standard signal in Fig. 8(b). The ment were listed in Table 3. It is shown that the developed yarn
second is the signal come from the sensor. sensor can measure the BPM accurately.

Table 3
The experimental results for measuring the BPM using the yarn-based sensors
sample BPM The first notch time (s) The last notch time (s) Periods BPM Error (%)

A 10 2.6 56.49 9 10.02 0.2

30 2.196 28.124 13 30.08 0.26
B 10 0.7 54.58 9 10.02 0.2
30 0.92 28.784 14 30.15 0.5
402 C.-T. Huang et al. / Sensors and Actuators A 141 (2008) 396–403
Fig. 9. The results of experimental system.
5. Conclusions
In this paper, an innovative yarn-based sensor was devel-
oped and its functionality was verified by using a respiration
monitoring system. Yarn-based sensors were fabricated by using
piezo-resistive fibers, elastic, and regular polyester fibers. Sin-
gle and double wrapping methods were utilized to fabricate the
yarn-based sensor. It is shown that the relationship between the
resistance and the strain of the single CCF fiber can be described
as a linear function. However, the resistance curve for a single
wrapping yarn should be modeled as a second equation because
the slippage occurs between the CCF and the core yarn. Due to
the symmetric structure, the double wrapping yarn could provide
higher linearity in the resistance curve and thus can be served
as a better sensing element. It was also found that the sensitivity
of the double wrapping yarn is lower because the lateral force
might compress the core yarn such that the shape of the core
yarn is changed to alleviate the deformation of the CCFs. This
study also investigated the issue of the twist per meter and found
that there are no significant effects on different TPM. Finally,
experiments were conducted on a respiration monitoring sys-
tem. The results show that the yarn-based sensor can track the
respiratory signals precisely which validate the effectiveness of
the yarn-based sensors.
References
[1] J. Farringdon, A.J. Moore, N. Tilbury, J. Church, P.D. Biemond, Wear-
able sensor badge & sensor jacket for context awareness, Philips Research
Laboratories, UK, IEEE ISWC’99 (1999) 107–113.
[2] R.B. Katragadda, Y Xu, A Novel Intelligent Textile Technology
Based on Silicon Flexible Skins, Electrical & Computer Engineering,
Wayne State University, Detroit, Michigan, USA, ISWC (2005) 78–
81.
[3] J.B. Lee, V. Subramanian, Weave patterned organic transistors on fiber for
E-textiles, IEEE Trans. Electron Devices 62 (2005) 269.
[4] A. Bonfiglio, D. De Rossi, T. Kirstein, I. Locher, F. Mameli, R. Paradiso,
G. Vozzi, A feasibility study of yarns and fibers with annexed electronic
functions: the ARIANNE project, in: A. Lymberis, D. de Rossi (Eds.),
Wearable Health Systems for Personalised Health Management, IOS Press,
Amsterdam, The Netherlands, 2004, pp. 324–329.
[5] D. De Rossi, F. Carpi, F. Lorussi, A. Mazzoldi, E.P. Scilingo, A. Tognetti,
Electroactive fabrics for distributed, conformable and interactive systems,
IEEE Sens. (2002).
[6] F. Lorussi, E.P. Scilingo, M. Tesconi, A. Tognetti, D. De Rossi, Wearable
sensing garment for posture detection, rehabilitation and tele-medicine, in:
Proceedings of the 4th International IEEE EMBS Special Topic Conference
 C.-T. Huang et al. / Sensors and Actuators A 141 (2008) 396–403
on Information Technology Applications in Biomedicine, Birmingham,
UK, April 24–26, 2003, pp. 28–290.
[7] R. Wijesiriwardana, T. Dias, S. Mukhopadhyay, K. Mitcham, Knitted strain
gauges, in: Presented at the SPIE Int. Symp. Microtechnologies for the
New Millennium Maspalomas, Canary Islands, Spain, May 19, 2003, pp.
5116–5131.
[8] R. Wijesiriwardana, T. Dias, S. Mukhopadhyay, Resistive fiber-meshed
transducers, in: Proceedings of the IEEE Int. Symp. on Wearable Comput-
ing, October 2003, pp. 200–205.
[9] R. Wijesiriwardana, K. Mitcham, T.Dias, Fibre-meshed transducers based
real time wearable physiological information monitoring system, Proceed-
ings of the 8th International Symposium on Wearable Computers, ISWC’04
(2004) pp. 40–47.
[10] R. Wijesiriwardana, Inductive fiber-meshed strain and displacement trans-
ducers for respiratory measuring systems and motion capturing systems,
IEEE Sens. J. 6 (3) (2006) 571–579.
[11] E.P. Scilingo, F. Lorussi, A. Mazzoldi, D. De Rossi, Strain-sensing fabrics
for wearable kinaesthetic-like systems, Sens. J. IEEE 3 (4) (2003) 460–467.
[12] D. De Rossi, F. Carpi, F. Lorussi, R. Paradiso, E.P. Scilingo, A. Tognetti,
Electroactive fabrics and wearable man-machine interfaces, in: Wearable
Electronics and Photonics, Woodhead Publishing Limited, UK, 2005.
[13] D. De Rossi, F. Lorussi, A. Mazzoldi, P. Orsini, E.P. Scilingo, Monitor body
kinematics and gesture through sensing fabrics, in: Proceedings of the 1st
Annual International Conference On Microtechnologies in Medicine and
Biology, Centro Interdipartmentale di Ricerca, Pisa University, 2000, pp.
587–592.
[14] F. Lorussi, W. Rocchia, E.P. Scilingo, A. Tognetti, D. De Rossi, Wearable,
redundant fabric-based sensor arrays for reconstruction of body segment
posture, IEEE Sens. J. 4 (6) (2004) 807–818.
[15] D. De Rossi, F. Carpi, F. Lorussi, A. Mazzoldi, R. Paradiso, E.P. Scilingo, A.
Tognetti, Electroactive fabrics and wearable biomonitoring devices, Autex
Res. J. 3 (4) (2003) 180–185.
[16] G. Loriga, N. Taccini, D. De Rossi, R. Paradiso, Textile sensing interfaces
for cardiopulmonary signs monitoring, in: Proceedings of the 27th Annual
Conference on Engineering in Medicine and Biology, IEEE, 2005.
[17] R. Paradiso, G. Loriga, N. Taccini, A wearable health care system based on
knitted integrated sensors, IEEE Trans. Inf. Technol. Biomed. 9 (3) (2005)
337–344.
[18] R. Paradiso, G. Loriga, N. Taccini, A. Gemignani, B. Ghelarducci,
WEALTHY—a wearable healthcare system: new frontier on E-textile, J.
Telecommun. Inf. Technol. 4 (2005) (Poland).
[19] E.P. Scilingo, A. Gemignani, R. Paradiso, N. Taccini, B. Ghelarducci, D.
De Rossi, Performance evaluation of sensing fabrics for monitoring phys-
iological and biomechanical variables, IEEE Trans. Inf. Technol. Biomed.
9 (3) (2005) 345–352.
[20] C.-T. Huang, C.-F. Tang, C.-L. Shen, A Wearable Textile for Monitoring
Respiration, Using a Yarn-Based Sensor, ISWC’06 (2006), pp. 141–142.
403
Biographies
Ching-Tang Huang, graduated from National Taiwan University in 1988, and
received the M.S. degrees in Mechanical Engineering from National Taiwan
University in 1993. Now, he is a Ph.D. student in Mechanical Engineering,
National Taiwan University. Simultaneously, he is also the chief of system and
information section, Taiwan Textile Research Institute. The major research topic
is about the smart textile, focusing on how to develop sensors and actuators,
which are tiny and could be embedded in the textile, or the textiles are sen-
sors or actuators themselves, and how to implement the smart clothing with
miscellaneous sensors and actuators on the basis of maintaining the traditional
comfort.
Chien-Lung Shen, was born in Tainan, Taiwan in 1976. He received his M.S.
degree in Department of Automatic Control Engineering at Feng Chia Univer-
sity, Taiwan in 2000. From 2001, he has been joining Department of Product
Development, Taiwan Textile Research Institute as a researcher. From 2005,
he has been joining Institute of Biomedical Engineering System Engineering
at National Yang-Ming University as a candidate for doctor’s degree. His main
research interests are electronic and electric circuits design, computer program-
ming, biomedical and sensor applications.
Chien-Fa Tang, received the M.Sc. degree in textiles from the UMIST (Univer-
sity of Manchester Institute of Science and Technology), UK, in 1997. He has
worked on the TTRI since 1990. His research interests include electrical inter-
face, biomedical signal processing, digital/analog circuit design, and system
control.
Shuo-Hung Chang, received the B.S. degree in 1974 from National Chen
Kung University, Taiwan and the M.S. and Ph.D. degrees in 1981 and 1985
from the University of Cincinnati. From 1984 to 1990 he worked with
the IBM T.J. Watson Research Center in Yorktown Heights, NY, involved
in advanced computer peripheral devices, such as printer, data storage and
information display. Since 1990, he has been at the National Taiwan Uni-
versity, where he is a professor in Mechanical Engineering Department. He
is the Director of the Nano-Electro-Mechanical Systems (NEMS) Research
Center and the Deputy Director of the Center for Nano-Science and Tech-
nology. He was a visiting scholar with National Institute of Standards
and Technology (NIST) and the Stanford University in 2000 and 2003,
respectively. His research interests involve electro-elasticity theory and mod-
eling, ultrasonics, sensors and actuators, nanometer positioning and carbon
nanotubes.
Dr. Chang is a member of the IEEE, American Society of Mechani-
cal Engineers (ASME), Chinese Society of Mechanical Engineers (CSME),
Chinese society of Mechanism and Machine Theory, and a member
of the executive committee of the Nanotechnology and Micro-System
Association.