J. Biomedical Science and Engineering, 2010, 3, 822-827 JBiSE
doi:10.4236/jbise.2010.38110 Published Online August 2010 (http://www.SciRP.org/journal/jbise/).
Published Online August 2010 in SciRes. http:// www. scirp. org/journal/jbise
Tele-care for emergency announcements
Ching-Sung Wang1, Chien-Wei Liu 2, Teng-Wei Wang3
1Department of Electronic Enginee ring, Oriental Institute of Technology, Taipei, Taiwan, China;
2Department of Information Management, St. Mary’s College, Yilan, Taiwan, China;
3The Third Department of Clinical Research Institute, Peking University, Beijing, China.
Received 5 June 2010; revised 21 June 2010; accepted 22 June 2010.
ABSTRACT
The first aid and immediately help are very impor-
tant in an accident. The earlier detection and treat-
ment we do, the better prognosis and health patients
have. In the senior populations, it is more important.
Once seniors have an accident, not only physically
injure in their body, but also mental and social ability,
and may have severe sequela. Concerning about these
populations, this research design a simple, practical,
and portable device of real-time monitoring body
activity with sphygmomanometer and pulsimeter.
When an accident occurs, the signals go through mo-
bile phone, immediately notify the remote ends and
provide first time help.
Keywords: Real-Time; Monitoring; Mobile;
Sphygmomanometer; Pulsimeter
1. INTRODUCTION
In 1991, American Heart Association suggests “The
Chain of Survival” -Early Access, Early CPR, Early
Defibrillation, and Early Advanced Care. It emphasizes
that immediately after cardiac arrest, early CPR in the
first 4 min. and early advance care in the first 8 min
would have about 43% success rescue rate, on the con-
trary, lower than 20%. Cumnius, R.O.’s research points
out that there is about 7-10% success rescue rate de-
creasing per minute delayed after patients’ heart arrest
[1]. In traumatic care, there is so called “golden hour”
concept that means the first sixty minutes after the oc-
currence of multi-system trauma. It is widely believed
that the victim’s chances of survival are greatest if they
receive definitive care in the operating room within the
first hour after a severe injury (Committee on Trauma,
1993; Division of Trauma & EMS, 1992). For the cere-
brovascular accident, it also points out “golden 3 hours”,
giving a proper deal in the first 3 hours after the accident
occurring can increase the chance of survival. Whether
what kind of emergency, they focus on immediately re-
sponse and properly deal, early detection and early
treatment.
Accident is more important for the seniors, morbid
populations and for those who have unstable life signs.
According to studies, once seniors fall down, they would
reduce their living activity [2,3], indirectly leveling
down their life quality.
Based on people’s habits, using mobile phone, this
research suggests a device, real timely monitoring hu-
man body’s activity with sphygmomanometer and pul-
simeter, combining with mobile and Bluetooth etc. rela-
tive technology to edit a simple, practical, and suit for
the current mobile phone that includes Bluetooth device.
When the accident occurs, it can immediately and effec-
tively provide relative rescuing information to remote
GPS etc, expecting that is able to improve first aid effect.
And, it provides senior population an effective prevent-
ing accident guarantee, enabling them to avoid decreas-
ing their life activity caused by an accident.
2. SYSTEM ARCHITECTURE
This system divides into two parts, which are client side
and server side respectively. Client side comprises BAD
(Body Activity Detector) with sphygmomanometer and
pulsimeter, GPS (Global Position System) device and
mobile phone. Once BAD detected any unusual body
responses, mobile phone would immediately delivers the
emergent message by DTMF (Dual Tone Multi Fre-
quency) coding with sphygmomanometer, pulsimeter
and the GPS coordinates to the monitoring server [4].
The server, which includes PSTN (Public Switched
Telephone Network) phone controller and a monitoring
terminal, that is responsible for monito ring the client and
emergency management. Figure 1 shows the system
architecture.
2.1. Client Side System
As considered the mobility, BAD with sphygmoma-
nometer, pulsimeter and GPS devices are separated in
client side device, yet information delivery by Bluetooth.

C.-S. Wang et al. / J. Biomedical Science and Engineering 3 (2010) 822-827 823
Figure 1. System architecture (P: pulsimeter; S: sphygmo-
manometer).
Details of BAD with sphygmomanometer, pulsimeter
and GPS devices are mentioned in the following:
1) BAD: The main function is to monitoring the un-
usual body activity, and communicates with mobile
phone by Bluetooth HFP (Hands-Free Profile) [5]. When
BAD detects any unusual body responses, the emergent
signals and the GPS coordinates information will be
delivered to the mobile phone, and they simultaneously
transport the signals to the cell phone by Pulsimeter and
Sphygmomanometer, and the mobile phone sends such
message to the monitoring server through the public
mobile network immediately.
2) GPS Device: receiving GPS signals. GPS device
could deliver GPS signals to BAD by Bluetooth SPP
(Serial Port Profile) [6] every 5 seconds to keep BAD
updated the latest position information.
3) Pulsimeter and Sphygmomanometer: when Pul-
simeter and Sphygmomanometer receive any urgency
messages from BAD, they immediately measure the
urgency messages and transport the results to the cell
phone by the BAD of Bluetooth. Then Pulsimeter and
Sphygmomanometer transport them every 10 seconds.
2.2. Server Side System
Server side is constructed by PSTN phone controller and
the monitoring terminal. When server side PSTN phone
controller receives the incoming call from client side
mobile phone, it will be fo rced to conn ect and output th e
sound to the monitoring terminal. The monitoring ter-
minal reads the DTMF code and once the terminal de-
termines the emergency signals are correct, it would
send out the emergen cy tone and pop out the emergency
screen.
3. HARDWARE ARCHITECTURE
SHORT-LISTED
3.1. Client Side Hardware Architecture
Figure 2 details the block diagram of BAD hardware
architecture. BAD is mainly constructed by a micro-
processor, an accelerometer, Bluetooth transceiver, voice
Figure 2. BAD hardware architecture.
device and other accessories. Its functions are detection
of the body reaction signals, anomaly analysis, and the
control of signal transfer and receive. Explanations of
each device are listed below:
1) Microprocessor: SC14431 [7] is used as the con-
troller of BAD in this research. SC14431contains a
RISC (Reduced Instruction Set Computer) micro con-
troller, DSP (Digital Signal Processor), and PCM (Pulse
Code Modulation) codec. Micro controller is used for
operation of Bluetooth protocol stack, and analysis of
the signals of accelerometer. DSP and PCM codec are
used for sound signal operation and decoding. SC14431
is made by SiTel.
2) Accelerometer: MXA2500 Dual Axis Accelerome-
ter [8] is used as the accelerometer is this research.
MXA2500 is an electro-mechanical integrate device
which can be used to detect the change of acceleration.
The change of acceleration is a control point of the body
condition in this research. MXA2500 communicates
with the microprocessor through the A/D (Analog/Digi-
tal) interface. MXA2500 is made by MEMSIC.
3) Bluetooth Transceiver: BC03 [9] is used as the
baseband IC of Bluetooth transceiver, communicates
with the microprocessor through UART and PCM inter-
face. BC03 is made by CSR.
Figure 3 details the block diagram of GPS device
hardware architecture. The hardware structure of GPS
device is similar to BAD, which contains the micro-
processor, GPS module, Bluetooth transceiver and the
Figure 3. GPS device hardware architecture.

824 C.-S. Wang et al. / J. Biomedical Science and Engineering 3 (2010) 822-827
accessories. Microprocessor is used for the operation of
Bluetooth protocol stack and receiving of the informa-
tion of GPS module. Start2 GRF2I [10] is used as the
GPS module RF IC and Start2 GSP2e is used as the GPS
module baseband IC, which deliver the data to the mi-
croprocessor through UART interface. Start2 GRF2i and
GSP2e are made by SiRF.
Sphygmomanometer [11-14]: The sphygmomanometer
includes LDO (Low Dropout Regulators) as power con-
troller, pressure sensor, filter, amplifier circuit and ac-
cessories. The research is using the pressure sensor of
SCC series which is manufactured by HoneyWell com-
pany. The pressure sensor will transmit different message
according to the changing pressure. The message passes
through the procedures of magnifying (magnify the mi-
cro-message form sensor), and filtering (remove noise),
then all the analog messages will be sent to the micro-
controller via A/D (Analog/Digital) interface. Figure 4
shows the block diagram of sphygmomanometer hard-
ware architecture.
Pulsimeter [11-14]: The Pulsimeter includes LDO,
pressure sensor, filter, amplifier circuit and accessories.
The circuit is using the SCC series of HonyWell com-
pany, but it is a different sen sitivity pressure sensor. Put-
ting sensor pad on pulse of hand, the pressure sensor will
convert the normal pulse beat to a larger voltage output
by magnifying, fillering, comparing (transform to digital
signal) then inputs it into microcontroller. Figure 5 is
shown the block diagram of Pul s imeter.
3.2. Server Side Hardware Architecture
Figure 6 details the block diagram of server side hard-
ware Architecture. Server side includes a PSTN phone
controller and a m oni t ori ng t erminal. Furtherm ore, PSTN
Figure 4. Sphygmomanometer hardware architecture.
Figure 5. Pulsimeter hardware architecture.
Figure 6. Server side hardware architecture.
phone controller is constructed by ATA (Analog Tele-
phone Adapter), microcontroller, and the accessories,
which takes the incoming call from the client. W681388
[15] is used as the baseband IC of ATA for ring detection
and coding of signals of voice. PIC24FJ64 [16] is used as
microcontroller, which receives the command of the
monitoring terminal through the UART interface and con-
trols W681388 Hook ON/OFF and voice output through
SPI(Serial Peripheral Interface) interface. W681388 is
made by Winbond. PIC24FJ64 is made by Microchip.
4. WIRELESS PERSONAL AREA
NETWORK BASE ON BLUETOOTH
Bluetooth provides point to point and multipoint wire-
less connection according to the in ternet concept. W ithin
any active communication scope, any devices are treated
the same. The first one requesting communication is
called master, and the passive one to accept signal is
called slave. A master and one or more salves construct
the Piconet of Bluetooth [17,18]. Due to the reason that
not all the mobile phone with Bluetooth function in the
current market support Serial Port Profile, the GPS de-
vice used in this research cannot connect with all the
mobile phone with Bluetooth via Bluetooth function. In
order to cover most of the mobile phone with Bluetooth
in the market in this research and make an active Piconet,
we use the Bluetooth device of BAD as master, and the
Bluetooth devices of mobile phone and GPS as slave.
The Bluetooth of BAD connects with the Bluetooth of
mobile phone and GPS and form a PAN (Personal Area
Network) upon Bluetooth. Figure 7 indicates the Blue-
tooth Piconet in this research.
5. SINGAL PROCESSING
BAD in this research contains two different modes ac-
cording to different body conditions, which are body
stimulation mode and body activity detection mode re-
spectively. Here are the details below:
5.1. Body Stress-Reaction Mode Signal Processing
In this mode, it means the system detect the body condi-
tion has changed from a low activity level to a continu-
ing high consumption level and it represents there is
something happened [19]. Then BAD turns into the

C.-S. Wang et al. / J. Biomedical Science and Engineering 3 (2010) 822-827 825
Figure 7. Client side bluetooth piconet (P: pulsimeter;
S: sphygmo-manometer).
emergency processes right away. For example if the user
falls down or breaks out epilepsy, the change of energy
reaction would trigger the emergency task. Figure 8
indicates the sample accelerometer signal in the low ac-
tivity level of body. Figure 9 indicates sample of accel-
Figure 8. Sample of accelerometer signal in the low activity
level of body.
Figure 9. The sample accelerometer signal in the high activity
level of body.
erometer signal in the high activity level of body. Figure
10 indicates the body stress-reaction mode procedure
flow chat.
Microprocessor reads the signal from accelerometer
every 10 milliseconds, and pass the signal through one
order passive HPF (High Pass Filter). Refer to Eq.1 [19],
and then calculate the RMS (Root Mean Square). Refer to
Eq.2, if the RMS is upper th an the preset dynamic RMS
threshold for a period of time, it means that emergency
has occurred. The system would beep out the warning
sound in the first place and if the user does not turn off
the alarm, the system would start emergency procedure.


()( 1)( 1)
() R CXnXnYn Yn TRC
 



(1)
HPF is used to screen out the signals over 1HZ. While
the signal is over 1HZ it means it is not the system tar-
geted signals. It perhaps just comes from the shaking or
noises of the devices. 1HZ HPF RC = 0.16, T = 0.01
(100HZ) [19].

2
1
1n
i
Yrms Yi
n
 (2)
RMS: The accelerometer provides absolute analog
(AC) outputs, so we need to transform the AC signals
into the equal DC signals.
5.2. Body Activity Detection Mode Signal Processing
This mode is opposite to the body stress-reaction mode,
which means it detects the body activity turnin g fro m the
Figure 10. Body stress-reaction mode procedure flow
chat.

826 C.-S. Wang et al. / J. Biomedical Science and Engineering 3 (2010) 822-827
normal level to a low activity level. It represent some-
thing wrong might happened as well [19], and BAD
starts the emergency procedures immediately. For ex-
ample if the user passes out while exercising or if there
is no move reaction while working. Figure 11 presents
the body activity detection mode procedure flow chart.
Microprocessor reads the signal from accelerometer
every-10 milliseconds, and pass the signal through one
order passive HPF (High Pass Filter). Refer to Eq.1 [19],
and then calculate the RMS (Root Mean Square). Refer
to Eq.2, if the RMS is lower than the preset static RMS
threshold for a period of time, it means that emergency
has occurred. The system would beep out the warning
sound in the first place and if the user does not turn off
the alarm, the system would start emergency procedure.
5.3. Emergency Mode Signal Processing
When BAD determines there is a unusual condition, it
would beep out the warning sound in the first place to
inform the user there is a problem detected, and if the
user-does not turn off the alarm, BAD would deliver the
ATD dialing command to the mobile phone through
Bluetooth HFP [5], and mobile phone would build up a
voice connection with the terminal. While remote party
answered call active, BAD deliver the emergency signals
with sphygmomanometer, pulsimeter and GPS informa-
Figure 11. The body activity detection mode procedure
flow chat.
tion to the monitoring server via HFP by DTMF [5] till
the server is disconnected. Figure 12 presents the signal
process of emergency procedure.
Figure 12. The signal processing of emergency procedure.

C.-S. Wang et al. / J. Biomedical Science and Engineering 3 (2010) 822-827 827
6. CONCLUSIONS
This research provides a simple, practical, and portable
device of real-timely monitoring human body activity to
design a BAD with sphygmomanometer, pulsimeter and
GPS devices, combining mobile phone technology.
These devices detecting user being in both dynamic and
static emergency by body stress-reaction mod e and body
activity detection mode separately, and suggests a solu-
tion in the emergency situation. Especially for senior
care and patients who do face a risk, we expect people,
who carry with this device, would have more effective
care and help from others.
REFERENCES
[1] Cummins, R.O., Eisenberg, M.S., Hallstrom, A.P. and
Litwin, P.E. (1985) Survival of out-of-hospital cardiac
arrest with early initiation of cardiopulmonary resuscita-
tion. American Journal of Emergency Medicine, 3(2), 114-
119.
[2] Albarede, J.L., Lemieux, A., Vellas, B. and Groulx, B.
(1989) Psychological factors in falls in elderly patients.
Canadian Journal of Psychiatry, 34(2), 94-96.
[3] Maki, B.E., Holliday, P.J. and Topper, A.K. (1991) Fear
of falling and postural performance in the elderly. Jour-
nal of Gerontology, 46(4), 123-131.
[4] Wang, C.S., Lee, J.H. and Chu, Y.T., (2007) Mobile
telemedicine application and technologies on GSM. Bio-
informatics and Biomedical Engineering, ICBBE, the 1st
International Conference, 6-8 July 2007, 1125-1128.
[5] Bluetooth Special Interest Group (2005) Bluetooth
Specification, Hands-Free Profile (HFP) 1.5, November
25.
[6] Bluetooth Special Interest Group (2001) Bluetooth spe-
cification. Serial Port Profil, 1.1, February 22.
[7] National Semiconductor Inc. (2005) SC14440/431/432,
SC14435, SC14436/437/438. Baseband Processor for
PP/FP DECT and WDCT, 1.1, June 16.
[8] MEMSIC Inc. (2007) MXA2500G/M Improved, Ultra
Low Noise ±1.7 g Dual Axis Accelerometer with Abso-
lute Outputs, Rev. G, June.
[9] CSR Inc. (2006) BlueCore3-ROM CSP Product Data
Book, November.
[10] SiRF Inc. (2003) SiRF starIIe/LP Chip Set Data Sheet,
Rev 1.1, February.
[11] Nissila, S., Sorvisto, M., Sorvoja, H., Vieri-Gashi, E. and
Myllyla, R. (1998) Non-invasive blood pressure meas-
urement based on the electronic palpation method. Engi-
neering in Medicine and Biology Society, Proceedings of
the 20th Annual International Conference of the IEEE, 4,
1723-1726.
[12] Y.-B. Lin and I. Chlamtac, (2000) Wireless and mobile
network architectures. Wiley Computer Publishing, New
Delhi.
[13] Wang, C.-S. (2010) Mobile and wireless technologies
applying on sphygmomanometer and pulsimeter for pa-
tients with pacemaker implementation and other cardio-
vascular complications. Journal of Biomedical Science
and Engineering, 3(1), 47-51.
[14] HoneyWell, (2002) SSC Series Pressure Sensors 0-5 psi
Through 0-300 psi.
[15] Winbond Inc., (2006) W681388 Single Programmable
Extended CODEC/SLIC, Ver. 1.0.
[16] Micrpchip Inc., (2007) PIC24FJ64GA004 Family Data
Sheet.
[17] Bluetooth Special Interest Group (2004) Specification of
the Bluetooth System Core, Ver., 2.0.
[18] Pravin Bhagwat Reefedge Inc., (2001) Bluetooth: Tech-
nology for short-range wireless apps, IEEE Internet
Computing, 5, 96-103.
[19] MEMSIC Inc., (2005) Body Activity Detection Refer-
ence Design, August 25.