Communications and Network, 2013, 5, 132-135
http://dx.doi.org/10.4236/cn.2013.53B2025 Published Online September 2013 (http://www.scirp.org/journal/cn)
Low Power Design of BAN Node for Healthcare System
Tao Xue, Fengye Hu, Liying Zhao, Dafei Zang
College of Communication Engineering, Jilin University, Changchun, Jilin, China
Received January, 2013
ABSTRACT
In this paper, we design and implement a body area network (BAN) based on ZigBee to detect physiological parameters
data. The main content can be divided into the design of end node (EN) and sink node (SN). EN collects data by
physiological parameter sensor, processes the collected data and sends the data to SN. SN receives all EN’s data, and
transmits data to PDAPersonal Digital Assistan t in certain format. PDA sends the received data to the internet or
saves it to database. The performance is evaluated through numerical results. The results indicate that the system not
only have superiority on complexity, but also have a low power consumption.
Keywords: BAN; EN; SN; Physiological Parameters
1. Introduction
Body area network (BAN) has received widespread at-
tention recently, and got a growing number of scholars
involved in its research and development. BAN is a
sub-domain network of personal area network (PAN),
which is mainly composed of wireless sensor nodes dis-
tributed on, near, or within a human body. Its develop-
ment is of great help to the development of medical, en-
tertainment and public health. BAN is not only a new
universal health care, disease surveillance and preven tion
solution, but also an important part of Internet of things
(IOT). The aim of this paper is to provide a ubiquitous
computing platform integrated with hardware, software
and wireless communication technology. The develop-
ment of it also provides the necessary conditions for the
future development of universal health care [1]. IEEE’s
WG15 group is engaged in the BAN’s standards. The
group used to assert that ZigBee is a perfect way of
communication for BAN. In this paper, we design and
develop a network using ZigBee to detect people’s
physiological parameters.
The system structure is shown in Figure 1. The system
consists of sink node (SN), end node (EN) and display
part. EN’s work is collecting data by physiological pa-
rameter sensor, and processing the collected data and
sending the data to SN through wireless module. SN’s
work is receiving all end nodes’ data, and transmitting
data to PDA in certain format. PDA sends received data
to the internet or saves it to database. Doctors could
monitor the patient’s real-time physiological parameters
data remotely or examine the patient’s historical physio-
logical parameters data.
Figure 1. System structure.
2. Hardware Design of Ban’S Node
2.1. A Overview of ZigBee
ZigBee works on the free frequency of ISM2.4GHz, and
has the characteristics of short-range, low power, low
speed and low cost [2]. The shortest distance it can
spread is 10 m, its power is around 1mW under normal
circumstances and transfer rate is 250 kbit/s. There is no
doubt that ZigBee is at a disadvantage in such aspects as
transmission distance and transfer rate. With its low
power consumption and streamlined protocol stack,
however, it is more suitable for wireless sensor network
than other communication modes. Furthermore, ZigBee
has unique advantages over other wireless communica-
tion modes. The advantages focus on the following three
points:
ZigBee has less damage on the human body. Lit-
erature [3] indicates that the higher the frequency
of electromagnetic waves, the shorter the wave-
length would be. The result of this is that the
fluctuation of wave’s energy when it spreads is
more severe, and its intensity decay more greatly.
Therefore, it is more likely to b e absorbed by our
skin and does less harm to our body when the
high frequency waves pass through our body.
C
opyright © 2013 SciRes. CN
T. XUE ET AL. 133
ZigBee’s standard working frequency is 2.4 GHz,
which is much higher than traditional 433 MHz
and also has less harm to human.
ZigBee’s energy consumption is very low. Zig-
Bee adopts a variety of technologies to reduce its
energy consumption, such as reducing the trans-
mission bandwidth to let node’s work mode
switch more quickly, using Direct Sequence
Spread Spectrum (DSSS) and s o o n [4 ] .
ZigBee chip is cheap and has mature technology.
All these lead that it is more suitable for mass
production. A typical ZigBee chip is CC2530F-
256, which is also used in my design. The chip’s
price is only $2, and has technical support such
as a half-open source stack, debugging applica-
tions etc.
2.2. EN Design
The structure of EN is shown in Figure 2. It is mainly
consist of collection part, controller part and wireless
transmission part:
Collection part’s function is physiological signal col-
lection and processing. The physiological signal includes
temperature and heart rate. Take heart rate for an exam-
ple, the type of sensor used here is SC0073. SC0073 is a
high performance, low-cost small piezoelectric pressure
sensor, its sensitivity is greater than 0.3 mV/pa and re-
sponse frequency is 1000 Hz. The sensor’s output signal
is analog signal. The characteristic of this signal is weak
and mixed with high-frequency noise. The system needs
to use certain amplifier circuit and filter circuit to get
signal whose voltage value is within 0-5 V, frequency is
less than 10 Hz.
Control part’s function is to transform analog signals
into digital signals, and then judge if the signal is wrong
by measuring if the measuring value is in the correct
range. The control chip used here is CC2530. Since
CC2530 has a 14-bit AD converter, an 8051 core and
ZigBee wireless module, it can realize analog-to-digital
conversion, controller and wireless transceiver by itself.
We need to write and read specific register to send and
receive data since the chip has been integrated with
wireless transmission module.
Node uses two AA batteries to support energy. Volt-
age stabilizing chip used in syste m is AVX’s TAJB- 106
M016 chip, the input vo ltage value is 5 V and the output
voltage value is 3.3 V. The amplifier circuit, filter circuit
and CC2530 can get energy through this design.
2.3. SN Design
The struc ture of SN is shown in Figure 3. It mainly con-
sists of serial port part, controller part and wireless
transmission part:
SN’s structure is similar to that of EN. It also uses
CC2530 to realize wireless data communication and con-
trol functions. The difference between SN and EN is that
SN use CC2530’s serial port to send the data to the
computer. The reason is that the communication between
control chips is usually finished by serial port. Using
serial port to communicate not only can complete the
current design, but also can make it easy to communicate
with other ordinary systems. Serial is also used to com-
municate with PDA system in practice, the difference is
that the serial output voltage of the control chip is +5V/0
V, but the computer’s serial port is +12V/-12V. Here
needs MAX232 to finish voltage level switch.
Figure 2. EN structure.
Figure 3. SN structure.
3. Software Design of Ban’S Node
The compiling environment used to compile node’s pro-
gram is IAR’s Embeded Workbench for MCS-51 V7.51
A. Simulator used here is CC Debugger, and the emula-
tor is connected with the node via the JTAG in terface.
The code uses TI’s Z-Stack-CC2530-2.30 protocol to
realize some function. Z-Stack is a semi-open-source
protocol stack code. It supports both ZigBee2007 and
ZigBee2007 Pro. All wireless modules meeting the same
standard of ZigBee can be interconnected [5]. The struc-
ture of the protocol stack is shown in Figure 4. In gen-
eral, the user needs to add three additional documents to
complete a project. They consist of a main file which is
used to store task time processing function, a file that is
used to save the main file header and an op erating system
interface file which is used to save the task handler func-
tion [5]. In the preparation of these three programs, all
the user need to do is to find the system function that can
finish the request and use it [6]. For the equipment de-
fined in the new system, users can program device driver
Copyright © 2013 SciRes. CN
T. XUE ET AL.
134
function themselves or program to operate the hardware
themselves.
3.1. Software Design of EN
The main work of the EN’s software can be described as
the following three parts: collecting the physiological
data, judging the validity of the collected data and send-
ing the data to SN through wireless module. The flow
chart of EN is shown in Figure 5.
Each node begins to run after power on. Firstly, it is
necessary to initialize each node, such as: initialize the
state of the wireless port, the pin which is connected to
sensor and status indicator. After all initialization is fin-
ished, each node begins to search for net which the EN
could join in and file a petition until join successfully.
After joining the network, node needs to collect data by
converting the analog data of port connecting with the
sensor. Then each node judges if the collecting data is
valid and determines whether to send to the sink node.
Each node records the failure time through variable flag
simultaneously and sets the status indicator to warm
when there are more than 5 times failure in series. Enter
the status of receiving and waiting for data asking if node
collect data successfully. The no de will come back to the
status of collecting data after sending data successfully.
3.2. Software Design of SN
The main work of the SN is to receive the physiological
data sent by EN and send the received data to serial port.
The flow chart of sink no de is sho w n in Figure 6.
The first thing after power on is also initialization . The
Initialization includes initializing the wireless module,
initializing the serial port and initializing the state indi-
cator. Wireless net is built through sending net message
to surrounding. After building net successfully, SN sends
net-built message to the serial port to indicate that net is
built successfully. After all this is done, the SN begins to
receive the end nodes’ join request and save their PANID.
If there is no node that wants to join in after a certain
time, the sink begin to ask for data by recording PANID
and send the data to the serial port.
Figure 4. Z-Stack protocol stack code.
Figure 5. Sink node structure.
Figure 6. Sink node structure.
4. System testing
4.1. The Setting of Serial Port
The parameters of serial debugging assistant are set as
follows: port is set to com1 and baud rate is set to 57600.
Parity bit is used to prevent communication interference
[7]. System will recognize fault if the odd bits change
when parity is set to odd parity. This way of calibration
is often used in serial port’s long-distance transmission,
Copyright © 2013 SciRes. CN
T. XUE ET AL.
Copyright © 2013 SciRes. CN
135
so it is unnecessary to adopt it here. The function of data
bit is to show the data length of received data. It is set 8
here because computer and sink use ASC-ii to commu-
nication. Stop bits is used to represent the last bit of a
packet and its function is to give a chance of calibrating
the time. The typical value of Stop bits is 1 [8].
……
4.2. Everment of Test
SN is connected to computer through serial port. Start the
serial debugging aid on one computer with system of
windows XP SP3 and make sure all the serial drivers are
normal. The test is divided into two steps. The first step
is to place the sensor in the indoor environment and ob-
serve the output. The second step is to take a healthy man
as a test object, place the temperature sensor in the man’s
armpit and the pulse sensor on the man’s wrist. Make
sure all of these two sensors are connect to man’s skin
tightly. Lastly, we calculate the statistic distribution of
measurement data.
Figure 8. Temperature distribution.
5. Conclusions
In this paper, a BAN based on ZigBee to detect physio-
logical parameters is designed and implemented. The
design can be divided into hardware and software of
nodes in the net. EN’s work is to collect, process and
send the data. SN’s work is to save, display the data and
upload the data to Internet. One main aim of the design is
to keep the system low power and work longer. Doctor
can monitor the patient’s body temperature and heart rate
(which can be expanded as needed) in time or examine
patient’s physical parameters of historical data through
this system. Numerical results indicate that the system
not only reduces node’s energy consumption, but also
simplifies the complexity of the system. The system also
performs well and has the potential to be popu larized.
4.3. Result of Test
The first test is at indoor environment and don’t let the
sensor touch anything. The temperature is around 20
and the heart rate is around 0 beats/min. However, be-
cause of the error, the value doesn ’t maintain.
Figure 7 and Figure 8 show the data test by sensors
placed on a adult man’s body testing in an hour. Since
the amount of data is large, a picture is paint to show the
value’s distribution. According to statistics58 groups of
measured heart-rate data is tested, including 54 groups
range from 70 beats/min to 80 beats/min. The other four
groups have 2 at 0 beats/min, 1 at 4 beats/min and 1 at 8
beats/min. The correct rate is around 0.931 according to
the measurement. The temperature includes 50 groups
range from 36 to 37, 4 at 0, 1 at 6 and 3 be-
yond 40. The correct rate is around 0.862. Although
there are some wrong data, the error data is easy to be
distinguished. A conclusion could be drawn that the de-
sign system can measure human’s physiological parame-
ters and draw real-time results.
REFERENCES
[1] J. B. Gong, R. Wang and L. Cui, “Research Advances and
Challenges of Body Sensor Network(BSN),” Journal of
Computer Research and Development, Vol. 47, No. 4,
2010, pp. 737-753
[2] X. F. Le, J. X. Tan, P. C. Wang and N. Hu, “Study on
Power SCADA System Based on ZigBee Pro Technol-
ogy,” 210st ed., Vol. 31. Telecommunications for Electric
Power System, 2010, pp. 40-44.
[3] S. Z. Wu, “Bioeffect of Electromagnetic Wave and Per-
son’s Health. Journal of South-Central University for Na-
tionalities (Nat. Sci. Edition), Vol. 29 No. 1, 2010, pp.
57-61.
[4] S. Farahani, “ZigBee Wireless Networks and Transceiv-
ers,” Newnes, Sep 2008.
[5] Texas Instruments,Inc,Z-Stack Developer’s Guide.
http://www.ti.com.cn,2010.
[6] S. W. Gao and C. Y. Wu, “Technical Hands-on Tutorials
of ZigBee,” Beijing Aerospace University Press, Jun
2009.
[7] D. Y. Li, W. J. Wang and X. Y. Meng, “Distributed
Computer System of Taking Even-odd Check as Multi-
Computer Communication,” Vol. 12 No. 1, 2002, pp.
19-21.
[8] NI, The basic concept of serial communication,
http://digital.ni.com/public.nsf/allkb/9F439B3D68D9072
F4825703000383197,Jul,3 2005.
Figure 7. Heart ra te distr ibution.