Andy Pletta, Adam Timmons, Tom Abbeg, Thomas McBeth (Students)
and Kurt Kosbar (Adviser)
Telemetry Learning Center
Department of Electrical and Computer Engineering
Missouri University of Science and Technology
This paper describes a project to implement a body area network to monitor the movements of a
human subject.--The sensor nodes can measure six degrees of movement by using a three axis
accelerometer and three axis gyroscope.--The data is transmitted wirelessly from the sensors to a
wearable microcontroller.--The microcontroller interfaces with a computer application that allows
a user to easily analyze and interpret the stored data.
During the past several years, advances in sensor technologies, wireless communications, and mobile computing have resulted in a tremendous increase in capacities for data collection and analysis.--At the same time, the costs of devices, and of systems in general, has declined.--One application area which has received considerable attention has been the development of systems for the
monitoring of human body functions and movement.
Central to many of these systems is the notion that the individual and collective movements of the
human body constitute a rich and dynamic set of data which, if available for analysis, can yield a
comprehensive view of the states or conditions of the specific movements being evaluated.--Two
broad application areas concerning body movement have emerged.--First, movements specifically
related to, or indicative of, health and medical conditions have been studied.--For example, the
movements of the body during sleep, during regular waking activities, and after severe or traumatic
injury can be examined with respect to movement patterns classified as normal or beneficial.-Second, athletic or sports-related movements have been studied with the intent of identifying and
measuring parameters of physical motion directly related to enhanced performance.
In some of these areas, localized measurements using individual sensors have been quite effective.-But to fully capture the underlying complexity of any particular movement, a multiple-sensor
network, and a programmable data collection and analysis module, would be helpful. Additionally, system performance and flexibility would be enhanced by the use of wireless communications
between the devices.--The application areas that can benefit from such a system include medical
diagnosis and treatment, as well as numerous sports and athletic applications. Some specific applications of this system might include:
Medical Applications:
● On-site or remote monitoring of movement to detect a decrease in speed or range of movement as compared to a specified norm.
● Detection of adverse bodily conditions, such as falling.
● Measurement of, and assistance with, physical therapy and rehabilitation from surgery,
stroke. etc.
Sports Applications:
● Golf swing analysis: professionals could use the system to analyze and improve a golfer’s
● Analysis of player movement efficiency or safety.
In the current market, there are many products that consumers can purchase to help monitor their
fitness.--Originally, pedometers and heart rate monitors were the only devices that could be used
in order to keep track of body motion.--With the invention of smartphones, applications were created to utilize the accelerometers and GPS to monitor body movement.--Several companies, such
as Garmin, Nike, Adidas, etc., have created devices to track fitness as well.--These devices
can upload data to smartphone apps or computers to allow the user to view the data.--However,
these devices only have a sensor in one location, which restricts what the devices are capable of
measuring.--Our device has three sensors that can be worn in several different locations.--The
ability to add and remove sensors makes our system flexible, allowing it to be used for many
different applications.
The project’s central objective was to implement an effective system of hardware and software
components that, operating together, can provide useful information about a person’s body movements.--It is important to have multiple sensors so more information can be obtained about a specific body movement or information received from one sensor can be verified for accuracy. Our
final project contains three digital sensors.--With more time, it would be possible to continue to
add more sensors, increasing the capabilities of the system.--Each sensor added requires changes
to the coding, not only on the microcontrollers, but also in our analysis application.
The second major goal of the project was to implement a simple wireless communication system
between the attached sensors and a central data processing module.--The team has successfully
accomplished this goal.--The addition of more sensors would be possible if time permitted as it
requires additional programming because data from each sensor must be parsed and additional
code written to separate and analyze it.
The final major goal of the project was to implement a computer-based application that can be
used to perform real-time or off-line data analysis.--A MATLAB program has been created that
performs basic analysis on the raw data received from the accelerometer and gyroscope.--This
analysis involves a plot of acceleration and rotation for each axis of all three sensors.--The direction that the sensor moved in can be determined from this data.--The acceleration plots show the
acceleration on each axis in g-force.--The rotation plot shows the rotation on each axis in degrees
per second.
The final product is composed of three sensors and the software needed to view the data.--The
sensors can be attached to a wrist band for comfort.--When the customer receives the device, they
will easily be able to wear the sensors, turn the device on, and begin using it.
The project team developed the following six specifications deemed to be necessary in order for
the device to function adequately (see Table 1).--The table shows the main highlights of the specifications and a more detailed description follows the table.
Table 1: Project Specifications Matrix
Battery life
Battery for each sensor must last
long enough to collect the data
Device Usage >= 45 minutes
Lightweight sensors
Sensors must be a comfortable
Weight of sensor <=200 grams
Size of sensors
Sensors must be small enough to
allow freedom of movement
Each subject retains total freedom of movement
Number of sensors
Number of sensors determines
system measurement capabilities
Number of sensors worn on
each subject to be at least 2
Transmit data wirelessly to a computer
Determine type of
Device must be able to transmit
data wirelessly over a convenient
Must be able to determine the
speed of movement of a person
wearing the sensors
Transmission range >=6 meters
Determine speed and acceleration of body part with 90%
The batteries must be powerful enough to last at least 45 minutes but also be fairly light so they
do not inhibit the movement of the user or cause discomfort.--The team wanted to be sure that the
user would be able to use the product without fear of losing power due to poor battery life. If the
device were to lose power, it could cause a loss of data.
Naturally, the lighter the sensors are, the more comfortable they will be to wear.--The weight of
the sensors on the wrist must not affect the person's movements, as this could alter the results the
systems is trying to achieve.--The group deemed that our sensors should not exceed 200 grams.
In addition to the weight of the sensors, the size of the sensors could also alter a person's movements.--If the sensors are too bulky, they could interfere with the user’s movement or cause discomfort.--Therefore, the group has made the specification that the device needs to be both lightweight and small.--There is no specific requirement in regard to volume, other than making the
the device small enough so that it does not restrict any movement.
It is important for the system to have multiple sensors.--If the system had only one sensor, it would
not be capable of giving the user enough information to accurately determine a person’s movement.--With two sensors, much more information can be gained.--The team has successfully
achieved the implementation of three sensors.
Another major specification is that the sensors must transmit wirelessly to a computer that will
collect and analyze the data.--If the sensors were wired it would cause a major inconvenience to
the user and would most definitely restrict movement.--Therefore, the sensors must successfully
transmit wirelessly and have a range of at least 6 meters.
Our last major specification is the ability to determine the type of movement using the devices. A
software program using MATLAB has been created that is able to take in the raw data and present it to the user in an easy to understand format.
The body movement monitoring system uses an Atmel ATmega256RFR2 Xplained Pro board
(Figure 1), an Atmel ZigBit extension module (Figure 2), and an InvenSense MPU-6050 accelerometer and gyroscope sensor (Figure 3).--The ZigBit and MPU-6050 use a two-wire interface
to communicate with each other, while using a wireless transceiver to communicate with another
board.--Both the Xplained Pro board and the ZigBit extension modules use the same microcontroller.--This microcontroller was chosen because of its built in wireless transceiver, 256K bytes
of flash RAM, and team members’ familiarity with programming Atmel AVR microcontrollers.
Figure 1: Atmel Xplained Pro Board
Figure 2: Atmel ZigBit Module
Figure 3: InvenSense MPU-6050 Accelerometer/Gyroscope
The Xplained Pro board is much larger than the ZigBit extension module and has a micro USB
cable that can be used for serial communication.--These features of the Xplained Pro board make
it well suited for the base unit of the system.--The job of the Xplained Pro board is to receive
wireless data from the sensors and send these values over the serial bus to be displayed on the
computer.--The ZigBit extension module is very small, which makes it well suited for the sensor.-The digital accelerometer and gyroscope is attached to the ZigBit extension module, and transmits
the data to the microcontroller via two wire interface.--The ZigBit extension module then wirelessly transmits this sensor data to the base unit.
These devices require a voltage of approximately 3 Volts.--The team found a perfect size battery
pack to fit the back of the ZigBit module but it was for 6 Volts.--This configuration was two
CR2032 batteries (CR2032 is a 3 Volt battery) in series.--The team modified the battery pack to
implement two CR2032 batteries in parallel, thus providing a 3V source, as well as extending the
battery life by being able to use two batteries at a time.
The team combined all of these devices into one unit as can be seen in Figure 4.--The battery pack
is mounted underneath the ZigBit and its wires (the black and red wires in the photo) wrap around
to power the devices.--The MPU-6050 is mounted directly on top of the ZigBit module.
Figure 4: Body Movement Monitor Sensor: ZigBit module, battery pack, and MPU-6050.
The MPU-6050 sensor combines the accelerometer and gyroscope.--The included accelerometer
can function at a range of ±2g, ±4g, ±8g, or ±16g.--The gyroscope can measure at ±250, ±500,
±1000, or ±2000 degrees per second.
The digital system that we created has three sensors that transmitted data from the sensors wirelessly to the base unit (the Xplained Pro board).--Once the base unit received the wireless transmission from the sensors, the data was formatted and transmitted to the computer using a UART
connection.--Each set of data begins with a character that corresponds to the sensor that transmitted
the information.--Having each data set begin with a unique character allows the data to be parsed.-The next set of numbers corresponds to the X, Y, and Z axis of the accelerometer and the X, Y,
and Z axis of the gyroscope.
The data is parsed and analyzed using MATLAB.--The code parses the data into six two-dimensional arrays.--The columns of the array corresponds to which sensor transmitted the data. For the
example above, the first column of the array would correspond to sensor B, the second column
would be sensor A, and finally the third column would be sensor C.--Using a multidimensional
array reduces the number of arrays needed to hold the data received from multiple sensors.
Two figure windows are generated for the data received by the sensors.--These two figures can be
seen below in Figure 5.
Figure 5: Figures generated from MATLAB code
The left-hand figure is the data received from the accelerometer.--Within this figure, there are
subplots that correspond to each sensor.--The number of subplots generated is related to the number of sensors used to measure body movements.--The order subplots are generated corresponds
to the sensor data that was received first.--In this example, data from sensor B was received first,
followed by A and C respectively.--The right-hand figure shows the data received from the gyroscope and has the same features as the accelerometer plot.
Before analyzing the data from the sensor, it is important to note the way the user has attached the
sensor onto their body.--The way the user attaches the sensor to their body directly effects which
axis the movement appears on.--The sensors A and B were attached to the users left and right arm
as shown in Figure 6.--The third sensor, sensor C, was placed in the pocket of the user.--A close
up the ZigBit module can be seen in Figure 7.
Figure 6: Sensor A (left wrist) and Sensor B (right wrist) attached to User
Figure 7: Close up of Sensor (note axis silkscreen)
The user, in this example, rotated is left wrist 180 degrees to the left, and then rotated his entire
left arm up.--At this point, his left wrist was facing away from him, and his fingers were point up
toward the ceiling.--The user then undid these movements, by rotating his entire left arm down
and then rotating his wrist to the right 180 degrees.--Knowing how the sensor is attached to the
person, the graphs can be used to analyze what types of movements were made.--A plot of the
gyroscope data for Sensor A can be seen in Figure 8 below.
Figure 8: Plot of Gyroscope Data for Sensor A (Attached to User’s Left Wrist)
Looking at the first two positive peaks (green and blue), it can be concluded that the left wrist
rotation was a much quick rotation than when the user rotated their arm up.--The two negative
peaks are mirror images of the positive peaks.--This means that the user undid his initial movements.--This is a very basic example of analyzing body movements using the data received from
the sensors.
This project demonstrated the feasibility of implementing a system of hardware and software components that, operating together, provide useful information about a person’s body movements.-The capabilities of the basic system implemented here can be extended by the addition of more
The successful implementation of a simple wireless communication system between the attached
sensors and a central data processing module showed that the system allowed for a very flexible
means of collecting and transmitting body movement data.
Finally, the project showed the feasibility of implementing a computer-based application for performing real-time or off-line data analysis.--Basic analysis of the raw data received from the accelerometers and gyroscopes showed that the speed and direction of individual body movements
could be plotted, thus providing a graphical representation of these movements, which can be used
in further analysis.
Analog Devices, “Small, Low Power, 3-Axis ±3 g Accelerometer,” ADXL335, Jan. 2009
[Revised Jan. 2010].
Atmel Corporation, “8-Bit AVR Microcontroller with Low Power 2.4 GHz Transceiver
for ZigBee and IEEE 802.15,” [Revised Sept. 2014].
Atmel Corporation, “Atmel ATMEGA256RFR2 Xplained Pro User Guide,” [Revised
Dec. 2013].
Atmel Corporation, “ZigBit 2.4 GHz Single Chip Wireless Module ATZB-S1-256-3-0-C
Datasheet,” [Revised Mar. 2014].
Atmel Corporation, “ZigBit Extension User Guide,” [Revised Jul. 2014].
Atmel Corporation, “ATDH1150USB CPLD JTAG ISP Download Cable User Guide,”
[Revised Apr. 2014].
InvenSense Corporation, “MPU-6000 and MPU-6050 Product Specification Revision
3.4,” [Revised Aug. 2013].
InvenSense Corporation, “MPU-6000 and MPU-6050 Register Map and Descriptions Revision 4.2,” [Revised Aug. 2013].
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF