Transcript Test Poster Font Arial – pt 44

Senior Project – Electrical Engineering- 2009
Biofeedback Training System for
Improved Athletic Performance
Matt Statton
Advisor – Professor Helen Hanson
Abstract:
The goal of this project is to produce a biofeedback training system
that leads to improved athletic training. Electrical signals produced
by muscles when they contract, measured by an electromyograph,
are used as an indicator of muscle fatigue. The measured signals
are processed using a microcontroller, which analyzes the signals
and provides a feedback response to the user when the amplitude
gets below a certain threshold. In the end, a biofeedback training
system was produced that can help people of all ages and fitness
levels maximize their benefits from exercise, reach their goals
faster, and improve their health.
Design Requirements:
•Measure electrical signals from muscles
•Use electrodes to measure electrical signals
•Amplify signal so it can be analyzed
•Analyze signal to determine level of muscle fatigue
•Analog-to-digital conversion of signal from muscle
•Measure maximum and relative maximum amplitude peaks of
signal
•Determine threshold at which muscle fatigue occurs
•Compare relative maximum to absolute maximum of signal to
determine level of muscle fatigue
•Provide feedback response to user when muscle fatigue occurs
Electromyograph:
Signal Processing:
The electromyogram was analyzed using a Silicon Labs
C8051F020 microcontroller. After analog-to-digital conversion
using the on-board 12-bit analog-to-digital converter, the
microcontroller was programmed to measure the average
amplitude of the signal during the first repetition. A running
average of the following repetitions was compared to the initial
average, and a feedback response was provided to the user when
the running average reached a threshold of muscle fatigue based
on the initial average.
Figure 1: EagleCAD schematic designed based on circuit Figure 2: EagleCAD board file created from schematic
diagram from
http://instruct1.cit.cornell.edu/courses/ee476/FinalProject
s/s2005/bsm24_ajg47/website/website/index.html
A MAX666CPA voltage regulator was used to regulate voltage from
a 9V battery to 5V to power the operational amplifiers. An LT1494
operational amplifier, with gain A = 1, was used to create a virtual
ground of VCC/2 = 2.5 V. An INA106 differential amplifier, with
internal gain A = 10, was used to amplify the differential input from
two surface electrodes placed on the proximal and distal portions of
the muscle. External resistors were used to increase the gain to A =
110. The electromyograph was created using an LPKF ProtoMat
C20S circuit board plotter from a schematic made in EagleCAD. All
parts were soldered into the board created by the circuit board
plotter. The inputs were connected to electrode leadwires, and the
output was connected to the 8051 microcontroller for processing.
Figure 5: Flow chart of electromyograph signal analysis program
Results:
Electromyogram of Fully Contracted and Slightly Contracted Muscle
2.65
2.6
Voltage (V)
2.55
2.5
2.45
2.4
2.35
2.3
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
Time (s)
0.4
0.6
0.8
1
Figure 6: Measured Electromyogram signal
displaying fully contracted, minimally
contracted, and resting biceps muscle
Figure 3: Front of electromyograph circuit
Figure 4: Back of electromyograph circuit
Acknowledgments:
Professor Hanson
Professor Hedrick
The electromyograph measured the
electrical activity of the muscles.
However, the microcontroller has not
yet been able to consistently
measure muscle fatigue based on
the amplified electromyogram data.
Further work is being done to enable
the microcontroller to more
accurately measure the level of
muscle fatigue of the user.
Ben Bunes
Internal Education Fund