See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/333118571

Electromyography (EMG) signal acquisition and processing by using surface

electrodes

Presentation · May 2019

DOI: 10.13140/RG.2.2.26222.46408

CITATIONS READS

0 3,712

1 author:

Jayanta Ghosh
National Institute of Technology, Durgapur
5 PUBLICATIONS   0 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Noise immune high frequency (GHz ranges) RF signals multiplier (Mixer) using 180nm CMOS technology. View project

Design of a Communication Architecture for Unmanned Aerial Vehicle (UAV) Swarm Networks View project

All content following this page was uploaded by Jayanta Ghosh on 15 May 2019.

The user has requested enhancement of the downloaded file.

 CSIR – CMERI
Central Mechanical Engineering
Research Institute

INTERNSHIP PROJECT REPORT

 INTRODUCTION

For the acquisition of the EMG signal, two electrodes

were placed in the front and back side of the
lower arm assuming that for each
movement the cross-talk content will be
mostly the same for each signal. The signal
contains different patterns for the different
movements and goal was to characterize
those patterns. For the classification of
acquired signal I am using techniques of
wavelet and autocorrelation to extract
relevant features.
The first few pages of this report have the review of the
various paper on the respective topics and next
half have our research glimpses.
RESEARCH PAPERS AND JOURNALS REFERRED

1. Surface EMG Signal Acquisition Analysis and

Classification for the Operation of a Prosthetic
Limb
Kasun Samarawickrama*, Sadun Ranasinghe, Yasoja Wickramasinghe, Wageesha
Mallehevidana, Vidarshi Marasinghe and Kanchana Wijesinghe
Department of Electrical and Electronic Engineering, General Sir John Kotelawala Defence
University,
Ratmalana, Sri Lanka.

In this paper SEMG signal parameters related to the upper limb speed
and flexion angle for one test subject is analysed. However, the
ultimate goal of this paper was a generalized algorithm for speed
calculations. Also, a low cost data acquisition unit to acquire EMG
signal is designed.
Out of the two methods to acquire EMG signal, non-invasive method
was selected in which single use and adhesive type electrodes were
used for EMG signal acquisition. Since the research was focused on
the upper limb, first electrode was placed on the bicep brachii, second
one on the bicep muscle tendon as reference electrode and the third
one on the elbow as the ground electrode.
Beaglebone Black signal processing unit interfaced with MATLAB
R2015a was used for processing the SEMG signal. The analog input
EMG signal was converted into digital signal inside the Beaglebone
and data was outputted via USB to PC. MATLAB R2015a software
was selected to visualize the signals as it is compatible with
Beaglebone.
Flexion angle was calculated by using goniometer(electrical
potentiometer) to perform the curve fitting and the potentiometer was
interfaced with Arduino ATMEGA 2560 to store the flexion angle
data to the PC. The techniques used for speed calculations are ;
Fourier Transformation and Wavelet transformation.

2. Design of surface electromyographic signal

acquisition system based on MATLAB
Dan Ren, College of Engineering & Technology, Eastern Liaoning University, Dandong
Liaoning

In this paper a surface electromyography signal acquisition system is

designed. The hardware parts are selected from the Thought
Technology. The structure of surface EMG signal acquisition system
consists of; Surface EMG sensor, Flexcomp Infiniti encoder, TT-USB
signal converter, PC software processing.
The surface EMG sensor used is a Myoscan Pro EMG sensor by
Thought Technology. The sensor is connected to one-time surface
electrodes.
The data acquisition device used is the Flexcomp Infiniti encoder. It
is the ideal data acquisition and physiological monitoring device for
power users. It can acquire data from any thought technology sensor.
The encoder resolution is 14 bits and the sampling rate is 2048 Hz.
The encoder has 10 channels and each channel is relatively
independent in the process of sampling.
The TT-USB receives the data from the Flexcomp Infiniti encoder via
fiber optics and sends it on to the PC via the USB port so that the PC
can correctly read and store the SEMG signal collected by the encoder
correctly.
PC communication software based on MATLAB is used for further
analysis and requirement of real-time system.
 3. Acquisition and analysis of EMG signals
to recognize multiple hand movements
for prosthetic applications
Giuseppina Ginia,∗, Matteo Arvettia, Ian Somlaib and Michele Folgheraiterc
aDepartment of Electronics and Information of Politecnico di Milano, Milan, Italy
bDepartment of Micro-technology and Medical Device Technology, Technical University of
Munich,
Munich, Germany
cDFKI, Bremen, Germany

The two ideas introduced in this paper are: the positioning of

electrodes and the method of data classification.
After analysing the movements necessary for grasping, five
movements for wrist hand mobility were individuated and a small
electronic device (controller) capable of registering those movements
is built which can be integrated into hand prosthesis. There are
basically two types of hand prostheses viz active and passive but the
one of concern in this paper is active prostheses.
The active prostheses use electrical motors powered by batteries. The
controller uses a geared automatic transmission to move to a lower
transmission rate when the sensor signals the grasping of an object.
The two main building blocks of controller are: acquisition of EMG
signals and the classifier.
For the acquisition of the EMG signal, two electrodes were placed in
the front and back side of the lower arm assuming that for each
movement the cross-talk content will be mostly the same for each
signal. The signal contains different patterns for the different
movements and goal was to characterize those patterns.
For the classification of acquired signal techniques of wavelet and
autocorrelation was used to extract relevant features for which a
neural classifier in cascade with wavelet analysis was used.
The controller uses a pattern recognition approach which includes
feature extraction and a classifier. Five statistical features were
extracted from the temporal sequence and the classifier was devised
as a multi-layered neural network whose inputs are features extracted.

4. Design and Development of a Low Cost EMG

Signal Acquisition System Using Surface EMG
Electrode
T. S. POO#1, K. Sundaraj#2
#School of Mechatronic Engineering,
Universiti Malaysia Perlis (UniMAP),
Kompleks Ulu Pauh,
02600 Arau,
Perlis.Malaysia.

This paper aims at developing of low cost physiotherapy EMG signal

acquisition system. EMG signal is collected from the bicep muscle
with the help of snap type SEMG electrodes from Medi-Trace mini
series electrodes which is then fed to the amplification stage because
the raw EMG signal is very small.

In the amplification stage, EMG signal is first passed to an

instrumentation amplifier with the gain of 100 and then it is fed to an
op-amp with a gain of 20. The signal is then fed to a notch filter to
remove the 50Hz noise which is being generated in every power
source. Amplified and filtered EMG signal is rectified by a diode to
remove the negative part of the signal. Low pass filtering is done to
obtain the outer envelope of the EMG signal by eliminating 3Hz
frequency component in the signal which is then fed to a conditioning
ADC (AD7710) to sample the signal into 24bit data. The sampled
signal is then send to PIC for transferring the digitized data to the
computer. Plotter software is used to plot the linear envelope EMG
signal in the computer.

5. MYOS: An Efficient Multi-Channel EMG Data

Acquisition and Analysis System
I. K. Kitsas’, L. J. Hadjileontiadisl, I. N. Papakonstantinou’, and C. Kotzamanidis‘
‘Department of Electrical & Computer Engineering, Aristotle University of Thessaloniki,
Thessaloniki, GREECE
2Denartment of Phvsical Education and Athletics. Aristotle Universitv of Thessaloniki.
Thessaloniki. GREECE

In this paper a computer based data acquisition and analysis system

MYOS is designed enabling multi-channel acquisition, data storage
and analysis. MYOS circumvents the various drawbacks of most of
the traditional acquisition system which is basically the inability to
acquire signals from multiple single unit acquisition instruments.
MYOS is divided in two modules: The acquisition interface and the
data analysis interface. In the acquisition interface, BNC adapter is
serving as an interface between instruments and DAQ card. The DAQ
card used is PCI-6040E. It is a multifunction input/output signal
acquisition card supporting 16 single-ended analog inputs. MYOS
acquisition interface is built on the National Instruments Labview 6.1
environment as an integrated stand-alone application. Second module
viz data analysis interface is a software application built on MATLAB
6.5 environment. The independence between the two modules and the
support of multiple file formats provides great compatibility.

6. Application of EMG signals for controlling

Exoskeleton robots
Christian Fleischer*, Andreas Wege, Konstantin Kondak and Gu¨ nter Hommel
Technical University of Berlin, Institute for Computer Engineering and Microelectronics,
Berlin, Germany
This paper surveys two exoskeleton systems developed in their
laboratory. First system is a lower-extremity exoskeleton with one
actuated degree of freedom in the knee joint. Second system is an
exoskeleton for a human hand with 16 actuated joints, four for each
finger. Basically, the overview of the design and control methods is
presented in this paper. The common approach to control exoskeleton
robots is based on using force sensors but there are some drawbacks
to this approach.

Also, two different approaches for application of EMG signals for

motion control of exoskeleton robots are given: the first approach is
called dynamic human body model (DHBM) and the second one is
direct force control (DFC). A main issue of these approaches is
conversion of measured EMG signals into muscles forces
(myoprocessor). In this paper an adaptive algorithm is presented for
the myoprocessor and the integration of the calibration procedure
necessary into the system.

Exoskeleton for the leg

The system is composed of an orthosis with an actuator, a

microcontroller connected to a PC and hall sensors to measure the
knee angle. For data acquisition and processing, three Delsys 2.3
differential electrodes which have an inbuilt amplifier of gain 1000
and a band pass filter is used. The EMG signal and force signals are
sampled with 12-bit ADC. The recorded EMG signals are then
rectified and low pass filtered with a second order Butterworth filter.
The force measurements are also low-pass filtered at 2Hz during
calibration to avoid misalignment of data. The post-processed EMG
signals are then converted into muscle forces by the EMG-to-force
function and the resulting knee torque produced by the actuator is
calculated. To control the exoskeleton both the above mentioned
approaches can be used.

Exoskeleton for the hand

Fingers are moved by a construction of levers actuated through pull
cables guided by flexible sheaths. Pulleys at the levers allow
bidirectional movements. Finger joint angles are measured by hall
sensors and force sensors integrated between the levers and finger
attachments measure forces during flexion and extension at the finger
joints. Surface EMG electrodes (Delsys 2.3) measure muscle activity
at eight points on the forearm. EMG sensor data can be used to
control the hand exoskeleton but there are some difficulties in the
application of algorithm. Blind source separation can be used to
overcome the problem. Another idea is to use EMG signals to
recognize user’s intention for a specific gesture.

7. Review of Surface Electromyogram Signals: Its

Analysis and Applications
World Academy of Science, Engineering and Technology
International Journal of Electrical and Computer Engineering
Vol:7, No:11, 2013

This paper focuses on surface electromyogram (SEMG) signals and

reviews the works on the surface EMG signal processing and its
analysis for engineering research in diverse areas. This paper gives a
brief description on the generation of myoelectric signals and various
techniques for processing that can be applied for SEMG signals. The
various techniques for recording signal, acquisition and amplification,
noise removal and analysis of SEMG signal are explained. Also a
brief description on application of SEMG signal in various fields are
given such as estimation of muscle fiber conduction velocity
(MFCV), control of prosthetic devices, clinical diagnosis and in
developing of an automatic speech recognition (ASR) .
 8. Different techniques for EMG signal processing
J. Paul
Bialystok Technical University
Wiejska 45C, 15-351 Bialystok
Phone: +48 85 7469222
e-mail: [email protected]

This paper relates to the upgrading existing methodologies, filtering,

processing, decomposition and modelling of EMG signal.
This paper introduces a procedure for filtering electromyography
signals in which a polynomial filter based on microprocessor zilog
8(Z8) is used. Communication with filter is established via serial port
(RS232). Wavelet transform is proposed as an efficient mathematical
tool for local analysis of non-stationary and fast transient signal.
Wavelet transform can be implemented by means of a discrete time
filter bank. Analysis was done using MATLAB 6 wavelet toolbox. In
this paper the regressive (time series model) has been used to study
the EMG signal. SEMG signal can be represented as an
autoregressive model with the delayed intramuscular EMG as the
input. To study the performance of the regression model of muscle
activity, several realizations of a number process models were
generated and the model coefficients were estimated.
 Sensors used in lower limb exoskeleton

1. Design of a Walking Assistance Lower Limb

Exoskeleton for Paraplegic Patients and Hardware
Validation Using CoP
Jung-Hoon Kim1, Jeong Woo Han2, Deog Young Kim3 and Yoon Su Baek2,*
1 Yonsei University, School of Civil and Environmental Engineering, Construction
Robot and Automation Laboratory, South Korea
2 Yonsei University, School of Mechanical Engineering, South Korea
3 Yonsei University, College of Medicine, Department of Rehabilitation Medicine,
South Korea

Forces acting on a walker can be separated in two categories:

forces exerted by contact and forces transmitted without contact.
The former is related to centre of pressure (CoP) and latter is
related to zero moment pressure (ZMP). Both these concepts are
useful for the control of the dynamic equilibrium of biped
walking robots. The point associated with contact forces is
referred as CoP while ZMP is considered to be related to gravity
plus inertia forces.

CoP is used in the assistive exoskeleton for the purpose of

detecting the human intention to walk and checking the stability
while ZMP is used in control of humanoid robots. In this paper
CoP is considered as a method of determining the intention of a
patient to walk.

To measure CoP in exoskeleton robot, two kinds of force

sensors units are installed in the exoskeleton: low profile force
sensors in foot modules to measure the human weight
transferred to the ground and a load cell at the shank (knee to
ankle) frame to measure the supporting force.
A load cell is a transducer that is used to create an electrical
signal whose magnitude is directly proportional to the force
being measured. Various load cell types includes: hydraulic,
pneumatic and strain gauge. Mostly strain gauge is preferred.

Following sensors are used to know the positions of all the

contact points where the forces are transferred from the
exoskeleton and the wearer to the ground:
• Angle sensors at all joints(14 DOF exoskeleton joints)
• A force sensor measuring the reaction between the
exoskeleton and the ground.
• A force sensor measuring the reaction between the wearer
and the exoskeleton’s foot.
• A ground contact sensor system in the exoskeleton foot
• The measurement of the torso angle a reference of
orientation. Inclinometer (tilt sensor) attached to the
midpoint of the waist is used to measure the torso angle.
Force sensor used is a cantilever-type load cell which is
designed for the operation range of human weight.
 2. Design of a Wearable Sensing System for a Lower
Extremity exoskeleton
Chunfeng Yue1, Hong Cheng1, Ye Chen2, Qinglong Deng2, Xichuan Lin1
1. School of Automation Engineering University of Electronic Science and
Technology 2. Graduate School of China University of Electronic Science and
Technology of China

This paper introduces a design for foot wearable sensing system

for the purpose of realizing intelligent control for an assistive
type lower extremity exoskeleton named AIDER. The wearable
sensing system employs 7 force sensors as a sensing matrix to
achieve high accuracy of ground reaction force detection. One
more IMU sensor is integrated into the wearable sensing system
to realize 6-DOF posture capture.

The mechanical structure of the wearable sensing system

contains 3 layers. The bottom layer is a made up of rubber to
avoid slipping, the middle layer is the sensor layer which
includes all the force sensors and the top layer is made up of
aluminium alloy which is the support for sensor installation. Top
layer is divided into two parts: the heel and the forefoot. Three
force sensors form a stable plane to sense forefoot pressure and
four force sensors form a trapezoid to bear heel pressure. IMU
sensors and processing circuit are installed in a circuit box
 3. Recent developments and challenges of
Lower extremity exoskeletons
Bing Chen a, Hao Ma b, Lai-Yin Qin c, Fei Gao b, Kai-Ming Chan a,
Sheung-Wai Law a, Ling Qin a,*, Wei-Hsin Liao b,*
a Department of Orthopaedics and Traumatology, Chinese University of Hong Kong,
Hong Kong, China
b Department of Mechanical and Automation Engineering, Chinese University of
Hong Kong, Hong Kong,
China
c Division of Biomedical Engineering, Chinese University of Hong Kong, Hong
Kong, China

To control the exoskeleton, it is essential to acquire different

types of motion data of the human-exoskeleton system during
movement. There are three types of biomechanical data
generally associated with the human motion: kinematic data
includes body posture and joint angles; kinetic data such as
human joint torque, ground reaction forces, interaction force
between wearer and exoskeleton; bioelectric data such as EMG
signals, brain signals. Measured motion data can be used to
recognize the wearer’s motion intention, analyse the motion
status and gait pattern. Different types of sensors are usually
equipped in the exoskeleton system to measure these motion
data.

Encoders, potentiometers and an inertia measurement unit

(IMU) are used to measure the kinematic data. Force/torque
sensors are used for kinetic data acquisition. On acquiring
motion data, an exoskeleton’s motion assistance can be initiated
according to the wearer’s intention.

Various sensors used in different exoskeleton:-

• In the multisensor system of CUHK-EXO, IMUs are
mounted on the backpack to obtain the wearer’s trunk
posture and the pressure sensors are connected in the
insoles and smart crutches to detect ground contact
condition. With this information, the wearer’s centre of
gravity can be calculated in real-time and the wearer’s
motion intention can be estimated by detecting the change
in motion data.
• In ReWalk exoskeleton, a tilt sensor is used to estimate the
wearer’s walking intention. The forward tilt of the
wearer’s upper body initiates the patient’s first step, and a
functional natural gait is generated with the repeated body
weight shifting.
• In the sensing system of the HAL-5 exoskeleton, EMG
sensors are attached on the wearer’s skin to detect the
extensor and flexor muscle activities of the knee and hip
joints.
• In the HEXAR exoskeleton, a muscle stiffness sensor was
developed to obtain the signal for the degree of expansion
of a muscle. The measured signals are used to detect the
wearer’s intention to control the exoskeleton.
 4. Movement Intention Prediction to Find a New
Exoskeleton Design with Light and Comfortable
Materials
Ing. Mauricio Plaza Torres ,Ing Fredy Bernal and Ing. Andrés Cifuentes
Universidad Militar Nueva Granada

In this paper study of the design of an exoskeleton and

implementation of superficial magnetic sensor and pressure
sensor to find patterns and parameters for predicting the
intention of movement is given.
Orientation estimation in human motion tracking system needs
various types of sensors. The most commonly used ones are
accelerometers, gyroscopes and magnetometers. Each of them
has varying advantages and problems. Accelerometer measures
the linear acceleration and gravity, gyroscope measures angular
velocities by integrating data over time to find orientation and
magnetometer measures the local magnetic field vector. To
avoid the problems filtering algorithm is used.

Position sensors are used to measure the distance a body has

moved from its reference and the resulting output is given as a
feedback to the control loop of the exoskeleton. The position
sensors must be able to detect the linear position and angular
position as the motion of human body can be rectilinear or
curvilinear. The limitation of the sensor (output signals contain
noise) can be eliminated by filtering algorithm. Kalman filter
algorithm is used as it can produce estimations of state variables
and uncertainties to predict the position. Offset angles are
calculated in order to overlap the sensor frame to the body
frame. Fuzzy logic is used to control the exoskeleton with the
movement intention of the user.
5. lower limb exoskeleton for experimental research on
gait control
Camila Souit, Dafne Santana Coelho, Marcelo Szylit, Franklin Camargo-Junior,
Milton Peres Cortez Junior, Arturo Forner-Cordero Member IEEE

The selection of sensor depends on the control strategy chosen.

The control strategy used in this paper for a robot which
interacts with human body is based on the relationship between
the interaction force and position (impedance control).

Sensors are needed to measure the interaction force between the

frame and the user as well as to measure the joint angle.
In this paper, the actuator module consists of an EC Maxon
motor, a harmonic reduction and bearings. The angle is
measured by a quadrature encoder fitted in the EC Maxon
motor. Extra space is not required for fixating the position
sensor. The encoder used is an incremental one (Maxon Motors
– HEDL 5540) and outputs quadrature signals which are
decoded by a quadrature decoder.

The interaction force between the frame and the user are
measured by force sensors over the frame. There are various
types of force sensors available such as piezoelectric sensors,
capacitive force sensors and piezoresistive polymers. A
commercial sensor with the required specifications that will fit
into the exoskeleton’s frame is preferred. Strain gauges are
mostly used to measure the interaction force. The exoskeleton
frame contains of two load cell and load cells consists of elastic
frame and four strain gauges coupled at full Wheatstone bridge.
 6. Design and control of hybrid actuation lower limb
exoskeleton
Hipolito Aguilar-Sierra1, Wen Yu1, Sergio Salazar2 and Ricardo Lopez

In this paper two types of actuators are used in the lower limb
exoskeleton: DC motor with harmonic drive and Pneumatic
artificial muscle (PAM). The DC motor provides with high
torque but it’s heavy. The high torque PAM actuator is light and
provides high power/weight ratio but has low control accuracy.
The shortcomings of the actuators are overcome by hybrid
actuation which takes the advantage of both the harmonic drive
and the PAM. The combination provides high accuracy of
position control and high power/weight ratio. The only
disadvantage of combination of two different types of actuators
is that control becomes complex. A special design process of the
hybrid control is considered.

The human-machine interface of the exoskeleton uses EMG

sensors. They are put on several muscles of the human lower
limb and send EMG signals to the computer in a wireless
manner. To obtain lower limb motions by muscle’s activities, 8
EMG sensors are placed on one leg. These sensors capture the
movements of the lower limb and generate a base of gait pattern.
The 8 EMG signals are filtered, recorded and sent to a
classification algorithm to generate different motion patterns. By
analysing the patterns, control commands are sent to the
actuators to move the exoskeleton. In this way the Human-
machine interface uses the patterns from muscle fibres to predict
the intention of the operator.
7. Design of a Lower Extremity Exoskeleton for Motion
Assistance in Paralyzed Individuals
Bing Chen, Hao Ma, Lai-Yin Qin, Xiao Guan, Kai-Ming Chan, Sheung-Wai Law,
Ling Qin and Wei-Hsin Liao

In this paper, the whole system of a lower extremity exoskeleton

developed by the Chinese University of Hong Kong (CUHK-
EXO) to help the paraplegics is explained. Human lower
extremity has seven DOFs for each leg with three for hip joint,
one for knee joint and three for ankle joint. Due to the weight
and space limitation, not all of the exoskeleton joints need to be
active. Among all the joints, hip and knee joints are active and
ankle joints are passive. In CUHK-EXO, pair of smart crutches
is developed to support the body weight and propulsion. The
active joints are actuated by DC motor (Maxon RE40, 150W)
through planetary gearboxes and bevel gears. The CUHK-
EXO’s motion range is large enough for the wearer to perform
the daily life motions such as STS (stand up/sit down) and walk.

The multi-sensor system of CUHK-EXO is used to obtain the

wearer’s motion intention. The hip and knee joints of CUHK-
EXO are equipped with encoders and potentiometers to get the
information about joint angles and angular velocity. An inertial
measurement unit (IMU) is mounted on the wearer’s trunk to
obtain the wearer’s trunk posture which in turn is used to detect
the wearer’s intention to STS. A pair of pressure insoles is
mounted on bottom of the CUHK-EXO feet to acquire ground
reaction force (GRF) and plantar pressure distribution in both
feet.

A pair of smart crutches with FSR sensors, IMUs,

microcontroller units (MCUs), Bluetooth modules and batteries
is designed for the exoskeleton system to provide comfortable
and stable assistance for the paraplegic patient. The FSR sensors
mounted at the bottom of the crutch provides the ground
reaction force (GRF) and the IMUs are used to acquire the
information of crutches posture which can be used to calculate
the position of crutch bottom. The step size of human-
exoskeleton can be determined from the position of crutch
bottom. The sensor information is sent to high level controller
which is a remote PC through Bluetooth module.
8. Quantifying The Human-Robot Interaction Forces
Between A Lower Limb Exoskeleton And Healthy Users
Ashish Rathore1, Matthew Wilcox1, Dafne Zuleima Morgado Ramirez2, Rui
Loureiro1, Member,IEEE

This paper presents the development of an alternative real-time

force monitoring apparatus using FSRs installed at the physical
human-robot interface (pHRI) of the lower limb robotic
exoskeleton. The human-machine interface between the user and
the exoskeleton involves multiple points of human-robot contact
at which a net flux of power generated by the exoskeleton is
transferred to the viscoelastic soft tissue of the patient through
two interfaces: connection cuffs and orthoses. The study of
interaction forces at the pHRI includes using mathematical
modelling; direct measurement using load cells and opto-
electronic sensors. However, the accuracy of mathematical
model is difficult to measure and also the single point sensors
used do not provide information on the distribution of force
across the human-exoskeleton interface.
One study developed ‘pressure pads’ to record pressure maps
for a healthy person and a SCI patient but drawback was that
integrating FSRs into a flexible pressure distributing pad may
cause excessive bending of the FSRs during applied loads which
can affect the accuracy of the result. In this paper 16 key
locations were identified to monitor and data was obtained from
10 healthy participants.
An exoskeleton REX Personal controlled by a joystick interface
is used for experiments. 16 Interlink Electronics FSRs 400 was
selected for force measurements. For the force data acquisition
Arduino Mega 2560 was used and a graphical user interface was
built in MATLAB.
Myoware Muscle Sensor Interfacing with Arduino
Myoware Muscle sensor (AT-04-001):

The Myoware Muscle Sensor from Advancer Technologies measures,

filters, rectifies, and amplifies the electrical activity of a muscle and
produces an analog output signal that can easily be read by a
microcontroller.
It operates with single power supply (+2.9V to +5.7V) with polarity
reversal protection; additional feature of this sensor is that we can
adjust the sensitivity gain.
Layout of sensor:
Parts Required :-
1. Myoware muscle sensor
2. Arduino uno
3. Soldering iron and solder
4. 22-gauge wire(3-conductors)
5. Electrodes with centre nubs to snap into myoware’s receptacles
6. A computer
Steps to be followed:-

• Find the three holes on the edge of the Myoware that are
labelled (+), (-), and SIG (signal). Solder a 22-gauge wire to
each hole. Solder the other ends to header pins (pins can be
found in an Arduino starter kit).
• Connect the myoware’s (+) wire to the +5 output header on the
Arduino.
• Connect the myoware’s (–) wire to the GND (ground) header on
the header.
• Connect the signal wire to A0 on Arduino.
• Connect Arduino to the computer via USB cable.
Placing electrodes:-

• Add electrodes to all 3 connectors on the sensor. Remove the

paper backing of the two electrodes on the circuit board.
• Clean the skin with rubbing alcohol to remove dirt, oil, or lotion.
• Put the circuit board in place. The side of the circuit board with
the wires should be close to the centre of the muscle and the
other side should be close to the end of the muscle.
• Peel the paper backing off of the electrode on the black wire and
place it somewhere away from the muscle you are sensing.
Arduino with DUINOPEAK ECG EMG Shield for
1 channel

➢Requirements:-
• DUINOPEAK ECG EMG Shield
• Arduino UNO
• Electrode cables ( SHIELD-EKG-EMG-PRO)
• ECG Gel electrodes
• USB cable
➢DuinoPeak ECG EMG Shield layout and
description:-

This is an ECG EMG Shield which allows for Arduino like boards to
capture electrocardiography (ECG) and electromyography (EMG)
signals. This shield allows to experiment with bio-feedback signal.
One can monitor the heartbeat; recognize gestures by monitoring and
analysing the muscle activity.

Powering up the board:

The DuinoPeak ECG EMG shield is powered by the host board it is

mounted on. There is an option to power either by 3.3V or 5.0V host
board (configured easily by a jumper). On powering the board PWR
LED must become RED.
ECG EMG shield description

1. Arduino shield connectors:- (CON1,CON2,CON3,CON4)

• These connectors follow the ARDUINO specification for

shield connection. The shield comes with soldered connectors
making it ready for mounting on compatible board.
• 6-pin (CON1 & CON2) and 8-pin (CON3 & CON4)
connectors mounted:

2. Trimmer TR1:

• It is a precise trimmer potentiometer for calibration and

it may be adjusted for gain.

3. Jumper description:

• 3.3V/5V:- This jumper controls the power circuit

whether powered by 3.3v or 5v. Default state is 3.3v.

• REF_E :- The position of the REF_E jumper depends

on the "host" board. If the "host" board provides voltage
on the AREF pin of the digital connector, REF_E has to
be open. If there is no voltage provided on the AREF
pin then REF_E jumper has to be closed. Default state
is closed.
• AIN_SEL:- This jumper is responsible for which
channel ECG EMG Shield would utilize. Default state
is in position 1. AIN_SEL is the analog pin where the
shield would send the data it acquired. If you have two
or more shields stacked - each should have AIN_SEL
set to different position.

• D4/D9:- This controls the pin D4 and D9. Default state

is D9. D4/D9 jumper controls which digital pin the
shield should use for acquisition of calibration signal
from the main board. If your main board uses D4 for
other purposes, then change it to D9, and vice versa.

• CAL:- used for feedback of the calibration and requires

additional cable. Default state is open.

Front view back view

➢Procedure:-

1. Download and extract the arduino IDE package from the

arduino website.
2. Download and place the two libraries required (TimerOne,
FlexiTimer2) in the arduino libraries.
3. Download the arduino demo code for ECG EMG Shield
and FTDI VCP drivers.
4. Jumpers of the shield are set as :
o REF_E –closed
o 3.3V/5V-5V position
o D4/D9-D9 position
o ANI_SEL-1 position(channel 1)
5. Connect the shield to the arduino board and also connect the
arduino to the USB.
6. Install the VCP FTDI drivers by going in device manager.
Also go to device manager and from advanced settings of the
recognized serial port (COMx) device set x to a free port
between 1 and 4 because the monitoring software used
(Electricguru TM) can read only from port1 to port4.
7. Start arduino IDE and open the demo code and
Set Tools->Board->arduino/genuine uno
Set Tools->Serial port->the COM we configured our board at
and click upload (->)
8. Download, install and start the monitoring software
Electricguru TM. Adjust the settings in electricguru TM:
Preferences->trace (waveform)->channel 1 (depending on
how many channels/shields we are using)
9. Choose the COMP port to which the arduino is connected to
from preferences->serial port…
10. Placement of electrodes:

Attach ECG Gel electrodes to the electrode cable.

White: centre of the muscle
Red: end of muscle, near joint
Black: reference point such as bone or irrelevant muscle
Attach electrodes to the skin by removing plastic and
pressing firmly on clean skin. Make sure there is still wet gel
for good contact.
 11. Plug in the electrode cable to the ECG EMG Shield arduino
to the PC.
12. Make sure to unplug laptop charger before plugging in the
arduino with electrodes on the skin.
13. Start monitoring!

➢Hardware setup:-
To record EMG signals using Grove-EMG
Detector and arduino

➢Requirements:-
• Grove EMG Detector
• Grove cable
• On-Off electrode
• DC jack to button connector cable 1000mm
• Arduino UNO
➢EMG Detector board overview:-

EMG Sensor v1.1

EMG Detector is a sensor which can detect and gather the small
muscle signals. Also, it amplifies and filters the signal. The output
signal can be recognized by arduino.

In standby mode, the output voltage is 1.5V. When active muscle is

detected, the output signal rise up, the maximum voltage is 3.3V. We
can use this sensor in 3.3V or 5V system.

• J2: grove interface, connect to analog I/O

• J1: EMG Disposable Surface Electrodes connector
• U1: INA331IDGKT, difference amplifier
• U2, U3: OPA333, Zero drift amplifiers
➢Procedure:-
1. Connect the GNDA pin of EMG Sensor to the GND of
arduino.
2. Connect the VCC pin of EMG Sensor to 3.3V on arduino.
3. Connect the VOUT pin of EMG Sensor to the A0 of arduino.
4. Placement of electrode :
Attach the on-off electrodes to the electrode cable.
White: centre of the muscle
Red: end of muscle, near joint
Black: reference point such as bone or irrelevant muscle
Attach electrodes to the skin by removing plastic and
pressing firmly on clean skin. Make sure there is still wet gel
for good contact.
 5. Plug in the electrode cable to the Grove EMG Sensor and
arduino to the PC.
6. Make sure to unplug laptop charger before plugging in the
arduino with electrodes on the skin.
7. Code used is : upload the following code and start monitoring

➢Hardware setup:-

➢Result:-
• Open tools -> serial monitor in the arduino coding application
to see the voltage readings streaming in.

• Open tools -> serial plotter in the arduino to see the raw
EMG waveform from the sensor.
Extension and flexion of arm:-

Extension of muscle

Flexion of muscle
View publication stats