Development of EEG measurement and

processing system in LabVIEW development

environment

International Review of } BARSY1, GYULA GYORI

BENCE GERGO } 2 and
Applied Sciences and 
PETER  SZEMES p
TAMAS 1
Engineering
1
11 (2020) 3, 287–297 Mechatronics Department, Faculty of Engineering, University of Debrecen, Debrecen, Hungary
2
DOI: PHARMAFLIGHT International Science and Service Center Private Limited Company,
10.1556/1848.2020.00151 Debrecen, Hungary
© 2020 The Author(s)
Received: July 7, 2020 • Accepted: October 9, 2020
Published online: October 29, 2020

ABSTRACT
ORIGINAL RESEARCH
Our research team has developed a system and methodology for measuring psycho-physiological pa-
PAPER rameters, which can be used to determine the level of fatigue and fitness of the person being measured.
This article describes the electroencephalography (EEG) part of this system. This article covers the
technical and mathematical background of EEG measurement, the selection and implementation of the
measurement tool in the development environment, and the development of the measurement and
processing algorithm. The result is a system that can detect, digitize, and process the digitized signal
from the brain, and save the processed signal in an XML database.

KEYWORDS
EEG, LabVIEW, XML, FFT, adatbazis, Muse, UDP protokoll, Bluetooth low energy, OSC protokoll

1. INTRODUCTION [1]
Electroencephalography (EEG) is a non-invasive measurement procedure for detecting
electrical signals from the brain and then processing these signals. With the help of signals
from the brain, we can determine the current state of the human mind and track the change
to certain external stimuli from the ground state. Fixing the baseline is a critical point for
EEG measurements, since every person has a different datum elsewhere, it is not possible to
add a single datum for each person. Compared to the correctly recorded baseline condition,
we can draw conclusions that the subject’s mental state has deteriorated or improved as a
result of certain stimuli. Deeper, more comprehensive analysis requires serious neurological
knowledge, but simpler, but more important analyses can be performed by simple mathe-
matical calculations.
For example, surveying the mental health of those involved in extreme work is an area of
application that has shown rapid development and need for further research. This includes
performing the aforementioned resting measurement, which is important for the subject to
be calm and relaxed. After resting, it is possible to re-examine the physical or mental subject
to see how your brain responded to the exercise. Typically, these jobs include pilots, fire-
fighters, air traffic controllers, ambulances.
The purpose of this research is to create a system that can quantify the fatigue and fitness
*Corresponding author.
status of a person and store this data in a database. The complete system includes an ECG
E-mail: [email protected]
gauge, an EEG gauge, a vascular stiffness gauge, a muscle oxygen level gauge, and a skin
temperature gauge. The data processed must be credible in order to provide an expert with
relevant data.
This paper is part of a larger research project, further results will be published in [18, 19].

Unauthenticated | Downloaded 10/21/23 03:59 PM UTC

288 International Review of Applied Sciences and Engineering 11 (2020) 3, 287–297

Fig. 1. EEG electrode mounting standard [3]

2. TECHNICAL AND MATHEMATICAL BASIS

OF EEG MEASUREMENT
2.1. Technical background of EEG measurement
The non-invasive EEG measurement procedure places low-
resistance metal macroelectrodes on the scalp according to Fig. 2. Bipolar and unipolar EEG measurement [4]
an international system. This system selects 4 anatomical
reference points on the head and places the electrodes at 10
and 20% distance between these reference points, respec- amplification) or display. While we know that we can draw
tively. However, it is now more common to use the 10–10 conclusions from the temporal display of brain waves, such
system, which requires electrodes to be positioned 10% from as epilepsy, blinking, etc., we need to perform further oper-
the reference point. The measuring system always measures ations on the already transformed digital signal for deeper
the potential difference between two points [2]. evaluation [7].
The names of the electrode locations are derived from There are many processing methods available to analyze
anatomical expressions, with F beginning with the frontal, EEG signal for different purposes, such as medical exami-
i.e., the front, P with the parietal, i.e. the back of the head, C nation of Brian Computer Interface (BCI) [16]: Artificial
starting with the central, i.e. the middle, T the temporal, i.e. Neural Networks (ANN), Support Vector Machines (SVN),
at the ear and O at the occipital (Figs. 1 and 2). and PCA, Principal Component Analysis. The above-
From the point of view of the measurement principle, mentioned methods are found to be very useful to extract
two types are distinguished: special features from noisy signals. Our research aim is to
A. Bipolar measurement is the measurement of the poten- find low computation and direct (simple) method to extract
tial difference between two electrodes. features, what is proportional to physical and psychological
B. Unipolar measurement, where the potential difference fatigue, we established our method in time domain, and
between two electrodes is also measured, but one of statistical analysis.
which is always an inactive reference electrode.
2.2. Mathematical background of EEG measurement
The electrodes used for measurement can be divided into [9]
two groups, distinguishing between dry and wet electrodes.
For smaller portable weighing systems, a dry electrode is During the EEG measurement, the digitized waveform from
generally used, and for multi-channel systems, a wet elec- the measuring system is processed for further evaluation for
trode is preferred. The dry electrode has a gold plating fitness/fatigue evaluation. If the signal is not properly
which, once applied, can be used immediately and provides filtered, additional noises may affect the sensitivity of the
good contact with the skin surface. It does not require any fitness/fatigue evaluation.
professional qualification [5]. The Butterworth filter is the most used filtering type in
The signals taken from the skull using the electrodes this discipline because the cutting characteristics of this filter
should be filtered and amplified, usually by the measuring are the best to avoid distorting the signal after filtering so
device on the head, but if necessary, the program can further that no information is lost during processing. According to
refine the amplification and filtering. Doing these is key to Butterworth transfer function (1), this type of filter rejects
signal processing because we can only process properly unwanted frequencies, and provides unified gain over the
noise-free and amplified signals so that we can extract useful bandwidth of wanted frequencies.
information from them [6]. The properly filtered time domain signal must be con-
Fig. 3 illustrates the structure of a classical EEG system verted to a frequency domain so that each of the notable
that does not perform any pre-processing (filtering, brain frequency bands can be determined. The method used

Unauthenticated | Downloaded 10/21/23 03:59 PM UTC

International Review of Applied Sciences and Engineering 11 (2020) 3, 287–297 289

Fig. 3. Generic EEG system block diagram [8]

for this is the Fast Fourier Transform, or FFT, which is a Fast Fourier Transform (FFT) is a (fast) algorithm that
faster version of the discrete Fourier transform. can be used to calculate DFT. The following equation gives
the DFT transformation formula:
1. Butterworth filter  
Butterworth filters are of the type IIR, or Infinite Impulse X
N −1 −j 2nik
N

Response filters. Butterworth filters have the following X½k ¼ x½i*e k ¼ 0; 1; 2 . . . N  1: (2)
i¼0
characteristics:

Damped amplitude function at all frequencies, where: X [n] values are the time domain values of the

The amplitude curve decreases monotonically from a sampled signal, N is the number of sample values.
given cut-off frequency, The FFT and the power spectrum are very useful for

Maximum flatness, unit response in transmission band measuring the frequency content of steady and transient
and zero in crop band, signals. The FFT gives an average frequency content over the

Half power frequency or 3 dwindling frequency is related entire measurement range. Therefore, FFT is useful when
to cutting frequency, the signal is steady state or when an average power content
is required at each frequency. The frequency spectrum
The great advantages of Butterworth filters are their
smoothness and monotonically decreasing frequency func-
tion, so they do not distort the signal even when properly
filtered (Fig. 4).
1
HðjuÞ ¼ sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2Nffi (1)
1 þ «2 uup

where: N is the filter order number, u 5 2pf is the frequency

variable, uP is the cut off frequency, « is the maximum gain
and H is the filter response (transfer function in the complex
frequency domain).
1. FFT algorithm [10]
FFT, or fast Fourier transformation, is essentially a faster
version of a classic Fourier transformation. In the direct
implementation of the discrete Fourier transform, we need
to perform approximately N2 complex operations on N
samples, which is a very time-consuming calculation oper- Fig. 4. Butterworth filter Bode diagram with different order
ation. numbers [9]

Unauthenticated | Downloaded 10/21/23 03:59 PM UTC

290 International Review of Applied Sciences and Engineering 11 (2020) 3, 287–297

obtained from the EEG signal can be classified into the The device is capable of real-time monitoring of brain
notable frequency bands shown in Table 1: waves on 4 EEG channels with 256 [Hz] sampling rate and
12 [bit] resolution, and a 3-axis accelerometer and gyroscope
are included to estimate head pose and movement. Pose and
3. SELECTING THE EEG MEASURING movement information will be useful later in the measure-
INSTRUMENTATION [11] ment as motion can cause disturbance in the signal, so if the
EEG waveform and the accelerometer are both jumping and
disturbed at the same time. Source of disturbance signal
The development procedures started with the selection of
caused by the movement and jerking.
proper EEG device that is properly authenticated, can
Fig. 6 clearly shows that the device uses 4 channels, two
communicate wirelessly and is easy to install to patient and
on the forehead and two behind the ear, with the reference
use by non-technical experts as well. There are many EEG
electrode located in the center of the forehead (Fpz). This 4-
metering systems on the market for sport, meditation, and
channel EEG measurement system proved to be sufficient
medical applications, in variable price ranges. Our selection
(in technical and financial terms) for this project, and it is
was the Muse 2016 EEG instrument, which can detect EEG
signals with the same accuracy as a laboratory instrument, possible to extract the data needed to assess the subject’s
state of mind [13]. Bluetooth wireless connection provides a
and comes with a manufacturer-supplied Application Pro-
stable connection in our office and laboratory use, for our
gramming Interface (API) for quick integration into any
measurement, but it should be noted that it may not be
development environment and facilitating the development
suitable for all possible application areas. Wireless connec-
process [12].
tion may be disturbed by other wireless devices utilizing the
Muse EEG is a 4-channel EEG measurement device
same 2.4 GHz ISM band. Cooperation with other Bluetooth
capable of delivering real-time data of EEG signals via
health care devices, such as smart watches or sleep monitors,
wireless connection (Bluetooth Low Energy 4.2.) (Fig. 5)
The technical parameters of the Muse EEG are shown in was not tested.
Table 2.
4. MUSE EEG IMPLEMENTATION IN LABVIEW
Table 1. EEG frequency bands and their characteristics DEVELOPMENT ENVIRONMENT
Name Frequency Features
4.1. Pairing Muse EEG with your computer
d-delta Less than, 4 Fastest and strongest brain wave,
Hz very low frequency, high amplitude. A utility called Muse Direct needs to be installed from the
Usually deep sleep or sleepless Microsoft Store to connect the device to your computer. This
sleep. It is also found in deep application communicates with the device via a Bluetooth
meditation. connection, and then transmits the incoming data to the
q-theta 4–8 Hz Drowsiness, light sleep, deep processing program as an OSC (Open Sound Control)
relaxation or meditation. stream, UDP (User Datagram Protocol). Please note that
a-alpha 8–12 Hz Relaxation, tranquility or peace. Bluetooth was the lower level protocol and TCP/IP protocol
b-beta 12–30 Hz Normal waking awareness, high
was the higher level protocol, where UDP was applied (Fig. 7).
alert, active thinking, anxiety or
focus. Most people work in this
Clicking on the Bluetooth slider will start exploring and
band during the day. connecting to nearby Muse devices. After connection, you
g-gamma More than 30 Fastest brainwave. Hyperactivity or will receive information on signal quality and battery status.
Hz processing of information from
different brain areas.

Table 2. Muse EEG technical parameters

Wireless connection BT 4.2 BLE
EEG channels 4 Measurement Channels
256 Hz sampling rate
12 bit/sample
Reference electrode FPz (CMS/DRL)
position
Channel electrode TP9, AF7, AF8, TP10 (dry)
position
Battery Time Maximum 10 hours
(rechargeable Li-Ion)
Accelerometer Three-axis 52 Hz, 16 bit resolution,
range ±4 g
Gyroscope ±1,000 8/s
Fig. 5. Muse EEG [11]

Unauthenticated | Downloaded 10/21/23 03:59 PM UTC

International Review of Applied Sciences and Engineering 11 (2020) 3, 287–297 291

Fig. 6. Muse EEG distribution of electrodes [1]

Fig. 7. Pair Muse EEG with Muse Direct [own figure]

Communication must be configured to provide Lab-

VIEW with processable data (Fig. 8).
During configuration, you need to enter your computer’s
own IP address through any port, in this case 7,000, and
then choose which data to provide.

4.2. Implementing Muse EEG in a LabVIEW

development environment
Implementing Muse EEG in the LabVIEW Developer
Environment by completing the OSC data stream, LabVIEW
performs post-filtering of data, graphs and graphs raw and
calculated values, and saves measurement and configuration Fig. 8. Configure an OSC stream [own figure]
parameters in a .csv and .xml file extension.
1. Read OSC stream
packets. Muse Direct sends the data according to the
The OSC stream is transmitted through a UDP protocol following encoding:
to a processing program like the TCP/IP network protocol,

/notch_filtered_eeg: raw EEG from which 50 Hz network
except that the sending party does not wait for confirmation
noise is filtered
from the client whether it has received the data packet. This

/acc: accelerometer data
method provides faster communication than TCP (Trans-

/gyro: gyroscope data
mission Control Protocol) but may result in data loss for a
less resource intensive client. For this reason, the client, the LabVIEW can read these packages using an installable
processing program, must run on a powerful machine to directory by setting the appropriate parameters and ad-
minimize the time difference between iterations and to avoid dresses. Inbound data is merged into a waveform, trans-
packet loss. TCP packet waits for acknowledgment sent by formed from a quasi-digital signal to an analog signal [14].
receiver, so this wait may overload transmission buffers. Fig. 9 shows that the program first connects to the
A feature of an OSC data stream is that it sends data in appropriate UDP port, but if no data is exchanged for 3
packets, with different addresses and arguments in the seconds, the program will stop (and generates exception)

Unauthenticated | Downloaded 10/21/23 03:59 PM UTC

292 International Review of Applied Sciences and Engineering 11 (2020) 3, 287–297

1. Calculation of spectral analysis for EEG channels

The cleaned data set is then converted from a time
domain to a frequency domain using an FFT algorithm and
then plotted using a graph to measure the average spectrum.
According to our fitness/fatigue evaluation protocol we
developed, a measurement lasts in quiet and calm environ-
ment for 6 minutes and is divided into 3 and 3 minutes.
According to our investigation, it was suitable to provide
information on the trend for the first and second half of the
measurement [15] (Fig. 11).
The program calculates FFT for a total of 10 arrays, 4
channels for the first measurement, 4 channels for the sec-
ond half of the measurement, and first and second half av-
erages.
During the FFT calculation, the program also calculates
the power of each frequency range, i.e. the area under the
FFT graph. The area under the curve is used to determine
the asymmetry between the right and left hemispheres.
Obviously, the hemisphere that has higher average power at
Fig. 9. OSC streaming reader and visualization program detail [own the time of measurement is the more dominant hemisphere.
figure] Fig. 12 shows the frequency bands, which are marked
with a separate color (Figs. 13 and 14).
due to timeout. Once the connection is established, the data
stream starts according to the OSC address table, which
provides EEG, gyroscope, and accelerometer data from the
device.
1. EEG signal filtering by Butterworth filter
The implementation of the filter in the program was
inevitable as only the 50 [Hz] utility network components
are physically filtered by the device from the input data, so
the higher frequency interfering components are still present
in the signal. To filter out low frequency component dis-
turbances, such as blinking, muscle twitching, accidental
movement, the parameters of the filtering algorithm were set
to pass the components between 0.5 [Hz] and 35 [Hz]. With
this EEG data bandpass filter, higher frequency components
can also be filtered out.
Based on the results of some research [17], the optimal
number for filtering the EEG is 2, which can be used to
safely filter out interfering signals without distorting the
signal. 0.5 Hz was selected because our measurements were
done in a quiet and calm laboratory environment (Fig. 10).
The input to the filter is the array read from the OSC
stream, and the output is the filtered signal that can then be
processed.

Fig. 10. Bandwidth Butterworth Filter Program Detail [own figure] Fig. 11. FFT algorithm program snippet [own figure]

Unauthenticated | Downloaded 10/21/23 03:59 PM UTC

International Review of Applied Sciences and Engineering 11 (2020) 3, 287–297 293

3u.csv”. Since the date and time are saved, plus the seconds
to 3 decimal places, it is possible to accurately differentiate
and identify each measurement file.
After the measurement is started, the program creates an
XML database in which the configuration settings are stored.
The most important of these are the serial numbers of the
devices, the name of the person performing the measure-
ments, the resolutions of the devices, the sampling fre-
quency, the path of the measurement file created after the
measurement, the number of channels and the minimum
Fig. 12. FFT graph for EEG signal [own figure] and maximum values. This database can be used to validate
the output of the measurement in terms of configuration
parameters (Fig. 15).

5. TEST MEASUREMENT
Measurement and testing of EEG signals was carried out
with the assistance of a team of medical experts, including
helicopter pilots. Measuring EEG signals first and foremost
requires proper placement of the Muse, as this is the only
way the device can provide valuable, processable signals. The
device’s data sheet states that it works with a dry electrode
system, but for a perfect contact, it is advisable to moisten
the surface of the electrode with a slightly damp or blue
spirit cotton pad (Figs. 16 and 17).

Fig. 13. Right/left hemisphere asymmetry calculation part [own

figure]

Fig. 15. Configuration XML database [own figure]

Fig. 14. Visualization of cerebral hemispheres asymmetry for the

first and second half of the measurement [own figure]

1. Saving measurement data

At the end of the measurement, the measurement data is
saved with headers and units in the .csv file extension. This
format is selected because this extension can be interpreted
by several popular data processing and display programs, so
we have more options for re-measuring.
The filename composition applies the following system: Fig. 16. Professional medical assistance: Placing Muse on the test
“Muse_measurement_% Y-% m-% d. % H:% M:% S% subject. The measurement was done with ECG measurement.

Unauthenticated | Downloaded 10/21/23 03:59 PM UTC

294 International Review of Applied Sciences and Engineering 11 (2020) 3, 287–297

Before starting the measurement, the subject should be Measurements are tracked using graphs that transmit
asked to close his eyes and move as little as possible, as these measured data in real-time to provide an accurate view of
can cause interfering signals in the collected signals, which the measurement process. You can display more than one
could greatly distort the measurement output. graph on the graph at a time, so you can look for correla-
The installation was in all cases done by an expert per- tions between the waveforms of some measured values
son, continuously monitoring the data collected by the de- (Fig. 22).
vice during the measurement. During the measurement the The report calculates the data computed by all programs,
experts found everything was fine, and any fine-tuning was the power, relative and absolute distributions of each fre-
carried out on site (Figs. 18 and 19). quency range, and the percentage distribution of the right
During the test measurement, the gyroscope and the and left hemispheres, all of which appear on the expert panel
accelerometer proved to be useful as these sensors can be for the first and second half of the measurement (Figs. 23–
used to detect whether motion was caused by a disturbance 25).
signal (some disturbing signals such as motion, jerking) or For the sake of clarity, the first and second halves of the
other external sources, interference (Figs. 20 and 21). measurement were distinguished in the hemispheric dia-
gram by two colors, so that they can be clearly distin-
guished. And displaying a chart helps you keep track of the
trend.

6. CONCLUSIONS
In this paper, we presented a technical method and evalu-
ation logic to measure physical and mental fitness/fatigue.
Commercially available, Muse EEG device was used for EEG
measurement and LabView environment was used to
develop evaluation logic. Butterworth filter was applied to
remove unwanted noise from signal, then FFT was applied
to convert time domain to frequency domain. The signal was
classified according to common frequency bandwidths of
brain signals: Delta, Theta, Alpha, Beta and Gamma. The
area covered by FFT signals was also measured to decide the
dominant left or right hemispheres.
In the end, the system is working as expected, all
required functions have been successfully implemented. The
Fig. 17. ECG measurements was done in calm and quiet laboratory system is stable, even for long-term continuous measure-
to minimize external disturbance may has effect on patient’s mind ments. This paper is part of a larger research project, further
state results will be published in [18, 19].

Fig. 18. Screenshot of test measurement with gyroscope on bottom graph

Unauthenticated | Downloaded 10/21/23 03:59 PM UTC

International Review of Applied Sciences and Engineering 11 (2020) 3, 287–297 295

Fig. 19. Test Measurement Screenshot - EEG waveform

Fig. 20. Test measurement screenshot and EEG waveforms

Fig. 21. Test Measurement - EEG and accelerometer waveforms

Unauthenticated | Downloaded 10/21/23 03:59 PM UTC

296 International Review of Applied Sciences and Engineering 11 (2020) 3, 287–297

The system offers a number of enhancements that can be

used to extract even more data from the measurement data
set:

Application of Deep/machine learning algorithms to
classify multiple measurements

Creating a correlation matrix for the measurement pa-
rameters

Prediction of mental illness

Fig. 22. EEG figures for a 24-year-old male in the report

7. ACKNOWLEDGMENTS
The authors would like to thank every medical and technical
expert of PHARMAFLIGHT International Science and
Service Center PLC for their continuous support during the
research activity. The origin of the research topic was initi-
ated by Mr. Gyula Gy}ori.

REFERENCES

[1] A. Asif, M. Majid, and S. M. Anwar, “Human stress classification

Fig. 23. EEG spectrum of a 24-year-old male – first half measure- using EEG signals in response to music tracks,” Comput. Biol.
ment Med., vol. 107, pp. 182–96, 2019.
[2] M. Seeck, L. Koessler, T. Bast, F. Leijten, C. Michel, and C.
Baumgartner, “The standardized EEG electrode array of the
IFCN,” Clin. Neurophysiol., vol. 128, pp. 2070–7, 2017.
[3] T. Ahmad and V. Ramachandran, “Hyperspherical manifold for
EEG signals of Epileptic Seizures,” J. Appl. Math., vol. 2012, p. 22,
2012. https://doi.org/10.1155/2012/926358.
[4] J. Ahmed, Brain Machine Interface Using EEG Sci-fi to Reality
Neural Interface Engineering Brain Machine Interface Using EEG 1
Brain Machine Interface Using EEG, 2016.
[5] M. A. Lopez-Gordo, D. Sanchez-Morillo, and F. P. Valle, “Dry
EEG electrodes,” Sensors, vol. 14, pp. 12847–70, 2014.
[6] I Szirmai and A. Kamondi, “Kognitıv zavarok vizsgalata EEG-vel,”
Fig. 24. EEG spectrum of a 24-year-old male – second half of (Hungarian) Ideggyogyaszati Szemle, vol. 64, no. 01–02 2011, pp.
measurement
14–23, 2016.
[7] M. X. Cohen, “Where does EEG come from?,” Trends Neurosci.,
vol. 40, no. 4, 2017. https://doi.org/10.1016/j.tins.2017.02.004.
[8] V. Sir, A Book of Diagnostic Medical Instrumentation, 2017.
[9] D. L. Gy€orgy, Jelfeldolgozas es szamıtogepes ir
anyıtas, 2012.
[10] D. N. Istvan, Gyakorlati diagnosztika, FFT analızis, 2013.
[11] I. Inc, Muse, InteraXon Inc., [Online]. Available: http://www.
choosemuse.com/. [Hozzaferes datuma: 6. okt ober 2018.].
[12] E. Ratti, S. Waninger, C. Berka, G. Ruffini, and A Verma,
“Comparison of medical and consumer wireless EEG systems
for use in clinical trials,” Front. Hum. Neurosci., vol. 11, p. 398,
2017.
[13] L Ciechanowski, M Magnuski, and P Gloor, Muse Headband:
Measuring Tool or a Collaborative Gadget?, Springer International
Publishing, 2018.
[14] N. Instruments, Open Sound Control Library, National In-
struments, 2019, [Online]. Available: http://sine.ni.com/nips/cds/
view/p/lang/hu/nid/211428 [Hozzaferes datuma: 19 October,
Fig. 25. Bar graph of 24-year-old male cerebral hemispheric rate 2019].

Unauthenticated | Downloaded 10/21/23 03:59 PM UTC

International Review of Applied Sciences and Engineering 11 (2020) 3, 287–297 297

[15] N. Instruments, Ni Biomedical Toolkit, National Instruments, Trans. Biomed. Eng., vol. 60, no. 4, pp. 1100–10, April 2013.
[Online]. Available: http://sine.ni.com/nips/cds/view/p/lang/hu/ https://doi.org/10.1109/TBME.2012.2215960.
nid/211023 [Hozzaferes datuma: 18 October, 2018]. [18] D. Torma, Gy. Gy} ori, and K. Sarvajcz, “Processing and analyzing
[16] M. R. Lakshmi, T. V. Prasad, and V. C. Prakash, “Survey on psychophysiological data using NI DIAdem software,” Int. Rev.
EEG signal processing methods,” IJARCSSE, Int. J. Adv. Res. Appl. Sci. Eng., IRASE, vol. 11, no. 2, pp. 188–96, 2020. https://doi.
Comput. Sci. Software Eng., vol. 4, no. 1, 2014, ISSN: 2277 org/10.1556/1848.2020.20128.
128X. [19] D. Dezso, G. Gyori, and H. Giza Husi, “Development of ECG
[17] H. Higashi, and T. Tanaka, “Simultaneous design of FIR filter measurement and processing system in labview development
banks and spatial patterns for EEG signal classification,” IEEE environment,” IRASE, Under Publication.

Open Access. This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International License (https://creativecommons.org/
licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited, a link to the CC
License is provided, and changes – if any – are indicated. (SID_1)

Unauthenticated | Downloaded 10/21/23 03:59 PM UTC