Page 96 - ITU Journal Future and evolving technologies Volume 2 (2021), Issue 3 – Internet of Bio-Nano Things for health applications
P. 96
ITU Journal on Future and Evolving Technologies, Volume 2 (2021), Issue 3
cortex of the occipital lobe can enable the user to depending on the desired frequency range the filter
spell out the visualized words and letters as shown setting changes to bandpass filter. For instance, to
in [15]. However, to deliver such precise find out cognitive functioning EEG data, electrodes
commands in any system, deep learning of the data are attached to the frontal lobe, where to get rid of
is imperative. As associated in [13], the machine external electrical influence a notch filter is used. In
learning algorithm of deep learning concentrates addition, a bandpass filter of 5-50Hz is used, which
on the advancements of precision-control in is the brain activity wave except for the delta which
automated systems. This can benefit the obtained is experienced during meditation or dreamless
feedback generation of our paper to incorporate sleep and the amount of sample rate varies
concrete real-life applications in the future. depending on the outcome expectation but mostly
the used sample rate is 255Hz [18]. Moreover,
maximum frequency can be opted for according to
the preference which could be from 20Hz to 400Hz.
Again, to make the data transmitted to Arduino or
Python workable mapping of the data is required.
Moreover to that, mapped data is used in the
completion of tasks or controlling a system.
5. RESULT AND DISCUSSION
Continuing on from the previous discussion,
motion control of a car with the assistance of
Arduino has been successfully implemented by
utilizing the acquired brainwave data exclusively.
The Arduino-based robotic car has been through
multiple trials for motion control where the brain
wave signals were used to enforce the focus and
concentration state on a real-time basis [20].
Looking at Fig. 3, we can see that 3(a) displays the
bio potential of the brainwave in the frequency
domain while 3(b) and 3(c) display the head region
activity at the instances. In Fig. 3(a), 2 channels
were attached to the frontal lobe while acquiring
concentration state data. In 3(a) the FFT plot of the
potential values are majorly active in the alpha
(8-12 Hz) and beta (13-32 Hz) frequency ranges.
This information implies that the focus state is
Fig. 2 – Experimental feedback generation activated in the aforementioned bandwidths and
the car gains motion correspondingly. Thus, the
4.3 Parameter settings motion and direction of the car can be controlled
with concentration. However, the car immediately
In Section 4.1 and Section 4.2, hardware stops moving the very moment the focus state is
implementation and software control departed. Furthermore, the head region plots 3(b)
implementation were observed by utilizing the and 3(c), display high activity in the frontal and
data streaming of an EEG reading from the person. central lobe. The highlighted regions of the
Correspondingly, the EEG data generated from respective lobes are distinctly represented in the
mostly the parameter setting depends on the event of cognitive and motor functionalities.
pathway taken to filter the signal. Most of the time
84 © International Telecommunication Union, 2021
