Detection of Dangerous Situations Using a Smart Internet of Things System - New Contributions in Information Systems and Technologies - page 368

The IETF and its working groups worked on 6LoWPAN technology and Core
(Constrained RESTful Environments), which together, enable IPv6 with the 802.15.4
network providing access to a Web Service via CoAP (Constrained Application
Protocol). CoAP is a simple network oriented to Web Services REST.
The IoT application will be able to function in real time in order to provide better
surveillance of the child and to give a better assistance in case of detection of a
dangerous situation.

3.2 Application Architecture

The application architecture is all based on data collected. As illustrated in Fig. 2., the
WSN will read the biometric records of the child that may indicate a situation of fear
or illness. Those values will be compared with the values established as standard for a
child. If the data collected are different of the standard range, then those values may
be dangerous and will be recorded in database.
The validation of a state of danger is made taking into account a self-learning algo-
rithm that uses the calculation of the threshold, the average of each sensor of the
wearable device and consults the database of the system.
The calculation of a threshold is used to create standard values accordingly to each
child, because each person has unique standards of physiological signals, what is
dangerous for a child may not be dangerous for another child. Applying this principle,
the algorithm will be self-learning and will be always adjusting its threshold values
according to the time of the day and the child who is wearing the device.
The calculation of the average of each sensor is made for determine if the child is
in a dangerous situation. This is a weighted average that has a different equation
according to the situation of fear or illness.

Fig. 2. Application Architecture Proposal