INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 197
www.rsisinternational.org
Smart Water Safety for Children: IoT-based Monitoring and
Emergency Alert System
Irda Roslan
1
, Wong Joe Xien
1
, Ariff Idris
1
, Khadijah Wan Mohd Ghazali
1
, Mohd Faeez Alwi
2
1
Faculty Technology Maklumat dan Komunikasi, University Technical Malaysia Melaka, Malacca,
Malaysia
2
CTRM Aero Composites, Malacca, Malaysia
DOI:
https://dx.doi.org/10.47772/IJRISS.2025.910000016
Received: 28 September 2025; Accepted: 03 October 2025; Published: 01 November 2025
ABSTRACT
Drowning is a leading cause of accidental deaths among children, largely due to insufficient supervision and
delayed emergency responses. This paper presents the development of the Smart Water Safety for Children
system, an Internet of Things (IoT)-based prototype designed to enhance child safety during aquatic activities.
The proposed system integrates multiple sensors including a MAX30100 pulse oximeter, a water-level sensor,
and a NEO-6MV2 GPS module, all managed by a NodeMCU V3 microcontroller. These components collect
real-time physiological and positional data from the child, which is transmitted via Wi-Fi to a Firebase cloud
database and visualized through a mobile application developed on Android Studio. The application provides
guardians with real-time updates on heart rate, oxygen saturation, and location, issuing alerts when critical
thresholds are detected. Functional and physiological testing confirms the system's reliability in identifying near-
drowning scenarios and effectively notifying emergency contacts. This system demonstrates a cost-effective,
scalable, and accessible solution for enhancing aquatic safety among children.
Keywords- IoT, near-drowning detection, real-time monitoring, emergency alert, child safety
INTRODUCTION
Drowning ranks among the top ten causes of accidental deaths worldwide, with an estimated 372,000 fatalities
occurring annually. According to the "National Drowning Report" by Royal Life Saving Australia, drowning
incidents rose by 20% between 2020 and 2021. Several factors contribute to drowning, such as pre-existing
medical conditions like heart issues, substance abuse, and lack of swimming skills. Additionally, the primary
cause of drowning-related deaths in children is insufficient supervision. The 2014 Global Report on Drowning
highlights that children aged 1–4 years are at the highest risk due to lack of adult oversight, followed by those
aged 5–9 years [1].
In general, there are five distinct phases in drowning. The initial phase includes a brief attempt to catch one’s
breath, followed by the second phase, where the individual holds their breath to prevent water from entering the
lungs. Near-drowning incidents typically occur during the first and second phases. During these stages, blood
oxygen saturation decreases, while the heart rate accelerates to compensate for oxygen delivery disruptions and
ensure the body receives adequate oxygen. Factors such as heart attacks, loss of consciousness, or head and
spinal injuries can prevent a diver from resurfacing. A reduced oxygen level in the blood (hypoxemia) is a
common feature of all near-drowning cases. As drowning progresses, the larynx (air passage) involuntarily closes
to block air and water from entering the lungs [2]. The third phase begins when the individual stops breathing
and becomes unconscious. To prevent this progression, drowning must be detected no later than the third phase
[3]. Preventative measures include close supervision, swimming training, and teaching basic swimming rules
and etiquette. While these strategies have been effective to some extent, they are not foolproof.
Near-drowning alert systems are classified into two types: image processing systems and IoT sensor-based
systems. Image processing systems use algorithms to analyze live video footage for signs of drowning [1]. These
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 198
www.rsisinternational.org
systems often require the use of drones to cover large areas, leading to safety concerns and practical challenges
such as limited drone battery life. While image processing methods outperform sensor-based systems in terms
of accuracy, they are less effective in situations where drowning occurs in murky water, making swimmers
difficult to detect. IoT sensor-based systems, on the other hand, rely on devices like oximeters and water-level
sensors to monitor user conditions. Various drowning detection methods were analyzed based on their accuracy,
complexity, and cost. Sensor-based devices are generally classified as having low to moderate complexity and
cost, whereas image processing systems are regarded as more complex and expensive [1]. Thus, efforts to
develop sensory systems in the form of wearable devices for drowning detection are increasingly prevalent.
The objective of this system is to facilitate the rapid rescue of individuals experiencing near-drowning incidents.
It serves as a monitoring system that tracks the user’s heart rate and oxygen levels, making it particularly suitable
for parents to supervise their children while swimming. The system is designed to detect irregularities in heart
rate or oxygen levels and provide a drowning alert, including the user’s location. The sensors are connected to a
NodeMCU (ESP8266) microcontroller. Equipped with a Wi-Fi module, the ESP8266 enables notifications to be
sent directly to Android apps on users' smartphones.
Related Works
The integration of Internet of Things (IoT) technologies in drowning detection systems has garnered significant
attention in recent years. Researchers have developed various prototypes aimed at improving child safety in
aquatic environments by enabling real-time health monitoring and emergency alert capabilities. These systems
generally share core attributes such as network connectivity, mobile application integration, cloud-based data
storage, and the ability to process physiological data like heart rate, oxygen saturation, and movement. The
primary goal is to ensure rapid detection and intervention in near-drowning incidents, particularly among
children.
Ramdhan et al. [4] developed an EDDS system designed to alert parents and lifeguards when an abnormal
heartbeat is detected. The system employs Radio Frequency (RF) as the communication protocol between the
transmitter and receiver. A PPG sensor is used to monitor heart rhythm by utilizing a light source to measure the
heartbeat. The system incorporates an Arduino Pro Mini 328 (433MHz UART) microcontroller to process the
signals from the pulse sensor and transmit them to an access point (Raspberry Pi2). The access point, connected
to the internet, forwards the data to a database. Once an abnormality is detected, a warning signal is sent to
guardians via a webpage or Android app. The monitoring web page and smartphone app are updated every
second, enabling parents and lifeguards to continuously track the swimmer's status. The monitoring web page
displays essential information such as time, BPM values, and the swimmer's current condition.
Farid et al. [5] developed a child drowning alert system using an IoT PPG sensor to assist parents in monitoring
their children in the pool. The system includes a heart rate sensor and a NodeMCU microcontroller. The heart
rate sensor monitors the children's heart rate, while the NodeMCU runs an algorithm to differentiate between
normal and drowning heart rates. The system connects to a smartphone via Wi-Fi using the Blynk application.
If the detected heart rate exceeds the preset threshold, the microcontroller sends a warning notification to the
parents.
Monish et al. [6] developed a drowning alarm system utilizing RF transmission and a GPRS/GSM module. The
proposed system consists of three main components. The first component is the RF transmitter and receiver,
managed by an ATmega328 microcontroller from the AVR family, which is connected to an LCD screen and
powered by a 12V battery. The heart rate of the individual in the water is displayed on this LCD screen. The
receiver circuit, similar in design, uses an AVR microcontroller and an RF module powered by a 12V
transformer. In an emergency, when the individual's heart rate becomes critical, a distress signal is sent to the
receiver circuit, activating an LED and buzzer. A GPS module pinpoints the individual's location and transmits
it to the Blynk server via GSM, where it is displayed on the Blynk application. This allows lifeguards to perform
live tracking of the person in the water and identify their location using the Blynk app on a smartphone.
In another study, Purnomo et al. [7] developed a wearable health monitoring device using the MAX30100 sensor,
capable of tracking heart rate and SpO₂ levels in real-time. It was integrated with an Android application using
Firebase as cloud storage, enhancing accessibility and data logging. Although not specific to drowning, the
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 199
www.rsisinternational.org
technology has similar applications in water safety systems. Meanwhile, Khan et al. [8] proposed an IoT-based
child safety wearable with multi-sensor input, including GPS and environmental data. The system emphasizes
child location tracking and alerts in critical situations, with potential adaptation for aquatic environments. Their
work underlines the growing trend of merging GPS, GSM, and health monitoring into compact wearable
platforms. A study by Ramesh et al. [9] implemented a smart life-saving jacket designed for swimmers, which
incorporates heart rate sensors, accelerometers, and GSM modules. When critical heart activity or sudden motion
is detected, the system automatically inflates an airbag and sends emergency alerts. Such systems emphasize
how wearable IoT can transition from passive monitoring to active intervention.
Collectively, these studies showcase the rapid progress of IoT in health and safety monitoring applications. While
earlier systems relied on GSM and RF communication, more recent designs now leverage real-time cloud
services like Firebase and advanced mobile app platforms for better scalability, user experience, and cross-
platform functionality. These advancements contribute to the continuous improvement of drowning detection
systems tailored for children, improving parental response time and overall safety.
METHODOLOGY
The Smart Water Safety for Children System aims to help parents keep track of their children's swimming safety
by monitoring their heart rate and oxygen saturation levels. It also records the child's location when a near
drowning incident is detected. The system's primary objective is to enable timely rescue by sending SMS alert
notifications to parents. The system includes three sensors connected to a NodeMCU V3 microcontroller: a
water-level sensor, a heart rate and pulse oximeter sensor (MAX30100), and a GPS sensor (GPS neo-6mv2).
The MAX30100 sensor integrates pulse oximetry and heart rate monitoring capabilities, combining two LEDs,
a photodetector, optimized optics, and low-noise signal processing to capture heart rate and oxygen-level signals
accurately.
A child is considered to be in a near-drowning state when their oxygen saturation drops below 95%, their heart
rate exceeds 140 BPM, or other irregular heart rhythms are detected [10]. The system leverages a Wi-Fi module
to facilitate communication. When such critical conditions are identified, an alert notification containing the
child's current location is sent to the parents.
The NodeMCU V3 microcontroller gathers data from the sensors and transmits it to Firebase cloud storage.
Parents can access and monitor real-time heart rate and oxygen saturation levels through an Android-based
mobile application. This app, developed as part of the Smart Water Safety for Children system, is tailored to
control and monitor the system’s hardware. It provides a user-friendly interface, enabling parents to view their
child’s vital signs directly from their smartphone.
Project Methodology
Figure 1 depicts the Prototype Model as the methodology used to develop the system. The methodology
comprises six stages: Requirement Gathering, Quick Design, Prototype Development, User Evaluation,
Prototype Refinement, and Product Engineering.
Figure 1. Prototype Model
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 200
www.rsisinternational.org
In this system, the requirements specify that the prototype must include basic sensor functionalities to monitor
metrics such as heart rate, oxygen levels, and location. Consequently, the tool should incorporate these sensors
and provide the capability to view the results through an accompanying mobile application. Table 1 shows the
hardware and software requirement for this system.
Table 1 Hardware And Software REQUIREMENTS
Component
Description
NodeMCU V3
A microcontroller development board with Wi-Fi capability. It uses an
ESP8266 microcontroller chip
MAX30100
An integrated pulse oximetry and heart rate monitor sensor
Jumper Wire
Connect breadboard components to the microcontroller and sensors
GY-NEO6MV2
A low-cost yet powerful GPS receiver
Water-level sensor
A cost-effective high-level/drop recognition sensor is achieved by using a
series of parallel exposed wire traces that measure water droplets or volume
to determine the water level.
Arduino IDE
An open-source application compiler to run code on microcontroller
Android Studio
A platform for easily creating interfaces for managing and monitoring
hardware projects from Android mobile.
Fritzing
An open-source hardware initiative that makes electronics accessible
In the quick design stage, the essential functions, mobile application interface design, input and output, and usage
flow are all included. This can be done by using Fritzing that incorporates data gathered in earlier stage. Next,
an initial version of the prototype was constructed to showcase the system's functional model. During this phase,
the circuits and components were assembled and tested. Programming was developed simultaneously, as coding
is crucial for enabling the components to operate. Once the device is functional, additional coding is required to
ensure the Android mobile application operates effectively.
The completed prototype was tested and reviewed by users, making this stage essential for evaluating the
system's strengths and weaknesses. Valuable feedback was gathered from users, enabling improvements to the
system. During this phase, the mobile application's user interface, design, and functionality were also assessed.
Based on user feedback and requirements, the prototype will be updated with new features and adjustments
during the prototype refining stage. This iterative process will continue until users are satisfied with the
prototype. The refined prototype will then be used as a foundation to develop the final system, with enhancements
made until the final product is complete.
Once all necessary procedures and approvals are completed, the prototype entered its final development phase.
During this stage, the technology was fine-tuned and integrated into the final product. The completed version
was then undergoing thorough testing to ensure proper functionality and error-free performance. Ongoing
maintenance will be required to address bugs, manage updates, and deploy new software versions as needed.
Project Design
The architectural framework of the Smart Water Safety for Children System, as illustrated in Figure 2, presents
an integrated IoT-based solution aimed at enhancing child safety during swimming activities. At the center of
the system is the NodeMCU V3 microcontroller, which acts as the main processing unit responsible for
collecting, processing, and transmitting data from multiple connected sensors.
The system incorporates three primary sensors. The MAX30100 pulse oximeter monitors the child’s heart rate
and blood oxygen saturation (SpO₂), both of which are essential indicators of physiological health. If the heart
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 201
www.rsisinternational.org
rate rises above normal or oxygen levels fall below a safe threshold, the system recognizes this as a possible
near-drowning situation. To determine whether the child is submerged, a water-level sensor is used. This sensor
detects when water levels surpass a predefined value, confirming that the child is in a swimming or underwater
environment.
Figure 2. System Architecture
This helps in assessing the child’s condition in real-time. Additionally, a NEO-6MV2 GPS module is integrated
to provide accurate location tracking. This feature is vital for emergencies, allowing caregivers or emergency
responders to locate the child promptly.
The data collected from all sensors is transmitted over a Wi-Fi connection, using the ESP8266 module within
the NodeMCU, to a Firebase cloud server. A mobile application developed on the Android platform then
retrieves this data and presents it in an easy-to-understand interface. Parents can monitor real-time updates on
their child’s heart rate, oxygen levels, submersion status, and location, enabling swift and informed responses in
critical situations. A key feature of the system is its ability to send automated notification alerts when signs of
near drowning are detected, allowing guardians or emergency responders to act promptly. This provides a
practical and reliable method to improve water safety for children.
Figure 3 presents the physical prototype configuration of the Smart Water Safety for Children System. The setup
includes the NodeMCU V3 microcontroller, LED indicators, a buzzer for local alerts, and the sensors—
MAX30100 pulse oximeter, NEO-6MV2 GPS sensor, and a water-level sensor. This integrated design facilitates
both immediate local warnings and remote alerts, enhancing the overall responsiveness and reliability of the
system in emergency situations.
The operational flow of the proposed system begins with the user launching the dedicated mobile application on
an Android smartphone. Upon initialization, the application establishes a connection with the prototype hardware
via a shared local Wi-Fi network, ensuring real-time communication between the device and the phone. The
prototype, which is equipped with various sensors, then actively monitors the user's vital signs, specifically heart
rate and blood oxygen saturation (SpO₂). These physiological readings are captured and transmitted wirelessly
to the Firebase cloud server, acting as the centralized data repository.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 202
www.rsisinternational.org
Figure 3. Prototype setup
Once the data is stored in Firebase, the mobile application retrieves it and presents the readings through a user-
friendly interface, allowing real-time monitoring on the smartphone. The system is designed to detect signs of
potential danger while the user is in the water. If the water-level sensor detects a value above 400, it indicates
that the user is likely swimming or submerged. During this condition, the system continuously monitors the
user's vital signs. If the heart rate exceeds 140 beats per minute (BPM) and the SpO₂ level drops below 95%, the
system interprets these signals as symptoms of potential distress or near-drowning.
In response, an automatic alert notification is generated and sent directly to the user’s registered emergency
contact, such as a parent or guardian. This alert includes critical information such as the user's current health
status and their real-time GPS location, enabling immediate intervention. This process ensures prompt awareness
and response during emergencies, potentially preventing fatal outcomes due to drowning incidents.
The user interface of the Smart Water Safety for Children System application is designed with two main
functional fragments to provide comprehensive monitoring and quick access to critical information. Figure 4
illustrates the visual layout and design of this application interface, where Status Fragment and Location
Fragment can be found. These fragments are structured to enhance user experience and facilitate real-time
tracking of the child's safety status while swimming.
The Status Fragment serves as the primary dashboard for monitoring the child's physiological data. It displays
vital health indicators such as heart rate and blood oxygen saturation (SpO₂) levels. This real-time data is
presented in an intuitive format, allowing parents or guardians to continuously observe the swimmer's condition.
The interface is designed to highlight abnormal readings—for example, elevated heart rate or reduced oxygen
levels—ensuring that any sign of distress is immediately noticeable. Additionally, the swimmer’s overall status,
whether in a normal or near-drowning condition, is also indicated clearly in this fragment.
Figure 4. Interface Design
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 203
www.rsisinternational.org
The Location Fragment, on the other hand, provides geographic tracking capabilities. It features a dynamic map
that displays the swimmer’s current position, enabling users to visually locate the child in real time. The interface
also provides precise latitude and longitude coordinates, ensuring accurate location information is available,
which is crucial in emergencies. This geolocation data is updated continuously as long as the system remains
active and connected to the network.
Together, these two fragments offer a comprehensive monitoring solution by integrating both health data and
location tracking.
FINDINGS & ANALYSIS
This section presents the experimental findings derived from the development and validation of the Smart Water
Safety for Children System prototype. The analysis is structured into two key components: system functionality
testing and physiological data analysis under simulated near-drowning conditions. The primary aim is to evaluate
the system’s effectiveness in real-time monitoring, accurate data acquisition, and timely notification under
critical conditions.
Functionality Testing
To validate the operational reliability of the Smart Water Safety for Children system, a comprehensive
functionality test was conducted. The primary objective was to assess whether each subsystem including
hardware integration, software communication, and user interface responsiveness met the design specifications
for a real-time, IoT-based drowning alert solution.
The testing procedure encompassed individual and integrated component evaluations under simulated
conditions. Each test case focused on a specific system function, with predefined performance benchmarks.
These benchmarks were determined based on the system’s core requirements: real-time physiological data
monitoring, wireless transmission, user alert activation, and emergency communication. The key functionalities
tested included:
1. Wireless Network Connectivity: Establishing a stable and consistent connection between the prototype
and local Wi-Fi for uninterrupted data flow.
2. Sensor Input Validation: Ensuring accurate and continuous measurement of heart rate (BPM), blood
oxygen saturation (SpO₂), and water-level detection.
3. Cloud Synchronization: Verifying real-time data transmission and storage in the Firebase cloud database
using the ESP8266 Wi-Fi module.
4. Mobile Application Display: Confirming accurate rendering of physiological data on the Android
interface, synchronized with the cloud.
5. Emergency Alert Activation: Testing the system's automated response when physiological thresholds
(BPM > 140, SpO₂ < 95) are crossed, triggering buzzer, LED, and push notifications.
6. Location Tracking: Assessing the real-time retrieval and display of GPS coordinates within the mobile
application.
7. Emergency Call Feature: Validating the function that allows users to directly initiate a call to emergency
services through the application interface.
The outcomes of the functionality testing are detailed in Table 2. All functional components demonstrated
expected behavior, passing the evaluation criteria. These results confirm the prototype’s capability to perform
consistently under simulated near-drowning conditions and affirm its potential for real-world deployment.
Table 2 Functionality Testing Result
Test ID
Function Evaluated
Expected Outcome
Status
1
Wi-Fi Connectivity
Stable connection with
local network
Pass
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 204
www.rsisinternational.org
Physiological Data Analysis
Following system validation, controlled testing was conducted to simulate physical exertion and submersion,
replicating conditions associated with near-drowning. The test subject performed light swimming and cardio
exercises while wearing the prototype. Sensor readings were continuously logged and transmitted for analysis.
Key physiological metrics monitored included heart rate (BPM), SpO₂ levels, and water-level sensor values. The
system is designed to trigger an alert when the heart rate exceeds 140 BPM and oxygen saturation drops below
95%, provided the water-level sensor value is above 400, indicating submersion.
To evaluate the functionality and reliability of the Smart Water Safety for Children System, a series of repeated
measurement trials were conducted under controlled scenarios (Table 3 until Table 6). These tests aimed to
validate whether the sensor-based thresholds accurately detected near-drowning conditions while minimizing
false positives. Data were collected across four experimental conditions: resting state, swimming without
exertion, exercise without swimming, and a combination of swimming with exercise. Each scenario was repeated
five times to ensure consistency, with pulse rate (BPM), oxygen saturation (SpO₂), and water level serving as
the core indicators. The yellow highlights in the tables indicate instances where near-drowning was detected,
based on the criteria of water level exceeding 400, heart rate surpassing 140 BPM, and oxygen saturation
dropping below 95%.
Table 3 Resting State (No Exercise/Swimming)
Table 4 Swimming Only Condition
2
Sensor Input (Pulse, SpO₂,
Water Level)
Continuous and
accurate data
acquisition
Pass
3
Firebase Cloud Sync
Data transmitted and
stored in cloud
Pass
4
Android Application
Display
Correct display of live
data
Pass
5
Emergency Alert Trigger
Alert generated when
BPM > 140, SpO₂ < 95
Pass
6
GPS Location Retrieval
Accurate geographical
location on mobile app
Pass
7
Emergency Call Function
Initiate direct call to
emergency services
Pass
Test
Data
1
2
3
4
5
Water-Level
80
67
70
69
74
Heart Rate
(bpm)
83.8
79.5
84.2
89.7
70.8
Oxygen-
Level (%)
98
98
98
97
98
Test
Data
1
2
3
4
5
Water-Level
562
555
544
533
526
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 205
www.rsisinternational.org
Table 5 Exercise Only Condition
Table 6 Exercise And Swimming Condition
In the first scenario, the data for resting condition without swimming or physical exertion is gathered as shown
in Table 3. The water-level sensor consistently registered values below 100, confirming that the subject was not
submerged. The heart rate ranged from 70.8 to 89.7 BPM, with SpO₂ readings holding steady at 98%. These
values fall well within normal physiological ranges for a resting individual and serve as a control reference for
other test scenarios. The absence of alerts during this state confirms that the system does not falsely trigger under
normal, non-risk conditions.
Table 4 represents the second scenario where data for swimming without exercise is collected. In all the test
cases, the water level exceeded 400, indicating active submersion. Despite this, heart rate values remained below
100 BPM and SpO₂ levels maintained at 97–98%, suggesting that submersion alone does not induce a critical
health event. Importantly, the system correctly refrained from sending any alert notifications, validating that the
presence of water alone is not a sufficient trigger unless accompanied by abnormal physiological signals.
The third scenario involved exercise without submersion as shown in Table 5. The water level values remained
below 100, while the heart rate increased significantly—ranging from 110.4 to 140.5 BPM—due to physical
exertion. Oxygen saturation slightly decreased to 94–95%, reflecting increased metabolic activity. While these
readings suggest physiological stress, the system did not send alerts, as the water-level threshold indicating
submersion was not breached. This confirms that elevated BPM or minor reductions in SpO₂ alone are
insufficient to activate the alert system.
The fourth and most critical scenario which combined swimming with physical exertion as displayed in Table 6,
emulating a near-drowning event. In this condition, both heart rate and SpO₂ parameters crossed the system’s
predefined critical thresholds. Heart rate exceeded 140 BPM in two out of five trials, and SpO₂ levels dropped
below 95%, with water levels consistently above 400. Under these conditions, the system successfully triggered
Heart Rate
(bpm)
89.5
90.2
87.9
78.3
100.4
Oxygen-
Level (%)
98
98
98
97
97
Test
Data
1
2
3
4
5
Water-Level
56
64
52
58
55
Heart Rate
(bpm)
110.5
120.3
135.4
140.5
139.7
Oxygen-
Level (%)
96
94
95
95
94
Test
Data
1
2
3
4
5
Water-Level
513
512
509
507
508
Heart Rate
(bpm)
103.3
110.8
132.5
144.8
147.8
Oxygen-
Level (%)
95
94
95
94
94
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 206
www.rsisinternational.org
alert notifications, including real-time location tracking, verifying that the logic for detecting near-drowning
scenarios is operational and precise.
Overall, the findings confirm that the system reliably distinguishes between non-critical and critical states by
assessing a combination of water submersion, elevated heart rate, and reduced oxygen saturation. The multi-
sensor approach ensures contextual awareness, reducing false positives while enabling timely emergency alerts
when actual risk is present. The consistency and accuracy of the results across all test repetitions underscore the
robustness of the system and its practical applicability in real-time drowning prevention for children.
The analysis confirms that the Smart Water Safety for Children System prototype is capable of real-time
monitoring and accurate detection of potential near-drowning conditions based on predefined thresholds. The
integration of the MAX30100 pulse oximeter, GPS module, and water-level sensor, combined with cloud-based
synchronization via Firebase, allowed for timely and efficient alert delivery. The mobile application interface,
developed using Android Studio, demonstrated seamless data retrieval and visualization, enhancing user
accessibility and situational awareness.
While the system performed as intended under test conditions, certain limitations were noted. Environmental
factors such as water splashes, sensor misalignment, and fluctuating ambient light may affect optical sensor
reliability. In future iterations, enhancements such as sensor fusion algorithms and waterproof casing are
recommended to further improve accuracy and robustness.
CONCLUSION
The Smart Water Safety for Children System presents an innovative solution to support parents in actively
monitoring their children’s swimming activities and preventing near-drowning incidents. By integrating real-
time health tracking and geolocation features into a user-friendly Android application, the system enables
immediate alerts and emergency responses when critical conditions are detected. The application not only
displays the child's heart rate and oxygen saturation but also pinpoints their exact location during distress,
allowing parents to act quickly. If the child is out of visual or physical reach, the built-in emergency call function
offers a direct link to emergency services, ensuring swift intervention.
The system is specifically designed for children aged 2 to 12 years old, addressing a highly vulnerable age group.
While the prototype has demonstrated its potential in detecting life-threatening scenarios and facilitating early
rescue responses, its current form still has room for improvement. One limitation is the hardware’s vulnerability
to water exposure. Future development should focus on creating a fully waterproof version of the device to
ensure functionality in real-world aquatic environments. Additionally, incorporating more sensors such as
accelerometers to detect sudden movements or stillness, and temperature sensors to monitor changes in the
child’s body or environment, could greatly enhance the reliability of the detection system. Another proposed
improvement is the integration of a child-friendly airbag mechanism, which could help keep the child afloat
during emergencies, increasing the likelihood of survival before rescue personnel arrive.
In conclusion, the system demonstrates a promising approach to child water safety by leveraging IoT
technologies, and future enhancements could make it a critical tool in drowning prevention and emergency
preparedness.
ACKNOWLEDGEMENT
This research is supported by Faculty Technology Maklumat dan Komunikasi, University Technical Malaysia
Melaka.
REFERENCES
1. A. Alotaibi, “Automated and Intelligent System for Monitoring Swimming Pool Safety Based on the
IoT and Transfer Learning,” Electronics, vol. 9, no. 12, p. 2082, 2020.
2. “Near-drowning,” Medical Dictionary, Aug. 30, 2022. [Online]. Available: https://medical-
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue X October 2025
Page 207
www.rsisinternational.org
dictionary.thefreedictionary.com/near-drowning
3. E. Kałamajska, J. Misiurewicz, and J. Weremczuk, “Wearable Pulse Oximeter for Swimming Pool
Safety,” Sensors (Basel), vol. 22, no. 10, p. 3823, May 2022, doi: 10.3390/s22103823.
4. M. S. Ramdhan, M. Ali, E. Paulson, G. N. Effiyana, S. Ali, and M. Y. Kamaludin, “An Early Drowning
Detection System for Internet of Things (IoT) Applications,” Telkomnika (Telecommunication
Computing Electronics and Control), vol. 16, pp. 1870–1876, 2018, doi:
10.12928/TELKOMNIKA.v16i4.9046.
5. F. Zakwan, Z. Ayop, I. Roslan, S. Anawar, N. Othman, and N. Harum, “Developing Child Drowning
Alert Prototype System using IoT PPG Sensor,” Technology Reports of Kansai University, vol. 62, pp.
6253–6264, 2020.
6. P. Monish, R. Darshan, K. Ponvalavan, and M. Bharathi, “Developing Drowning Detection Systems,”
Journal of Physics: Conference Series, vol. 1997, presented at the Asian Conference on Intelligent
Computing and Data Sciences (ACIDS), Perlis, Malaysia, May 24–25, 2021. Published under license
by IOP Publishing Ltd.
7. M. H. Purnomo, S. Wahyuni, and I. Gozali, “Design of Wearable Pulse Oximeter Based on MAX30100
Sensor with Firebase Cloud Database Integration,” Journal of Physics: Conference Series, vol. 1823,
no. 1, p. 012028, 2021, doi: 10.1088/1742-6596/1823/1/012028.
8. S. Khan, R. Hassan, and M. Ahsan, “IoT-Based Wearable Device for Children Safety Using Multiple
Sensors and GSM,” International Journal of Engineering and Advanced Technology (IJEAT), vol. 8,
no. 6S3, pp. 154–157, 2019, doi: 10.35940/ijeat.F1031.0986S319.
9. T. Ramesh, K. Meenakshi, and R. Akila, “Design and Development of Smart Life Jacket for Drowning
Prevention Using IoT,” International Journal of Innovative Technology and Exploring Engineering
(IJITEE), vol. 8, no. 12, pp. 168–172, 2019, doi: 10.35940/ijitee.L3065.1081219.
10. “Drowning – Emergency management in children,” Children’s Health Queensland, Dec. 21, 2021.
[Online]. Available: https://www.childrens.health.qld.gov.au/guideline-drowning-emergency-
management-in-children/