2021 International Conference on Cyber Security and Internet of Things (ICSIoT)

Innovative Fire Detection and Alarm System for

Sustainable City Development
2021 International Conference on Cyber Security and Internet of Things (ICSIoT) | 978-1-6654-7878-6/21/$31.00 ©2021 IEEE | DOI: 10.1109/ICSIoT55070.2021.00015

Amevi Acakpovi Douglas Tetteh Ayitey Edward Nagai Adjaloko

Electrical/Electronic Engineering Biomedical Engineering Department Computer Science Department
Department University of Ghana Accra Technical University
Accra Technical University Accra, Ghana Accra, Ghana
Accra, Ghana [email protected] [email protected]
[email protected]

Abstract— Nowadays, a number of cities are evolving into system can be monitored locally or remotely, depending on
smart cities integrating automated, real-time, and reliable services the situation. Fire detection and alarm systems are all
to numerous inhabitants. However, the challenges of smart cities structured and introduced with the same fundamental target:
are equally numerous and chief among them in Ghana was the to discover a fire and provide an effective notification through
repeated and unstoppable fire outbreaks. Fire is a highly
an alarm. However, in most cases of fire outbreaks, these
dangerous scenario, and it is critical to monitor it constantly and
detect it at the earliest stage to avoid serious devastation. This objectives are not satisfied and this may be due to a lack of
paper proposes an innovative method of efficiently and effectively sensitivity of the sensing system or lack of maintenance and
detecting early fire at a relatively low cost. This study adopted an monitoring of the overall system.
experimental approach in designing and constructing a low-cost When compared to a manual system, a remote alarm
artifact that made use of an Arduino board and a set of sensors system offers the advantage of being able to remotely monitor
including an MQ-2 sensor, Infrared sensor, LCD display, and a the property and take immediate action in response to the
buzzer. The device was programmed in C using the Arduino message received [1]. Wireless sensor networks, Ethernet,
development environment. The proposed system was able to detect image processing, and other types of digital communication
the early presence of smoke, trigger an alarm, and displayed can all be used to develop remote monitoring systems, which
constantly the status of the system on an LCD display. Consistent can then be used to monitor a number of locations [2]–[5].
tests proved that the developed fire alarm system was reliable, Even though these systems are dependable and offer
responsive, and cost-effective. The mass production of this device numerous benefits, they are complex, inefficient, non-
will help mediate the constant occurrence of fire in buildings, standalone, expensive, and feature redundant components. As
especially in developing countries. However, the availability of raw
a result, there is a need to build a system that is reliable and
materials for quick and effective products is also a challenge. The
responsive, as well as simple, easy to install, and cost-
merit of the paper relies on the intelligence programmed in the
microcontroller coupled with the operation of the sensor that
effective.
permitted early detection of fire yet at a relatively lower cost. I. LITERATURE REVIEW
Keywords— fire alarm system, fire detection, microcontroller, This section provides an overview of the latest
Arduino, remote monitoring, automatic system. developments in fire detection and prevention technology. It
Introduction is possible to classify the bulk of fire detection technologies
into two categories: vision-based and sensor-based
The development and planning of smart cities rely technologies (see Figure 1). This section dives into each of the
fundamentally on the careful design of the security and safety subjects listed above in further detail [6]–[9]
system of the building. A safe building provides assurance to
the occupant and creates a conducive atmosphere to transact As the number of security cameras for fire detection has
business. However, recently in Ghana, a number of buildings expanded, vision-based fire detection has become
were subjected to unpleasing scenarios of the fire outbreak increasingly widespread. The reason for this is that these
that were devastating. Even though some security and safety vision-based systems do not have a separate hardware budget
systems have been installed in those buildings, the majority of like traditional systems do. The disadvantage of using a video
the factors were related to the effectiveness and reliability of frame detection-based approach is that it is not suited for
the electrical systems installed. early-stage fire detection because there may only be smoke or
a few little fire flames to identify at that point in time.
Fire dangers can be extremely severe and result in human Therefore, current video frame detection-based systems may
death. The only way to prevent these losses is to act be deceiving when it comes to accurately recognizing and
immediately in an emergency situation. In industrialized reporting fires, particularly at the early stage, which is critical
countries such as the United Kingdom, Germany, Italy, and for the firefighting [10], [11].
others, it is a government requirement that all residences have
a fire alarm installed, to inform occupants and fire department Because of the grey and semi-transparent character of
workers in time for action. In middle-income countries such smoke, the sharpness of the borders of high-frequency images
as Ghana, fire alarm systems are not constantly monitored and may be compromised, resulting in an erroneous signal of the
checked from residences, resulting in significant losses and presence of fire. In addition, changes in backdrop color tones,
damage whenever there is a fire outbreak. segmentation of smoke-colored pixels, blur background,
lighting, and other parameters are used to distinguish smoke
Early detection of fires and notification of authorities can help from other types of background noise. The problem with this
to reduce property and life losses. A fire or smoke alarm technique is that it is impossible to distinguish between foggy

978-1-6654-7878-6/21/$31.00 ©2021 IEEE 30

DOI 10.1109/ICSIoT55070.2021.00015
Authorized licensed use limited to: SRM Institute of Science and Technology. Downloaded on December 02,2024 at 07:02:00 UTC from IEEE Xplore. Restrictions apply.
 weather and smoke when using it. An approach that is being sensor, LEDs, Buzzer, and the Liquid Crystal Display LCD.
considered for addressing this issue is motion analysis, which The components are grouped into the following units and
includes a vision-based technique for accurately identifying illustrated in the block diagram shown in Figure 1.
smoke. x Power supply unit
The use of sensors such as gas sensors, on the other hand, x LED unit
allows fires to be detected even before they become inflamed, x Fire and smoke sensor unit
depending on the type of gas that is being emitted. These x Display unit
sensor-based fire detection technologies are simple to set up, x Sound unit
inexpensive, and quick to deploy, making them an excellent
choice for many applications.
In order to provide a more desirable technique of fire
detection, computer-based fire detection is increasingly being
combined with sensor networks. However, it has the potential
to increase the cost and complexity of the system when it
comes to installation and deployment.
A. Empirical review
Recently, a number of initiatives have been launched to
develop systems that can detect and control fire outbreaks. Figure 1 System block diagram
The following paragraphs discuss works that have been A. The Power supply Unit
carried out in fire detection and control.
The Arduino Nano uses an ATmega328 microcontroller chip.
(Aydin et al., 2019) examined the potential use of fire The input and output pins are 14 and 6, respectively. There
extinguishing balls as part of a proposed system, where drone are also eight (8) analog input pins, an on-board resonator, a
and remote sensing technologies are utilized cooperatively as reset button, and holes for mounting pin headers. The
a supplement to traditional firefighting methods. The system operating voltage is 5V, the recommended input voltage is
consists of courting unmanned aircraft system (UAS) to detect between 7 - 12 Volts, and the DC current per I/O Pin is 40
spot fires and monitor the risk of wildfire approaching a mA. The figure below shows the Arduino Nano
building via remote sensing, communication UAS to establish
and extend the communication channel between scouting
UAS and firefighting UAS, and a fire-fighting UAS. One has
to be very skillful in controlling drones and also the system is
very complex which makes the system difficult to use.
(Wu et al., 2018) designed an intelligent smoke alarm system
with a wireless sensor network using ZigBee. The system
consists of a smoke detection module, a wireless
communication module, and intelligent identification and data
visualization module. The disadvantage of his system is that it
is very expensive and complex to design.
(Nsaif Shehab, 2018) designed and implemented a Factory
Security System that consists of a smoke sensor, a GSM Figure 2 Arduino Pro Mini Pin diagram
(Global System for Mobile communication) module, and a B. Light Emitting Diode (LED) Unit
sound module. When a gas leakage is detected, an SMS will
A light-emitting diode (LED) is a semiconductor diode that
be sent to a number. The disadvantage of the system is that
emits light when it is forward-biased. Thus, the anode
there is no device that can stop the gas leakage and hence
connected to the positive of the potential difference (p.d) and
when there is a fire outbreak the necessary device to
the cathode connected to the negative side of the potential
extinguish the fire is not included in the system which may
difference connected across the circuit. The symbol for the
cost loss of properties.
LED is shown below with the corresponding picture.
(Sonsale et al., 2014) designed an Intelligent Fire
Extinguisher System. The features are intelligent fire
detection and suppression, locating the position of fire
origin, effective power control of electricity, reporting
through an SMS or email, and effective usage of water
supply. Among the sensors used is a gas sensor that detects Figure 3 schematic symbol of an LED
any type of smoke; however, this can send a false alarm and
hence not reliable.
C. Fire and smoke sensor unit
II. SYSTEM DESIGN A smoke detector or a point heat detector is typically found
The proposed system is a fire detection and alarm system. It in a standard fire alarm system. An MQ-2 gas sensor was
consists of the following components to integrate the employed as a smoke detector and an IR flame sensor in this
system`s goal: Arduino Pro Mini, sensor MQ2, IR flame project. Based on a test done on the two sensors, a

31

Authorized licensed use limited to: SRM Institute of Science and Technology. Downloaded on December 02,2024 at 07:02:00 UTC from IEEE Xplore. Restrictions apply.
 temperature sensor was not chosen for this design. The MQ-
2 gas sensor detects LPG (natural gas) and smoke (carbon
monoxide), most notably, making it ideal for detecting gas
leaks or smoke. A measurement can be performed as soon as
possible because of its high sensitivity and short reaction
time. The sensor's sensitivity can be adjusted using the
potentiometer located beneath the gas sensor module. The IR
flame sensor, on the other hand, is a sensor that is most
sensitive to ordinary light and is known as a flame sensor.
This sensor detects flames and detects wavelengths from the
light source ranging from 760 nm to 1100 nm. If a high
temperature is applied to this sensor, it is readily damaged.
As a result, this sensor can be set at a predetermined distance Figure 6 LCD screen configuration circuit
from the flame. The flame detection may be done from a
distance of 100cm, with a detection angle of 600. This The LCD basically works on the concept of light polarization
sensor's output is either analog or digital. These sensors can of liquid crystals under the influence of an electric field. The
be found in firefighting robots, such as flame alarms. (See LCD contains a backlight behind the liquid crystal array
Figure 4 and 5). which acts as a light source. When an electric field is applied
across certain fluids, it changes the way they allow light to
pass through them thus it changes the orientation of the liquid
crystal molecules and as a result, light cannot flow through
them. Hence, by applying a suitable potential difference, one
can control whether or not light passes through the LCD
pixels.
E. Sound Unit
The sound unit used is a Buzzer (Figure 7). When the
microcontroller outputs a high signal, the buzzer in this
circuit is activated. A buzzer is an auditory signaling device
that can be mechanical, electromechanical, magnetic,
electromagnetic, electro-acoustic, or piezoelectric. An
Figure 4 Diagram of an MQ-2 sensor oscillating electrical circuit or another audio signal source can
power a piezoelectric buzzer. A button press can be indicated
by a click, beep, or ring. It comprises several switches or
sensors linked to a control unit that determines whether or not
a button was pressed or if a pre-set time has elapsed. It usually
displays a light on the appropriate button or control panel and
emits a continuous or intermittent buzzing or beeping sound
as a warning.

Figure 5 Schematic circuit of an IR sensor

D. Display Unit
The information obtained from the level sensors is displayed
on the liquid crystal display (LCD). The LCD features an in-
built controller that allows it to be easily controlled by an
external microcontroller. The following sections discuss
some of the LCD screen's features. (Figure 6).

Figure 7 Circuit Diagram of a Buzzer

III. ASSEMBLY OF THE FIRE DETECTION AND ALARM SYSTEM
This section explains in detail how all of the different parts
were assembled and packed. The project's housing is a plastic
case with screws. Figures 9, 10, 11, and 12 show the step-by-
step assembling of the circuits.

32

Authorized licensed use limited to: SRM Institute of Science and Technology. Downloaded on December 02,2024 at 07:02:00 UTC from IEEE Xplore. Restrictions apply.
 A. Software
This section represents the software development process.
Flowcharts depict the procedures of the main program and
subsequent programs for fire detection and liquid LCD.

B. Flowchart of the main program

The flowchart in figure 8 shows the logic and sequential flow
of the software program that runs in the microcontroller
utilized for the processing and control unit. When the system
is first turned on, the yellow led illuminates, indicating less
smoke in the region, and the system runs a test run, testing Figure 9 System schematic circuit in Proteus
the number of smoke particles in the area. If it exceeds 400,
the green LED turns out, and the red LED illuminates,
indicating substantial smoke in the region. If a fire is brought
closer to the IR flame sensor, the buzzer sounds, and the LCD
panel displays the "fire detected." As a result of several
testing, an experimentally chosen threshold of 40 ℃ was
chosen, and this value will not allow the IR flame sensor to
be oversensitive.

Figure 10 System construction phase I

Figure 11 System construction phase II

Figure 8 Program flowchart

Figure 12 Completed system

33

Authorized licensed use limited to: SRM Institute of Science and Technology. Downloaded on December 02,2024 at 07:02:00 UTC from IEEE Xplore. Restrictions apply.
 C. Coding based fire detection system though their features allow for
The C++ programming language was used to create the code multipurpose control and monitoring. The results obtained
that runs in the microcontroller's flash memory. The Arduino from the measurement and tests conducted have shown that
IDE is the interactive development environment (IDE) in the system performs well under all the conditions.
which the code was created, compiled, and downloaded into
the microcontroller. Additional simulations were carried out This implementation is cost-effective and reliable. The
using the Proteus software. system is an automatic fire alarm system with a simple
interface for fire detection as compared with many previous
studies [19], [24]–[26] and also [27]–[30]. Fire detectors use
IV. RESULTS various sensors, generally smoke sensors. The sensor input
This section presents the results of the fire detection system data is connected with the Arduino controller which tests a
designed and constructed. The test results are shown in the high (1) when smoke is detected. LCD display and the buzzer
proceeding subsections. were connected with Arduino and were triggered upon
detection of fire. Precisely, the buzzer provides notification
A. System Test Results by sounding an alarm, and the LCD screen displayed the text
Tests were done to observe the system’s performance. The “fire detection”. This system can be applied in residential
tests were completed by applying heat near the IR flame places, offices, and hotels as highlighted by (Mahzan et al.,
sensor. Figure 13 shows the condition when the system is in 2018). Safety is also assured as the system requires low
ready mode (ready to detect fire but detected none). Figure current output to operate effectively. The system can also be
14 displays when the IR flame sensor detects fire and an alert expanded to perform other parameter measurements for the
message appear on the LCD display with the buzzer also early detection of building fires. There should be a minimum
making a sound, the red led also turns on with the yellow led of two or three smoke detectors, placed at vantage locations
ggoingg off. in a home for optimum monitoring and control.
The advantages of the system are mainly the early reaction,
which ultimately saves lives and property, and the cost-
effectiveness.

Table 2 Budget of the Designed System

Cost
No. Items Quantity Unit
(GHC)

1 Plastic Case 1 12 12.00

Arduino Pro 80.00

2 1 80
Mini
3 LED 2 5 5.00
Figure 13 system ready condition mode
4 LCD 1 15.10 15.10

Gas Sensor 10.06

5 1 10.06
(MQ-2)
IR Flame 10.06
6 1 10.06
Sensor
7 Buzzer 1 2.41 2.41

8 Jumper wire 1 set 10.00 10.00

Total 144.63

Figure 14 Complete system tests I VI. CONCLUSION

Making cities and human settlements safe, resilient, and
V. DISCUSSION sustainable will require that homes and offices are designed
In this paper, an attempt has been made to design a fire alarm to provide the occupancy with some sort of assurance that
system using smoke sensors and a Microcontroller for their place of work or residence can provide them with the
efficient use of electricity. The simple fire detection system basic security especially with regards to fire. Over the years,
in this study reduces overall processing time as compared to the homeowner has had lots of valuable asserts due to
the Internet of Things IoT [16]–[19] and GSM [20]–[23] incipient fire that was not detected at the early stages and with

34

Authorized licensed use limited to: SRM Institute of Science and Technology. Downloaded on December 02,2024 at 07:02:00 UTC from IEEE Xplore. Restrictions apply.
 time became huge to the extent that fire wardens find it very [8] A. Khondaker, A. Khandaker, and J. Uddin,
difficult to curtail. This neglected event has led to a loss of “Computer vision-based early fire detection using
lives and properties. The existing technologies make use of enhanced chromatic segmentation and optical flow
the embedded system with some features such as GPS, WiFi, analysis technique,” International Arab Journal of
and sensors to help in early detection of fire, precise location Information Technology, 2020, doi:
reporting, and notifications to the occupants of the building 10.34028/iajit/17/6/13.
and the fire service station for immediate action to reduce the [9] F. Bu and M. S. Gharajeh, “Intelligent and vision-
cost associated with fire outbreaks. However, the cost of based fire detection systems: A survey,” Image and
these smart devices is relatively expensive and hence the Vision Computing, 2019, doi:
development of this affordable and easy-to-install and 10.1016/j.imavis.2019.08.007.
maintain fire alarm system. [10] D. Pritam and J. H. Dewan, “Detection of fire using
image processing techniques with LUV color
The development of a fire alarm system using the Arduino space,” 2017. doi: 10.1109/I2CT.2017.8226309.
Uno microcontroller was presented in this study. The [11] S. Bharathi, S. Gokilapriya, N. Elango, and P.
designed and constructed prototype included a feature that Vidhya, “Fire Detection and Fire Signature Using
allowed users to confirm early detection of fire. The fire Color Models for Security,” International Journal of
alarm system sounds an alarm and a fire alert message was Current Research and Modern Education (IJCRME),
displayed on an LCD screen to inform users. This system 2016.
made use of low-cost, dependable equipment that could be [12] B. Aydin, E. Selvi, J. Tao, and M. J. Starek, “Use of
utilized to create a fire alarm. The system detected smoke and fire-extinguishing balls for a conceptual system of
beeped an alarm, as well as displayed the system status on an drone-assisted wildfire fighting,” Drones, 2019, doi:
LCD panel. It is proposed that the system be advanced in 10.3390/drones3010017.
order to detect the location of a fire, making firefighting [13] Q. Wu et al., “Intelligent Smoke Alarm System with
easier and ensuring the safety of people and property. Wireless Sensor Network Using ZigBee,” Wireless
Communications and Mobile Computing, 2018, doi:
10.1155/2018/8235127.
ACKNOWLEDGMENT [14] J. Nsaif Shehab, “Design and Implementation of
The authors gratefully acknowledge the reviewers of this Factory Security System,” Journal of Engineering
paper for their valuable advice and guidance. and Sustainable Development, 2018, doi:
10.31272/jeasd.2018.1.13.
REFERENCES [15] P. Sonsale, R. Gawas, S. Pise, and A. Kaldate,
[1] S. Suresh, S. Yuthika, and G. A. Vardhini, “Home “Intelligent Fire Extinguisher System,” IOSR Journal
based Fire Monitoring and warning system,” 2017. of Computer Engineering, 2014, doi: 10.9790/0661-
doi: 10.1109/ICTBIG.2016.7892664. 16185961.
[2] P. Barmpoutis, P. Papaioannou, K. Dimitropoulos, [16] I. M. Al Shereiqi and M. mad Sohail, “Smart Fire
and N. Grammalidis, “A review on early forest fire Alarm System Using IOT,” Journal of Student
detection systems using optical remote sensing,” Research, 2020, doi: 10.47611/jsr.vi.882.
Sensors (Switzerland), 2020, doi: [17] A. Mahgoub, N. Tarrad, R. Elsherif, A. Al-Ali, and
10.3390/s20226442. L. Ismail, “IoT-based fire alarm system,” 2019. doi:
[3] M. Abdollahi, T. Islam, A. Gupta, and Q. K. Hassan, 10.1109/WorldS4.2019.8904001.
“An advanced forest fire danger forecasting system: [18] S. Tiwari and S. Bandopadhaya, “IoT Based Fire
Integration of remote sensing and historical sources Alarm and Monitoring System,” Shuvabrata
of ignition data,” Remote Sensing, 2018, doi: Bandopadhaya International Journal of Innovations
10.3390/rs10060923. & Advancement in Computer Science IJIACS ISSN,
[4] V. Sherstjuk, M. Zharikova, and I. Sokol, “Forest 2017.
Fire Monitoring System Based on UAV Team, [19] S. Suresh, S. Yuthika, and G. A. Vardhini, “Home
Remote Sensing, and Image Processing,” 2018. doi: based Fire Monitoring and warning system,” 2017.
10.1109/DSMP.2018.8478590. doi: 10.1109/ICTBIG.2016.7892664.
[5] E. H. Chowdhury and Q. K. Hassan, “Development [20] A. Ashwitha, K. Arjun, and H. Prashanth, “GSM
of a new daily-scale forest fire danger forecasting Enabled Smart Fire Alarm Controlling System with
system using remote sensing data,” Remote Sensing, SMS Alert,” International Journal of Computer
2015, doi: 10.3390/rs70302431. Applications, 2016, doi: 10.5120/ijca2016910797.
[6] G. Roque and V. S. Padilla, “LPWAN Based IoT [21] J. Zheng and Y. Tu, “Based on GSM Wireless Fire
Surveillance System for Outdoor Fire Detection,” Alarm System,” International Journal of Computer
IEEE Access, 2020, doi: Applications, 2016, doi: 10.5120/ijca2016908035.
10.1109/ACCESS.2020.3003848. [22] B. A. Karim and H. N. Awl, “Smart Home System
[7] A. Kushnir and B. Kopchak, “Development of Based on GSM Network,” Kurdistan Journal of
Computer Vision-Based Automatic Flame Detection Applied Research, 2018, doi:
Algorithm Using Matlab Software Environment,” 10.24017/science.2018.1.4.
Fire Safety, 2020, doi: [23] N. N. Mahzan, N. I. M. Enzai, N. M. Zin, and K. S.
10.32447/20786662.36.2020.05. S. K. M. Noh, “Design of an Arduino-based home

35

Authorized licensed use limited to: SRM Institute of Science and Technology. Downloaded on December 02,2024 at 07:02:00 UTC from IEEE Xplore. Restrictions apply.
 fire alarm system with GSM module,” 2018. doi: SMS Alert,” International Journal of Computer
10.1088/1742-6596/1019/1/012079. Applications, 2016, doi: 10.5120/ijca2016910797.
[24] I. M. Al Shereiqi and M. mad Sohail, “Smart Fire [28] J. Zheng and Y. Tu, “Based on GSM Wireless Fire
Alarm System Using IOT,” Journal of Student Alarm System,” International Journal of Computer
Research, 2020, doi: 10.47611/jsr.vi.882. Applications, 2016, doi: 10.5120/ijca2016908035.
[25] A. Mahgoub, N. Tarrad, R. Elsherif, A. Al-Ali, and [29] B. A. Karim and H. N. Awl, “Smart Home System
L. Ismail, “IoT-based fire alarm system,” 2019. doi: Based on GSM Network,” Kurdistan Journal of
10.1109/WorldS4.2019.8904001. Applied Research, 2018, doi:
[26] S. Tiwari and S. Bandopadhaya, “IoT Based Fire 10.24017/science.2018.1.4.
Alarm and Monitoring System,” Shuvabrata [30] N. N. Mahzan, N. I. M. Enzai, N. M. Zin, and K. S.
Bandopadhaya International Journal of Innovations S. K. M. Noh, “Design of an Arduino-based home
& Advancement in Computer Science IJIACS ISSN, fire alarm system with GSM module,” 2018. doi:
2017. 10.1088/1742-6596/1019/1/012079.
[27] A. Ashwitha, K. Arjun, and H. Prashanth, “GSM
Enabled Smart Fire Alarm Controlling System with

36

Authorized licensed use limited to: SRM Institute of Science and Technology. Downloaded on December 02,2024 at 07:02:00 UTC from IEEE Xplore. Restrictions apply.