See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/280829016

Development of Fire Alarm System using Raspberry Pi and Arduino Uno

Conference Paper · December 2013

CITATIONS READS

5 13,243

1 author:

Norlida Buniyamin
Universiti Teknologi MARA
99 PUBLICATIONS   562 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

LHM lens for hyperthermia Cancer Treatment View project

Ultrasound fat measurement sensor View project

All content following this page was uploaded by Norlida Buniyamin on 10 August 2015.

The user has requested enhancement of the downloaded file.

 2013 International Conference on Electrical, Electronics and System Engineering

Development of Fire Alarm System using Raspberry

Pi and Arduino Uno
Md Saifudaullah Bin Bahrudin Rosni Abu Kassim
Faculty of Electrical Engineering Norlida Buniyamin
University Teknologi MARA Faculty of Electrical Engineering
Selangor, Malaysia University Teknologi MARA
[email protected] Selangor, Malaysia

Abstract— The proposed Fire alarm system is a real-time of fuzzy neural network based on the characteristics of fire
monitoring system that detects the presence of smoke in the air detection signal so it will has a self-learning and adaptive
due to fire and captures images via a camera installed inside a capability. However, it is only a study on proposing a new
room when a fire occurs. The embedded systems used to develop approach for fire detection.
this fire alarm system are Raspberry Pi and Arduino Uno. The
key feature of the system is the ability to remotely send an alert Cao Shunxia et.al [2] designed a wireless intelligent home
when a fire is detected. When the presence of smoke is detected, alarm system consisting on anti-theft feature, anti-fire feature,
the system will display an image of the room state in a webpage. and anti-harmful gas leak feature using Single Chip
The system will need the user confirmation to report the event to Microcomputer (SCM) AT89C51 and voice chip ISD1420.
the Firefighter using Short Message Service (SMS). The Two SCMs were used to display the gas concentration and
advantage of using this system is it will reduce the possibility of alarm host as the alarm signals were sent by using wireless
false alert reported to the Firefighter. The camera will only transmission. When the sensor detects smoke, a voice message
capture an image, so this system will consume a little storage and will be sent to the relevant department. However, if an error
power. occurs during the detection, a false alarm will be submitted
because this system did not include any user confirmation.
Keywords-component; Raspberry Pi; Arduino Uno; fire; alarm;
webserver Jun Hou et.al [3] proposed an intelligent home security
system using Zigbee to monitor important locations inside a
I. INTRODUCTION home through a surveillance camera. When the system was
triggered by any penetration, the user will be notified through
Fire is an undesirable event that could bring a great loss of SMS and Multimedia Message Service (MMS). The
social wealth and human life. To prevent this losses, various temperature and gas sensor were connected to the system
alarm systems have been developed such as smoke detectors, motherboard using Zigbee modules and forming a Wireless
temperature sensor based systems etc. As technologies evolved Sensor Network (WSN). Even though it can be included as one
and instruments such as temperature sensors, camera etc of the most advanced system, the system motherboard used to
becomes affordable, various automated fire alarm systems are manage the WSN was too expensive.
now available. In conjunction with the cheaper instruments,
internet based and wireless broadband technologies, have also Rakesh V S et.al [4] improved real-time surveillance
improved and there are now various systems that enables system for home security system using Beagleboard SBC,
cheap, high rate data transmission and wireless networking. Zigbee and FTP Webserver which monitor important locations
The availability of cheap credit card sized single board inside a house using camera and detecting smoke. When smoke
computer such as the Raspberry Pi has enabled the creation of or intruder movement is detected, the system sends warning
numerous automated and monitoring system that has low messages through SMS to cell phones, starts capturing real-
power consumption, faster processing ability at a lower cost. time video for fixed duration and makes the alarm on. But this
The fire alarm system proposed in this paper integrates the use system only sends SMS to warn the user and cannot broadcast
of affordable instruments, connectivity and wireless the live streaming video as the system record the video only.
communication. Moreover, the single board computer used is expensive and has
lower technical specifications compare to the Raspberry Pi.
A. Background study and overview The solution for the problems from the previous research as
Various research and numerous prototypes of automated stated above is to develop a new fire alarm system that alerts
security surveillance system have been developed using the user instantly when any events occur and asks for
various platforms. Recent improvements in GSM, web server permission from the user to report to the Firefighter. The
and microcontroller technologies have led to the developments purpose of this study is to implement a fire alarm system using
of various fire alarm systems. For example, Yu Qiongfang et.al a cheaper single-board computer, the Raspberry Pi, combined
[1] proposed an intelligent fire alarm system using fuzzy neural with a microcontroller board, the Arduino Uno and the use of
network. It process data from the sensors and calculates model mid-level and high level programming languages to write the

978-1-4799-3178-1/13/$31.00 ©2013 IEEE 37

 2013 International Conference on Electrical, Electronics and System Engineering

program. The project is a Bachelor degree final semester The GSM shield controller used is SIM900 and it was
project that has the objective to enable the student to apply attached to the Arduino Uno. Its function is to send SMS to the
knowledge gained from the previous semesters. Firefighter. The SIM900 GSM shield has low power
consumption. It has its own GSM communication module and
II. SYSTEM ARCHITECTURE processor which is programmed using JAVA [10]. It can be
used to send and receive SMS, voice calling and receive data
A. Selecting a Template (Heading 2) packet. The AT commands is used to enable the GSM shield to
send the SMS to the recipient.
Figure 1 shows the system architecture. The structure of
this fire alarm system is composed of five components, which
are Raspberry Pi Model-B single-board computer, Arduino B. Sensor
Uno single-board microcontroller, gas sensor, GSM shield and The sensor used in this system is gas sensor QM-NG1 to
webcam. Raspberry Pi was selected due to its good technical detect the presence of smoke. It can be placed at any locations
specifications, high performance for data processing and is inside the room, but it is best placed at the center of room. The
cheaper than other single board computers available in the selection of the sensor was made due to its high sensitivity for
market. The Arduino Uno is used to reduce the tasks of the poisonous gas, long inductive time and long life span [11].
Raspberry Pi by processing the analog signal from sensor and
send SMS using the GSM shield. The webcam is used to
capture the image of the environment when the sensor detects
C. Hardware connections
any abnormality in the air.
The USB webcam is connected to the Raspberry Pi by
using an USB cable and the GSM shield was attached to the
Arduino Uno. A Local Area Connection cable is plugged in to
the Ethernet port of Raspberry Pi to allow a connection to the
internet. The ports of Raspberry Pi used to connect with the
Arduino Uno are shown in Fig. 2a including the pins of the
Arduino Uno. Port 24 of Raspberry Pi is connected to Pin 13
of Arduino Uno, Port 25 is connected to Pin 12, and Port 23 is
connected to Pin 11. Port 24 is used by the Raspberry Pi to get
warnings from Arduino Uno if any events occur. Port 25 is
used to inform the Arduino Uno if the user confirmed the
alert, while Port 23 is used to inform the Arduino Uno if the
Fig. 1: Block diagram of the fire alarm system user cancelled the alert.

III. METHODOLOGY

A. Raspberry Pi
Raspberry Pi project development started on 2006. It is an
inexpensive computer that uses Linux-based operating system,
equipped with a 700 MHz ARM-architecture CPU, having a
512 MB RAM and featured with two USB ports and an
Ethernet controller [5]. It is able to handle full HD 1080 video
playback and by using the onboard Videocore IV graphics Fig. 2: The ports of Raspberry Pi and the pins of Arduino Uno
processing unit (GPU), it was able to demonstrate the ported
version of the Quake 3 game. It allows flexibility in the choice
of programming languages and installation of software that However, the output voltage from the digital output pins of
could be used [6]. It has the ability to serve webpage by Arduino Uno is 5V, while the input ports of Raspberry Pi can
installing Apache HTTP Webserver on it [7]. Arduino Uno only handle 3.3V. Any input voltage higher than 3.3V applied
with GSM Shield. to the input ports of Raspberry will permanently damage the
Raspberry Pi [12], so a voltage divider circuit was designed to
Arduino Uno is an ATmega328-based microcontroller reduce the output voltage from the Arduino Uno. The
operated using 5.5 Volts input voltage supply and has a calculation for the voltage divider circuit is shown below and
maximum operating frequency of 20 MHz [8]. The power the image of the voltage divider circuit is shown in Fig. 3:
supply for the Arduino Uno can be either from a USB
connection, DC power supply, or both. It is a high performance
device which has low power AVR 8-bit Microcontroller with
32K bytes in-system and advanced reduced instruction set
computing [9].

978-1-4799-3178-1/13/$31.00 ©2013 IEEE 38

 2013 International Conference on Electrical, Electronics and System Engineering

The analog output signal from the sensor will be processed development of the programming in the Raspberry Pi is done
by the Arduino Uno. The sensor got the supply voltage from by using the IDLE software. High level programming language
Arduino Uno 5V pin, and the ground pin of the sensor was selected for writing the OS application program for the system
connected to ground pin of Arduino Uno. The analog output is Python programming language. The IDLE software is the
signal pin of the sensor was connected to analog input pin of Integrated Development Environment (IDE) for Python and it
Arduino Uno. Fig. 3 shows the connection between the sensor was already installed in the OS. A SimpleCV library for
and Arduino Uno. Python was installed to enable the operation of capturing the
image done by the webcam.
The Arduino Uno used open-source Arduino Uno integrated
development environment (IDE) for compiling and uploading
programs to the board. The Arduino Uno programs can be
written using C or C++ language. These two languages were
categorized as an intermediate-level language [13]. The
Arduino Uno is used to monitor the signal output from the
sensor.
Fig. 3: The Voltage Divider Circuit and connections between To enable the user from outside of the local network to
Arduino Uno and the sensor access the webpage, it must be visible in the internet. Port
forwarding technique was used to allow the public devices to
A monitor, an USB keyboard and an USB mouse were used access the internal server that has a private IP address [12].
for developing the software in the Raspberry Pi. The block Two ports were port forwarded, which are Port 80 for HTTP
diagram of the system is shown in Fig. 4. The monitor is port and Port 443 for Apache. Then, a hostname was
connected to the HDMI port of Raspberry Pi using a HDMI registered www.no-ip.com and linked to the Wide Area
cable, while the keyboard and the mouse were connected to the Network (WAN) IP address. The registered hostname is
USB ports of the Raspberry Pi. http://saifubahrudin.no-ip.biz. To keep the current IP address
in sync with the No-IP host, a client was installed to
continually checks and automatically updates the Domain
Name Services (DNS) if the IP address ever changed.

IV. SYSTEM DESCRIPTION

The proposed fire alarm system that used Raspberry Pi as a
master device and Arduino Uno as a slave device was
implemented. The sensor was placed at the center of the room.
It is sensitive to the ionization of the surrounding air, resulting
in the changes of its resistivity. The value of the output voltage
from the sensor circuitry varies linearly with the resistivity of
the sensor, so any changes in the sensor resistivity will result in
the changes of the output voltage. The output voltage of the
sensor circuitry is read in 1024 bit resolution by the Arduino
Uno. At start, the Arduino Uno read the first reading from the
sensor and set the threshold value. The calculation of the
threshold value is shown below:

Fig. 4: Full block diagram of the system

As for the ports configurations of the Raspberry Pi, Port 24
was defined as the input port to receive any alert from Arduino
D. Software Uno, Port 25 was defined as the output port and it is used to
The Raspberry Pi used a Linux-based operating system to inform the Arduino Uno if the user has already confirmed the
operate. In this project, Raspbian Wheezy operating system alert, while Port 23 was defined as the output port and is used
(OS) was chosen because it is recommended by the to inform the Arduino Uno if the user cancelled the alert. While
manufacturer. The Raspbian Wheezy is the modified version of for the pins configurations of the Arduino Uno, Pin 13 was
the Debian Cheezy OS. The OS image was extracted to an SD defined as the output pin for sending the alert to Raspberry Pi,
card by using the Win32 Disk Manager. The configurations of Pin 12 and Pin 11 were defined as the input pin for receiving
the OS such as login details and IP address settings were done. the information from the Raspberry Pi whether the user
The login at startup was disabled to make sure that if the power confirmed or cancelled the alert.
supply to the Raspberry Pi was cut down, the Raspberry Pi will
never ask for login details. The Apache HTTP webserver was The Raspberry Pi waits for any input from the Arduino Uno
installed to make the Raspberry Pi as a webserver. The on its Port 24. Upon detecting the event, the Arduino Uno will
send a HIGH output which is 5V from its Pin 13 to the Ports 24

978-1-4799-3178-1/13/$31.00 ©2013 IEEE 39

 2013 International Conference on Electrical, Electronics and System Engineering

of the Raspberry Pi and the voltage will be reduced as the

current flows through the voltage divider circuit. An image will
be captured using the webcam. The image will be displayed on
the webpage and the Raspberry Pi will wait for the user
confirmation on the validity of the alert. The user can do the
confirmation by clicking a submit button on the webpage. At
the moment the user click, the PHP programming of the
webpage will automatically open a text file named Logger file
and log a command data. The Python programming will open
the text file, and write a word “send” in the Logger_file. Then
the Raspberry Pi will send a HIGH output which is 3.3V
through Port 25 to Pin 12 of the Arduino Uno. The Arduino
Uno will then generate an SMS alert and send it using the GSM
shield.
If the user cancels the alert, the PHP programming of the
webpage write a word “cancel” in the Logger_file. The Python
programming read the content of the Logger_file, then the
image will be deleted and the Raspberry Pi will send a HIGH
output through Port 23 to Pin 11 of the Arduino Uno. The
Arduino Uno will continue to read the analog signal from the
sensor and the Raspberry Pi wait for other alert from the
Arduino Uno. The flowcharts for the programming in the
Arduino Uno and the Raspberry Pi are shown in Fig. 5 and Fig.
6.
Fig. 6: The flowchart of the Python programming in Raspberry Pi

V. RESULT AND DISCUSSION

The system was tested in two different situations; in the
first test, the sensor was placed inside a room and some rubbish
was burnt in the room to fill it with smoke and in the second
test, the sensitivity of the sensor was changed after the sample
reading for threshold value was taken. The first test is defined
as True Warning while the second test is False Warning. The
result will be displayed in the webpage. The webpage can be
accessed at http://saifubahrudin.no-ip.biz.
At first, the user will be prompted to fill the login details as
shown in Fig. 7. After logged on, the user will be redirected to
the display webpage as shown in Fig. 8, which will display the
image captured later.

Fig. 7: The login page

Fig. 8: The default webpage

A. True warning
As the sensor detects the increase of coal gas in the air and
Fig. 5: The flowchart of the C programming in Arduino Uno the value of the measurements surpassed the threshold value,

978-1-4799-3178-1/13/$31.00 ©2013 IEEE 40

 2013 International Conference on Electrical, Electronics and System Engineering

the camera captured an image and it was displayed on the gas, it will be assumed as device error. It is dangerous to leak
webpage. Fig. 9 shows the display of the image in the the methane gas in the room, so the sensitivity of the sensor
webpage. After the submit button was clicked, the PHP was changed after a sample reading to determine the threshold
programming modified the content of the Logger_file by value was taken. After the values of the measurements exceed
writing a word “send” in it. The user will be redirected to the the threshold value, an image was captured. The image was
report webpage as shown in the Fig. 10. The Python displayed on the webpage as shown in Fig. 12. After the cancel
programming read the content of the Logger_file and a HIGH button was clicked, the PHP programming modified the
output was sent through Port 25 of the Raspberry Pi. The content of the Logger_file by writing a word “cancel” in it. The
Arduino Uno reads the signal as a confirmation for sending user will be redirected to the cancel webpage named cancel.php
SMS to the Firefighter. The Arduino Uno activates the GSM as shown in the Fig. 13. The Python programming read the
shield and sent the address of the scene in the form of SMS by content of the Logger_file and a HIGH output was sent through
using AT Commands. Fig. 11 shows the screenshot of the Port 23 of the Raspberry Pi. The Python programming deleted
message sent by the Arduino Uno. the image and the content in the Logger_file. The Arduino Uno
reads the signal from the Raspberry Pi as a cancellation and
continued to monitor the measurements from the gas sensor.
After ten seconds, the user will be redirected back to the
default webpage.

Fig. 9: The image displayed on the webpage.

Fig. 12: The false warning image displayed on the webpage

Fig. 10: The webpage redirected from the display webpage

Fig. 13: The cancel webpage

VI. CONCLUSION
This study presented the development of a fire alarm
system using the Raspberry Pi and Arduino Uno
microcontroller. The developed prototype offered a feature that
enabled verification that a fire actually occurred. The fire
alarm system warns the user by first sending an alert and asks
for confirmation before submitting a report/alert to the “Fire-
fighter”. This system used low cost, reliable instruments that
were suitable to develop a fire alarm and affordable as a final
Fig. 11: The screenshot of the message sent by Arduino Uno year project that enabled students to apply knowledge acquired
during the engineering program.
B. False warning
A false warning test was made to show what happened VII. ACKNOWLEDGEMENT
when any error occurs during the operation of the system. The
source of error can vary. It could be due to sensor failure or gas The authors are grateful to Universiti Teknologi MARA for
leakage. However, in this project, gas leakage detection was the use of the equipments and Laboratories.
not covered, so if the sensor detects the increase of methane

978-1-4799-3178-1/13/$31.00 ©2013 IEEE 41

 2013 International Conference on Electrical, Electronics and System Engineering

VIII. REFERENCES [8] A. R. Krishna, G. S. Bala, A. Sastry, B. B. Sarma and G. S. Alla,

"Design And Implementation Of A Robotic Arm Based On
[1] Y. Qiongfang, Z. Dezhong, F. Yongli and D. Aihua, "Intelligent Haptic Technology," International Journal of Engineering
Fire Alarm System Based on Fuzzy Neural Network," 2009 Research and Applications (IJERA), vol. 2, no. 3, pp. 3098-
International Workshop on Intelligent Systems and Applications, 3103, 2012.
p. 1, May 2009.
[9] A. Abdullah, O. Sidek, N. A. Amran, U. N. Za’bah, F. Nikmat,
[2] C. Shunxia and C. Yanda, "Design Of Wireless Intelligent Home H. Jafar and M. A. Hadi, "Development of Wireless Sensor
Alarm System," Industrial Control and Electronics Engineering Network for Monitoring," 2012 International Conference on
(ICICEE), 2012 International Conference on, p. 1511, August Advanced Computer Science and Information Systems
2012. (ICACSIS), p. 109, 2012.
[10] S. Monk, Programming Arduino Getting Start With Sketches,
[3] J. Hou, C. Wu, Z. Yuan, J. Tan, Q. Wang and Y. Zhou, McGraw-Hill, 2011.
"Research of Intelligent Home Security Surveillance System
[11] Ningbo Hi-Tech Park SKKY Science Technology Co. Ltd., QM-
Based on ZigBee," International Symposium on Intelligent
NG1, Zhejiang, 2010.
Information Technology Application Workshops, pp. 554-557,
2008. [12] Shawn and M. Richardson, Getting Started With Raspberry Pi,
Sebastopol: O’Reilly Media, Inc, 2012.
[4] V. Rakesh, P. Sreesh and S. N. George, "Improved real-time [13] H. Schildt, C++: The Complete Reference, Mcgraw-Hill
surveillance system for home security system using Beagleboard Osborne Media, 2003.
SBC, Zigbee and FTP Webserver," 2012 Annual IEEE India L. Chun-yuan, "Design of intelligent fire alarm system based on
[14]
Conference (INDICON), pp. 1240 - 1244, 2012. GSM network," 2011 International Conference on Electronics
and Optoelectronics (ICEOE), vol. 1, pp. VI-393, July 2011.
[5] L. Garber, "UK Teams Build Supercomputer For $4050," IEEE
Intelligent Systems, p. 19, November 2012. [15] B. Lee and J. Jeon, "An Embedded Router for Internet
Communication Among Private Networks," ICIT 2006. IEEE
[6] Jaseman, "The Pioneers," The MagPi, pp. 7-8, May 2012. International Conference on Industrial Technology, 2006., p.
[7] B. Schaefer, "PiGauge," The MagPi, p. 14, December 2012. 688, 2006.

978-1-4799-3178-1/13/$31.00 ©2013 IEEE

View publication stats
42