Content uploaded by Marianne A. Azer
Author content
All content in this area was uploaded by Marianne A. Azer on May 31, 2022
Content may be subject to copyright.
Design and Implementation of Robust
Firefighting/Intruder Detection System Using Fuzzy
Logic Decision Control
(FIDFUZ)
Marianne A. Azer
National Telecommunications Institute
Nile University
Cairo, Egypt
Mazer@nu.edu.eg
Ahmed ElShafee
Faculty of Engineering
Ahram Canadian University
Cairo, Egypt
aelshafee@gmail.com
Abstract— This research focuses on using quantifiable
methods for using the IoT as a main support to
firefighting/intruder detection. From our research, we have
found numerous researches associated to supplying remote
services by means of portable sensors and communication
technologies. We represent in our research a unique Smart
Firefighting/Intruder Detection System with the support of Fuzzy
Logic Decision Control (FIDFUZ). The projected system has an
innovative value which is the Fuzzy Logic Decision Support
System application that deals with the predicted inaccuracy and
the doubt in the sensor’s information acquired, as well as
minimizing the rate of false positive and true negative. FIDFUZ
full architecture is presented, applied and checked by simulation
and using real data.
Keywords— Firefighting, Fuzzy Logic, Intruder Detection, IoT,
Smart Home
I. I
NTRODUCTION
Over the past few years, there has been a concern on the
growing interest of controlling and automating adjacent
environment using the monitoring devices and sensors.
Internet of Things (IoT) is one of the principal of upcoming
technological inventions and its usage is increasing drastically
over time [1].
IoT can be regarded as a network of heterogeneous
"Things". Things are meant to be physical objects which are
enabled by ambient intelligence and communication
capabilities for continuous data exchange. They can link to
each other's or to the cloud at anytime, anywhere, through any
network with absolute abstraction. Things have various sizes,
and computing capabilities and they range from RFID tags to
Vehicle on Boards Units (OBUs).
A noticeable objective of IoT, is to switch Human-to-
Machine communications into Machine-to-Machine
communications in order to decrease human collaboration with
its environment to the least levels, hence to escalate the ease of
their life style and enhance human's quality of life [2].
After the significant progression of wireless
communications, the trend of IoT has ascended. By
incorporating many research models such as embedded
systems, wireless networks, cloud computing, and automation,
IoT has empowered many advanced services which include
smart homes, healthcare assistance for elderly, smart cities,
transportation, and entertainment [3].
IoT relies on low cost, light weight communication
technologies such as GSM, Zigbee, Bluetooth, WiFi [4] [5] [6]
[7][8][9] [10] [11] [12] [13] [14].
Our research gives attention to smart firefighting and smart
intruder detection as both applications are under IoT smart
homes umbrella.
In this paper, we introduce Smart Firefighting/Intruder
Detection System supported by Fuzzy Logic Decision Control.
The proposed innovative system uses the Fuzzy Logic
Decision Support System in order to work with expected
inaccuracy and uncertainty in the acquired sensors' information
[15], and minimize false positive and true negative rates as
well.
The remainder in this paper is organized as follows.
Section II represents the background of different paradigms
used and executed in the research. Section III discusses some
of the latest researches that are related to IoT applications.
Section IV presents the architecture, functionality and
experimental results of the proposed Smart
Firefighting/Intruder Detection System supported by Fuzzy
Logic Decision Control (FIDFUZ). Finally in section VI,
conclusions and future work are presented.
II. BACKGROUND
In today’s world the requirement for fire prevention is
drastically increasing. Public attention is provided to enhance
the use of science and technology for Firefighting and
monitoring. IoT technologies meet the requirements of
firefighting as well as Wireless Sensor Network (WSNs). IoT
is extremely diverse and scalable to support the massive range
of equipments for fire monitoring and fighting. IoT based
firefighting methodology can support any of the following
functionalities:
ª*&&&
• Hazard source monitoring,
• Fire monitoring,
• Fire-fighting rescue,
• Fire early warning, prevention and early disposal.
Besides, it is used efficiently to improve the fire
brigade, fire frightening and emergency rescue capabilities.
On the other hand, the security of an organization is a very
important aspect and should be taken into consideration to
assure that there is no breach and no entry of unnecessary
people in any premises. Consequently, having an intrusion
detection system has become essential in every industry such
as institutions, companies or even homes. IoT can be used to
manage such situations by building a simple monitoring system
that depends on sensors and WSNs as itreduce the probability
of error and ultimately saves money by reducing security labor
cost.
Fuzzy logic algorithms are based on information
fusion, which analyzes and combines multi sensors or multi
sources of data under certain criteria to complete data
processing for decision making. The algorithm fuses situation
parameters from multi sensors to decide whether the alarm
detected by one sensor represents a True "Positive" alarm or
not. This method overcomes the limitation of using single
sensor. This algorithm effectively advances the reliability and
reduces the rate of false "Positives" [16] [17].
III. RELATED
WORK
There are several researches that are related to
providing remote services by using portable sensors along with
communication technologies. B. Avitall, et al. presented in [18]
a remote health monitoring system using a physician gadget
that interactively monitors the patient’s health by questions and
receiving answers. A technique of monitoring patients within
their own homes using electronic devices plus wireless
technology was discussed by N. Alshurafa, et al. in [19]. J.
Yick, et al, presented system architecture that provides real-
time sensor’s information, and real-time feedback for the user,
and forwards the actual user’s information to the telemedicine
server [20]. James M. Tien introduced a survey regarding the
usage and combination of Internet of Things (IoT), Real Time
Decision Making (RTDM) and Artificial Intelligence (AI)
paradigms for supplying real time remote services [2].
Moreover, M. JASINKI, et al. publicized a fuzzy logic
algorithm for decision support system controlling IoT based
shunt active power filter [15]. In [16] Hamid Medjahed, et al.
pointed out the use of the multimodal system called EMUTEM
for medical distress situations detection. Zanella, N. Bui et al
[21] discussed the Internet of Things for smart cities. T. Kelly
et al [22] deliberated the implementation of IoT algorithm for
environmental monitoring in homes. N. K. Suryadevara [23]
presented power management in intelligent buildings using
smart sensors and actuator. Furthermore, Konstantinous at.el
[24] stated an automated fire detection and alerting application
based on satellite and wireless communications. S. R.
Vijayalakshmi and S. Muruganand [25] discussed the
challenges in integrating Wireless Sensor Network and Internet
of Things for Environmental Monitoring.
IV. THE
PROPOSED
ROBUST
FIREFIGHTING/INTRUDER
DETECTION
SYSTEM
SUPPORTED
BY
FUZZY
LOGIC
DECISION
CONTROL
(FIDFUZ)
In this section, we present our proposed IoT based
system for firefighting and intrusion detection that is supported
by a fuzzy logic decision control. The system's overall structure
is depicted in Fig.1. It is composed of four main units: sensors'
station (Input and Data Collection Unit), Sensors' front end
(Data Aggregation Unit), Fuzzification server (Data
Classification, Fuzzification and Processing Unit) and Alarm
Activation/Deactivation module (Decision Control and Output
Unit). Each of these components is explained in details in the
following subsections.
A. Input and Data Collection Unit:
Sensors' station is considered to be the Input unit; it consists
of two separate groups of sensors. Group 1, is equipped for
sensing environmental information related to firefighting
module. It contains three different sensors to measure room
temperature, humidity level, and Co2 gas level. Fire is detected
when temperature sensor measures high level (above 40
degrees), humidity sensor measures very low humidity (less
than 30), and Co2 gas sensor measures high level as well as
(less than 240). Group 2, is equipped for sensing environmental
information related to intrusionr detection module. It also
consists of three different sensors to measure the rate of scene
change, the obstacles distanced from a certain location, and
noise level. Intrusion is detected when the PIR sensor
measures any variation in the captured scene, it takes a new
scan every 500 millisecond (every change is assigned to a new
reading on the sensor), ultrasonic sensor measures the distance
to the first obstacle from the sensor (say the room is 6 meter
width, then the sensor should read 6 m all the time unless there
is an intruder that intersects the ultrasonic signal from a closer
distance, then the reading will be less than 6 meters), and sound
sensor measures if any noise occurred in the monitored
location (less than 2.3 volts)..
B. Data Aggregation Unit
Sensors are connected to a front end Microcontroller based
module that is responsible of gathering all sensors’ data and
send them via Bluetooth (IEEE 802.15) to the fuzzification
server. Microcontroller based module is programmed to collect
sensors' data.
C. Data Classification, Fuzzification and Processing Unit:
All the data received from the sensor is imported to text file
(CSV, Comma Separated format) as shown in Fig.2. The data
is classified by the fuzzy server (a machine prepared with
Matlab Fuzzy Logic Tool Box) where the Fuzzy Inference Sets
(FIS), Fuzzy Membership Functions and Fuzzy Rules are
created and prepared to receive sensors' data as shown in the
Figures below. (Fig.3 to Fig.8 for firefighting module and
Fig.11 to Fig.17 for intrusionr detection module).
Fig. 2 FIDFUZ Captured Sensors' data by Microcontroller
based Module and Received by Server's Bluetooth Connection
Fig. 3 FIDFUZ Firefighting Fuzzy Inference Systems (FIS)
Fig. 1 FIDFUZ Architecture
Fig. 4 FIDFUZ Firefighting Fuzzy Membership Functions
(Temperator Sensor)
Fig. 5 FIDFUZ Firefighting Fuzzy Membership Functions
(Humidity Sensor)
Fig. 6 FIDFUZ Firefighting Fuzzy Membership Functions
(Gas Sensor)
Fig. 7 FIDFUZ Firefighting Fuzzy Membership Functions
(Fire Alarm Output)
Fig. 8 FIDFUZ Firefighting Fuzzy Rules Text Viewer
Fig. 9 FIDFUZ Firefighting Fuzzy Rules Graphical Viewer
Fig. 10 FIDFUZ Firefighting Fuzzy Rules 3D Viewer
Fig. 11 FIDFUZ Intruder Detection Fuzzy Inference Systems
(FIS)
Fig. 12 FIDFUZ Intruder Detection Fuzzy Membership
Functions (Ultrasonic Sensor)
Fig. 13 FIDFUZ Intruder Detection Fuzzy Membership
Functions (PIR Sensor)
Fig. 14 FIDFUZ Intruder Detection Fuzzy Membership
Functions (Noise Sensor)
Fig. 15 FIDFUZ Intruder Detection Fuzzy Membership
Functions (Intruder Alarm Output)
Fig. 16 FIDFUZ Intruder Detection Fuzzy Rules (Text
Viewer)
Fig. 17 FIDFUZ Intruder Detection Fuzzy Rules (Graphical
Viewer)
Fig. 18 FIDFUZ Intruder Detection Fuzzy Rules (3D Surface
Viewer)
Each sensor and the output have their FIS and Membership
functions. The function is composed of three interleaved
intervals in order to create the intersections between different
situations which is very beneficial in the
fuzzification/defuzzification processes.
D. Data Classification, Fuzzification and Processing Unit:
All the data received from the sensor is imported to text file
(CSV, Comma Separated format) as shown in Fig.2. The data
is classified by the fuzzy server (a machine prepared with
Matlab Fuzzy Logic Tool Box) where the Fuzzy Inference Sets
(FIS), Fuzzy Membership Functions and Fuzzy Rules are
created and prepared to receive sensors' data as shown in the
Figures below. (Fig.3 to Fig.8 for firefighting module and
Fig.11 to Fig.17 for intrusionr detection module). Each sensor
and the output have their FIS and Membership functions. The
function is composed of three interleaved intervals in order to
create the intersections between different situations which is
very beneficial in the fuzzification/defuzzification processes.
E. Decision Control and Output Unit
This unit is responsible for alarm activation/deactivation in
correspondence with the fuzzy server recommendations. If the
fuzzy server gives high possibility for danger situation (0.7 to
1.3 rate in case of firefighting and 1.0 to 1.7 rate in case of
intrusion detection). It follows that an alarm is activated
locally synchronously with a remote SMS notification sent to
the household and local authorities via GSM module connected
to the Microcontroller based Module, to start acting against the
dangerous situation detected.
V. EXPERIMENTAL
RESULTS
FIDFUZ was examined in different dangerous situations,
that were intentionally created to testthe system performance.
This is shown in Fig. 8, Fig. 9, Fig. 16 and Fig.17, the output of
the fuzzy server has two different formats. Fig. 8 and Fig. 17
show the rules interference and their effect on the output rate
for firefighting and Intruder detection modules respectively.
Fig. 9 and Fig. 18 show the rules interference and their effect
on the output rate for firefighting and Intruder detection
modules in a 3D surficial shape correspondingly. For
firefighting module, FIDFUZ sensesthe different situations
according to Table 1.
TABLE I. D
IFFEREN T SITUAT IONS SENSE D BY
FRIDFUZ
Situatio
n
Classifi
cation
Fuzzifica
tion
Server
Output
Temperatur
e (In
Celcius)
Average
Humidit
y Range
CO
2
Gaz
Level
Normal < 0.7 <40 (20, 30) <250
(No
Danger)
(0.7 1.3) >40 (0, 30) >250
As for intruder detection module, FIDFUZ is senses normal
(no danger) situation when fuzzification server output is less
than 1 corresponding to low PIR rate (< 1), fixed ultrasonic
reading (> 5 m) critical situations and low noise level (> 2.3
volts). Whereas, it detects danger situation when fuzzification
server output is > 1 and < 1.7 corresponding to high PIR rate (>
1), varied ultrasonic reading level (< 5 m) and high noise level
(< 2.3 volts). The results of FIDFUZ were compared to the real
experimental situations we’ve created to test the system
performance and it had matched with 95% True Positives rate
and 5% False Positives rate. The system also recorded 3%
False Negatives and 97% True Negatives.
VI. CONCLUSIONS
AND
FUTURE
WORK
In this paper, we focused on the area of smart home
monitoring critical situations, in which environmental
information is automatically collected with the help of sensors
and processed by special algorithms and fused in order to make
good decisions about dangerous situations at home. The
projected FIDFUZ is based on fuzzy logic that represents a fast
and easy tool for the interpretation of the fuzzy decision
regarding two different dangerous situations; firefighting and
intruder detection. It has been noticed that fuzzification process
of the input features caused a great impact on the final decision
process. It is possible to increase the performance further by
adding more related input variables and with more data to
enrich the knowledge in the rules.
Our proposed FIDFUZ was implemented and validated
both through simulations and by using real data. The
experimental results were accurate and robust. Amongst the
advantages of our proposed method, is the low computational
overhead which is inherited from the characteristics of fuzzy
systems. This approach allows the easiest combination between
data and adding other sensors. In the near future, FIDFUZ will
be re-implemented with other groups of sensors that can detect
different types of environmental dangerous situations
R
EFERENCES
[1] Cuka, Miralda, et al. "An Integrated Intelligent System for IoT Device
Selection and Placement in Opportunistic Networks Using Fuzzy Logic
and Genetic Algorithm." Advanced Information Networking and
Applications Workshops (WAINA), 2017 31st International Conference
on. IEEE, 2017.
[2] Bhatnagar, Anisha, Shipra Shukla, and Namrata Majumdar. "Machine
Learning Techniques to Reduce Error in the Internet of Things." 2019
9th International Conference on Cloud Computing, Data Science &
Engineering (Confluence). IEEE, 2019
[3] Salman, Tara, and Raj Jain. "A survey of protocols and standards for
internet of things." arXiv preprint arXiv:1903.11549 (2019).
[4] Lin, YuǦHsiu. "Novel smart home system architecture facilitated with
distributed and embedded flexible edge analytics in demandǦside
management." International Transactions on Electrical Energy Systems
(2019): e12014..
[5] Tanwar, Sudeep, et al. "An advanced Internet of Thing based security
alert system for smart home." 2017 International Conference on
Computer, Information and Telecommunication Systems (CITS). IEEE,
2017.
[6] Hinger, Ritik, and Rejo Mathew. "Smart Home Automation Systems."
International conference on Computer Networks, Big data and IoT.
Springer, Cham, 2018.
[7] Tsai, C.-F.; Hung, K.-C. Campus greenhouse monitoring with a simple
ZigBee-based sensor network. In Proceedings of the 2016 International
Conference on Advanced Materials for Science and Engineering
(ICAMSE), Tainan, Taiwan, 12–13 November 2016.
[8] Shariff, F.; Rahim, N.A.; Hew, W.P. Zigbee-based data acquisition
system for online monitoring of grid-connected photovoltaic system.
Expert Syst. Appl. 2015, 42, 1730–1742.
[9] Han, J.; Choi, C.-S.; Park, W.-K.; Lee, I.; Kim, S.-H. Smart home
energy management system including renewable energy based on
ZigBee and PLC. IEEE Trans. Cons. Elect. 2014, 60, 198–202.
[CrossRef]
[10] Arora, A.; Khera, A. Wi-Fi enabled personal computer network
monitoring system using smart phone with enhanced security measures.
Procedia Comput. Sci. 2015, 70, 114–122.
[11] Liu, F.; Zhao, H. The design of WiFi-based smart home communication
hardware adapter. In Proceedings of the 2015 Fifth International
Conference on Instrumentation and Measurement, Computer,
Communication and Control (IMCCC), Qinhuangdao, China, 18–20
September 2015.
[12] Raza, S.; Misra, P.; He, Z.; Voigt, T. Building the Internet of Things
with Bluetooth smart. Ad Hoc Netw. 2017, 57, 19–31.
[13] Yufeng, Z.; Ruqiao, J. Design and realization of the smart home control
system based on the Bluetooth. In Proceedings of the 2015 International
Conference on Intelligent Transportation, Big Data and Smart City
(ICITBS), Halong Bay, Vietnam, 19–20 December 2015.
[14] Collotta, M.; Pau, G. Bluetooth for Internet of Things: A fuzzy approach
to improve power management in smart homes. Comput. Electr. Eng.
2015, 44, 137–152.
[15] JasiĔski, M., P. Majtczak, and A. Malinowski. "Fuzzy logic in decision
support system as a simple Human/Internet of Things interface for shunt
active power filter." Bulletin of the Polish Academy of Sciences
Technical Sciences 64.4 (2016): 877-886.
[16] Vijayalakshmi, S. R., and S. Muruganand. "Internet of Things
technology for fire monitoring system." (2017).
[17] Santamaria, Amilcare Francesco, et al. "A two stages fuzzy logic
approach for Internet of Things (IoT) wearable devices." Personal,
Indoor, and Mobile Radio Communications (PIMRC), 2016 IEEE 27th
Annual International Symposium on. IEEE, 2016.
[18] B. Avitall, B. Peterson, J. Kletch , E. B. Grisworld, and P. Moran,
“Remote health monitoring system,” U.S. Patent 6,171,237, January 9,
2001.
[19] N. Alshurafa, J. Eastwood, S. Nyamathi, W. Xu, J. J. Liu, M.
Pourhomayoun, H. Ghasemzadeh, and M. Sarrafzadeh, “Battery
optimization in smartphones for remote health monitoring systems to
enhance user adherence,” in Proc. 7th International Conference on
Pervasive Technologies Related to Assistive Environments, 2014.
[20] J. Yick, B. Mukherjee, and D. Ghosal. ”Wireless sensor network
survey,” Computer Networks, vol. 52, no. 12, 2008, pp. 2292-2330.
[21] Janella, N. Bui, A. Castellani, L. Vangelista, and M. Zorzi, “Internet of
Things for Smart Cities,” IEEE Internet of Things, vol. 1, no. 1, pp. 22–
32, 2014.
[22] Elmustafa, Sayed Ali Ahmed, and Elbagir Yousef Mujtaba. "Internet of
things in Smart Environment: Concept, Applications, Challenges, and
Future Directions." World Scientific News 134.1 (2019): 1-5
[23] N. K. Suryadevara, S. C. Mukhopadhyay, S. D. T. Kelly, and S. P. S.
Gill, “WSN-Based Smart Sensors and Actuator for Power Management
in Intelligent Buildings," IEEE/ASME Transactions on Mechatronics,
vol. 20, no. 2, pp. 564-571, 2015.
[24] Konstantinous at.el, "An Automated Fire Detection and Alerting
Application based on Satellite and Wireless Communications", 2010 5th
Advanced Satellite Multimedia Systems Conference and the 11th Signal
processing for space communications workshop.
[25] S. R. Vijayalakshmi and S. Muruganand , "Challenges in Integrating
Wireless Sensor Network and Internet of Things for Environmental
Monitoring", World Scientific News,Vol. 41, 2016, pp.8-15.
