Forest Fire Detection System based on Low-Cost Wireless Sensor

Network and Internet of Things

ALI AL-DAHOUD

Department of Computer Science,

Al Zaytoonah University of Jordan,

Airport St, Amman,

JORDAN

MOHAMED FEZARI

Department of Computer Science,

Badji Mokhtar Annaba University,

ALGERIA

AHMAD AA ALKHATIB

Department of Cyber Security,

Al Zaytoonah University of Jordan,

Airport St, Amman,

JORDAN

MOHAMED NADIR SOLTANI

Department of Computer Science,

Badji Mokhtar Annaba University,

ALGERIA

AHMED AL-DAHOUD

Faculty of Architecture and Design, Department of Multimedia,

Al Zaytoonah University of Jordan,

Airport St, Amman,

JORDAN

Abstract: - Forest fires are one of the most devastating natural disasters that can have a significant impact on

the environment, economy, and human lives. Early detection and prompt response are crucial to minimize the

damage caused by forest fires. In recent years, Wireless Sensor Networks (WSN) and Internet of Things (IoT)

technologies have emerged as promising solutions for forest fire detection due to their low-cost and efficient

monitoring capabilities. This paper proposes a low-cost forest fire detection system based on WSN and IoT.

The system uses a network of sensor nodes that are strategically placed in the forest to monitor environmental

conditions such as temperature, humidity, and smoke. The sensor data is transmitted to a central server, where

advanced algorithms are used to detect and predict the occurrence of forest fires. The system provides real-time

alerts to forest authorities and users using a mobile application that shows the fire maps and the current updates.

The proposed system has been evaluated using based on experiments, and the results show that it can

effectively detect forest fires with high accuracy, low false alarms, and low cost. This system has the potential

to provide an efficient and cost-effective solution for forest fire detection and can play a vital role in protecting

the environment and saving lives.

Key-Words: - Wireless Sensor Networks, Sensors, Internet of Things, Forest Fire Detection

Received: December 19, 2022. Revised: April 9, 2023. Accepted: April 22, 2023. Published: May 26, 2023.

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2023.19.49

Ali Al-Dahoud, Mohamed Fezari,

Ahmad AA Alkhatib,

Mohamed Nadir Soltani, Ahmed Al-Dahoud

E-ISSN: 2224-3496

506

Volume 19, 2023

1 Introduction

Throughout 2021 and the summer of 2022, the

planet experienced a significant number of

destructive forest fires that spread across various

regions and inflicted massive harm and damage to

the environment. These fires caused the intrusion of

thousands of miles and emitted greenhouse gases.

Forest fires in July 2021, for instance, resulted in the

release of 350 million tons of CO2, marking the

highest level on record, while Siberia's hundreds of

forests also discharged record amounts of CO2. In

the same vein, Algeria suffered a series of fires in

the north of the country in July and August 2021,

causing substantial destruction. These forest fires

destroyed over 89,000 hectares of land in 35 wilayas

of the country, including areas like Tizi Ouzou,

Bejaia, Bouira, Setif, Blida, Médéa, El-Tarf,

Khenchela, Guelma, Tébessa, Tiaret, and Skikda,

where local authorities and the Ministry of Defense

recorded a total of 1,186 fires that claimed at least

90 lives, including 33 soldiers. The fires also

ravaged the agro-pastoral sector, wiping out

vegetation cover and animal resources. The extent

of the damage caused by the fires fluctuates from

year to year, with the yearly losses relating to the

commercial value of wood and cork estimated at

between 1 and 1.5 billion DA, according to the

Ministry of Agriculture, [1].

Satellite monitoring systems are utilized for

identifying forest fires, but they are only capable of

detecting fires when they have already spread over a

vast region. As a result, these methods are not very

effective, and approximately 80% of losses occur

due to the delayed detection of fires, as per a survey.

To tackle this issue, wireless sensor networks

(WSN) and Internet of Things (IoT) technologies

are being employed, [2], [3].

In an IoT-based system for detecting forest

fires, diverse sensors are deployed across the forest

area. Each node tracks the surrounding area of the

forest and gathers details such as temperature,

humidity, and gas levels. The collected data is then

transmitted to centralized cluster heads, and this

process continues until the information reaches the

router. Each cluster head supervises its designated

region and acts as a go-between for exchanging data

between neighbors and the router. The router

collects all the data gathered within the Rural

Community Security Framework (RCSF) and

transfers it to the receiver or base station through the

cloud, [4].

The work aims to propose a low-cost forest fire

detection system using Wireless Sensor Networks

(WSN) and Internet of Things (IoT) technologies.

The system utilizes a network of strategically placed

sensor nodes in the forest to monitor environmental

conditions and detect the occurrence of forest fires.

The data is transmitted to a central server, where

advanced algorithms are used to predict and detect

forest fires. The proposed system aims to provide

real-time alerts to forest authorities and users using

a mobile application that shows fire maps and

current updates.

2 Literature Review

In recent years, there has been an increasing interest

in the development of low-cost and efficient forest

fire detection systems using Wireless Sensor

Networks (WSN) and Internet of Things (IoT)

technologies. The primary goal of these systems is

to detect forest fires early and prevent them from

causing significant damage to the environment,

economy, and human lives, [5].

One of the significant advantages of WSN-

based forest fire detection systems is their ability to

monitor environmental conditions such as

temperature, humidity, and smoke levels in real-

time. This allows for the early detection of potential

forest fires, enabling authorities to take immediate

action to prevent them from spreading. Several

studies have proposed WSN-based forest fire

detection systems that utilize various environmental

sensors to detect changes in temperature, humidity,

and smoke levels, [6].

For instance, [7], proposed a forest fire

detection system based on a wireless sensor network

that utilizes temperature, humidity, and smoke

sensors to detect and locate forest fires accurately.

The system employs a novel algorithm that

combines wavelet packet decomposition and

principal component analysis to detect the fire

signals accurately, [7].

In addition to WSN, IoT technologies have also

been proposed for forest fire detection systems. The

integration of IoT devices and cloud computing

platforms can provide a powerful solution for forest

fire detection and management. For example, [8]

proposed an IoT-based forest fire detection system

that utilizes environmental sensors and a cloud

computing platform to detect and track forest fires

in real-time. The system employs an image

recognition algorithm that can detect the presence of

smoke and fire in real-time images captured by

cameras installed in the forest. [8].

The literature review by [9], provides a

comprehensive overview of various forest fire

detection techniques, including visual detection,

infrared detection, and acoustic detection. The

review also discusses wireless sensor networks

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2023.19.49

Ali Al-Dahoud, Mohamed Fezari,

Ahmad AA Alkhatib,

Mohamed Nadir Soltani, Ahmed Al-Dahoud

E-ISSN: 2224-3496

507

Volume 19, 2023

(WSNs) and their potential for forest fire detection,

[9].

In [10], the author proposes a smart and low-

cost technique for forest fire detection using WSNs.

The proposed technique uses temperature, humidity,

and light sensors to detect fire and send an alert to a

base station, [10].

In another paper, [11], the author proposes a

WSN-based forest fire detection and decision-

making system. The system consists of a sensor

network that collects data and sends it to a decision-

making module, which uses fuzzy logic to analyze

the data and decide whether a fire has occurred,

[11].

In [12], the authors present a smart system for

forest fire detection using a WSN. The proposed

system uses temperature and smoke sensors to

detect fires and send alerts to a control centre, [12].

In the book chapter. [13], the authors discuss forest

fire monitoring and the role of WSNs in early

detection and prevention, [13].

Finally, [14], propose a multivariate outlier

detection technique to improve the accuracy of

forest fire data aggregation in WSNs. The proposed

technique uses principal component analysis and the

Mahalanobis distance to detect outliers in the

collected data, [14].

Overall, the literature by [9], [10], [11], [12],

[13], [14], emphasizes the potential of WSNs for

forest fire detection and proposes various techniques

and systems to improve the accuracy and efficiency

of detection.

Moreover, several studies have proposed the use

of machine learning algorithms to enhance the

accuracy of WSN and IoT-based forest fire

detection systems. For instance, [15], proposed a

forest fire detection system that utilizes an artificial

neural network (ANN) algorithm to detect forest

fires accurately. The system utilizes temperature and

humidity sensors to collect environmental data,

which is then fed to the ANN algorithm for fire

detection, [15].

In conclusion, WSN and IoT-based forest fire

detection systems have the potential to provide an

efficient and cost-effective solution for early forest

fire detection and prevention. These systems utilize

environmental sensors, machine learning

algorithms, and cloud computing platforms to detect

and locate forest fires accurately, enabling

authorities to take immediate action to mitigate the

damage caused by forest fires.

The main difference that distinguishes our study

is that the first one proposes a low-cost forest fire

detection system based on WSN and IoT, which

uses a network of sensor nodes that monitor

environmental conditions such as temperature,

humidity, and smoke levels, and transmit the data to

a central server for real-time analysis and alerts. The

proposed system employs advanced algorithms with

simple components to detect and predict the

occurrence of forest fires and provides real-time

alerts to forest authorities and users through a

mobile application.

3 Proposed Solution

In the forest of El-Kala, located in the extreme

northeast of Algeria, a dense network of sensor

nodes has been installed. These nodes are organized

into clusters, with each node having a corresponding

"cluster head." The sensors are capable of

measuring ambient temperature, relative humidity,

and smoke levels. To determine their locations, each

sensor node is geotagged using Google Maps. Every

node sends its measurement data, along with its

location information (an identifier), to its

corresponding group header. The group header then

collects the data from all the nodes in its group and

transmits it to the gateway. This gateway is

connected to the internet and transfers all the

collected data from the network via WiFi.

Additionally, users have the option of inquiring

about the current temperature and humidity data for

a specific cluster area through Cloud-IOT. Further

details concerning sensor nodes and WSN

management will be covered in subsequent pages of

this document.

Figure 1 shows bellow the proposed system

Fig. 1: Proposed WSN solution for Forest fire

detection

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2023.19.49

Ali Al-Dahoud, Mohamed Fezari,

Ahmad AA Alkhatib,

Mohamed Nadir Soltani, Ahmed Al-Dahoud

E-ISSN: 2224-3496

508

Volume 19, 2023

4 Hardware Description

Based on the main architecture used in WSN design,

[16], [17], the principal parts of the node in our

design, as shown in Figure 2, are the CPU, sensors,

communication mode, and power module.

Fig. 2: WSN node architecture and designed

prototype

a. The ESP-32 is the main component of the node

design. The ESP32-WROOM-32, [18], is a highly

capable, generic Wi-Fi+BT+BLE MCU module

system-on-a-chip (SoC) microcontrollers are based

on Tensilica's Xtensa LX6 architecture and integrate

dual-mode Wi-Fi and Bluetooth management, as

well as a DSP. It is the successor to the ESP8266

and is ideal for wearable and mobile electronics, as

well as Internet of Things (IoT) applications. Its

features range from low-power sensor networks to

high-performance tasks such as voice encoding,

music streaming, and MP3 decoding. The ESP32

includes two low-power 32-bit LX6 Xtensa

microprocessors operating at either 160 MHz or 240

MHz. Its internal memory consists of [19], [20]:

448 KB of ROM for boot and main functions.

520 KB of on-chip SRAM for data and instructions.

8 KB of SRAM in RTC (Real-Time Clock) that is

called RTC FAST Memory, which can be used for

data storage and is accessed by the main CPU

during RTC boot from Deep-sleep mode.

8 Kbit of SRAM in RTC that is called RTC SLOW

Memory and is accessed by the ULP (Ultra-Low

Power) coprocessor during Deep-sleep mode.

1 Kbit of Fuse, with 256 bits used for the system

(MAC address and chip configuration), and the

remaining 768 bits reserved for client applications,

including flash encryption and chip ID.

b. Main sensors used in the application:

B1. Flame Sensor: This is a sensor capable of

detecting flames with wavelengths between 760 nm

and 1100 nm. This sensor uses infrared as a

transducer to detect flame conditions. This fire

sensor has a reading angle of 60 degrees and

normally operates at a temperature of 25 to 85

degrees Celsius, [21].

B2. The interface of the BME280 sensor measures

temperature, humidity, and pressure, and it is

straightforward to use since it does not require any

extra components and comes pre-calibrated. This

means that you can start measuring relative

humidity, temperature, barometric pressure, and

even approximate altitude very quickly. The

BME280 sensor is manufactured by Bosch and is a

next-generation digital temperature, humidity, and

pressure sensor that supersedes previous sensors

such as BMP180, BMP085, and BMP183, [22].

B3. The MQ-2 Gas Sensor is a low-cost sensor that

can detect various flammable vapors such as

propane and smoke, as well as natural gas. It is

highly sensitive and suitable for detecting smoke.

The sensor's sensitive material, SnO2, has lower

conductivity in pure air, but when flammable gas is

present, the sensor's conductivity increases with the

gas concentration. A straightforward circuit can

convert the change in conductivity to an output

signal that corresponds to the gas concentration,

making it easy for users to obtain gas concentration

information, [23], [24].

To conserve energy, the sensor nodes send short

message packets. Each packet includes a unique

identification number (NI) for each card, starting at

1. The ID number is sent along with the temperature

and humidity variables, which are floating-point

variables that occupy 8 bytes. The flame, smoke,

and prediction indicators are each composed of a

vector of 32 characters, occupying a total size of 32

bytes (with each character occupying 1 byte).

The data packet structure is defined by the following

variables: Identification number (NI) - 4 bytes,

Temperature in °C - 8 bytes, Humidity in % - 8

bytes, Flame indicator - 32 bytes, Smoke indicator -

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2023.19.49

Ali Al-Dahoud, Mohamed Fezari,

Ahmad AA Alkhatib,

Mohamed Nadir Soltani, Ahmed Al-Dahoud

E-ISSN: 2224-3496

509

Volume 19, 2023

32 bytes, Prediction - 32 bytes, Number of readings

- 4 bytes.

The temperature and humidity variables are

floating-point numbers that occupy 8 bytes each,

while the flame, smoke, and prediction indicators

are 32-character vectors occupying 32 bytes each.

The "number of readings" variable is an integer with

a size of 4 bytes.

Here is the flowchart of implemented software in

Figure 3:

Fig. 3: Implemented software on ESP-32

T: temperature, H: Humidity, Fd: fire detect, Fl: Flame,

ID: id node

4 Fire Prediction (Risk Rate)

Based on statistics related to fire detection in these

forest regions, we have developed a model of risk

rate. As the nodes retrieve data in each cycle, they

capture temperature, humidity, and flame

parameters, and calculate the percentage of risk of

fire presence. This information is then sent to the

group header as shown in below.

The indicators corresponding to each detection

parameter (Flame, Humidity, Temperature, Smoke)

are set to 1 if their threshold is crossed.

The temperature indicator is set to 1 if the

temperature is greater than or equal to 40°C.

The humidity indicator is true if the humidity is less

than 40%.

The smoke indicator turns on if smoke is detected.

The flame indicator lights up if a flame is detected.

The specified values for the system installation are

dependent on factors such as location, wind

patterns, and season. Hence, they must be

customized accordingly. To illustrate, in the case of

a region that experiences seasonal precipitation and

relatively low humidity, we have set the limits for

humidity at 40% RH, and the threshold for heat in

the shade at 40°C.

-If the flame detector rises high, the fire alert is

increased by 50% because the presence of flames at

tree branch height is extremely dangerous and a

clear indication of a forest fire.

-If smoke is detected, which is the case in most

scenarios because forest fires ignite mainly at grass

level, and smoke is the first element that rises, the

fire risk increases by 25% by putting it in the second

priority position after the flame.

-If the temperature measured by the BME280 is

high (> 40°C), the indication of a forest fire

increases by 12.5%.

-If the humidity is less than 40%, the fire warning is

raised to 6.25% because fires spread easily when the

air is dry.

Therefore, the risk of fire occurrence can be divided

into three zones:

a. Green zone: If no smoke or flame is detected,

the percentage of fire occurrence in this zone is

between 0% and 25%.

b. Orange zone: If no flame is detected, but the

smoke indicator is high, the risk of fire

occurrence is between 25% and 50%.

c. Red zone: If only the flame indicator is detected

and the smoke indicator is low or both

parameters are detected, the percentage of fire is

between 50% and 100%.

The threshold crossing of temperature and

humidity influences each zone. The percentage of

influence of temperature is higher than that of

humidity. For example, if the temperature is low,

there is a chance that there is a campfire whose

smoke reaches the sensors. However, if the

temperature regularly rises above 40°C, the fire

alarm is increased. If no flames or smoke are

detected, but the temperature is high and the

humidity is low, the risk of fire is increased, and the

corresponding alert is sent.

5 Experiment

A series of experiments were performed with a

sampling period of 3 seconds. The objective was to

study the best performance for fire detection under

certain climatic conditions. In this stage, a sensor

node was installed outside in the shade in the

absence of fire. The values of the parameters

presented below were obtained using the "serial

monitor" of the Arduino IDE software. In a second

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2023.19.49

Ali Al-Dahoud, Mohamed Fezari,

Ahmad AA Alkhatib,

Mohamed Nadir Soltani, Ahmed Al-Dahoud

E-ISSN: 2224-3496

510

Volume 19, 2023

scenario, the node was exposed to fire, and the

values of the same parameters were recorded. Then,

the data processing done at each node with the

ESP32 gives the percentage of the presence of a fire.

The graphs in Figure 4 illustrate the results of

temperature and humidity in the experiment of the

realized model: The x-axis contains the number of

readings taken every 3 seconds, and the y-axis

contains the temperature and humidity level as a

function of the number of readings. As the number

of iterations increases, the accuracy of the model

improves.

The test was carried out during the afternoon for

four parameters: temperature, relative humidity,

flame, and smoke, taking into account the climatic

conditions in the absence of fire, where the

approximation is: temperature = 27°C; humidity =

49%, flame: not detected, smoke: not detected. The

experiment lasted 75 seconds. The fire was lit after

15 seconds (5th reading number) at a distance of

one meter from the sensor node. The flame was

detected instantly, and the temperature and humidity

values began to change slowly. After 18 seconds,

the smoke indicator lit up, and the temperature

rapidly rose to 48°C while the humidity dropped to

21%. The risk rate changes according to the

threshold clearance of environmental parameters,

including temperature, relative humidity, flame, and

smoke, which are monitored by the system under

the previously mentioned climatic conditions.

Fig. 4: Humidity and temperature evolution in

experiment case

6 Web Platform Design

The monitoring of the monitored area can be done

from any location. It is enough for the user to be

connected to the Internet, enter the website, and

view the results. The server created at the gateway

has the role of transmitting the data to the base

station. The variables of each node of the network

illustrated in figure 4 are displayed to the user in

five frames. The first frame concerns the prediction

interval where the risk rate in % is expressed in

figure 5 A "Flame detected!" alert message will be

sent to the base station, displayed in the second

frame of the platform. A "Smoke detected!" alert

message will be sent to the base station in the third

frame. In the fourth and fifth frames, the

temperature in °C, the humidity in %, and the

"Reading Number:" are displayed as shown in

Figure 5.

Fig. 5: Web interface GUI designed

7 Conclusion

This study shows that WSN technology is a

promising and eco-friendly option for effectively

detecting forest fires in the country with more

accurate results. The study also concludes that the

proposed algorithm for calculating fire risk rate is

efficient and algorithmically simple compared to

others, as demonstrated through a real field

experiment during summer seasons. Future research

will focus on extending the model to include an

experimental evaluation of energy consumption to

keep the network operational for longer periods,

which could be achieved through a secondary power

supply with rechargeable batteries powered by a

solar panel. Additionally, distributing nodes into

groups and utilizing distributed sensing to find the

best information paths for real-time communication

could improve the system. Finally, embedding an

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2023.19.49

Ali Al-Dahoud, Mohamed Fezari,

Ahmad AA Alkhatib,

Mohamed Nadir Soltani, Ahmed Al-Dahoud

E-ISSN: 2224-3496

511

Volume 19, 2023

artificial intelligence algorithm could further

enhance the system design.

References:

[1] Ministère de l'Agriculture et du

Développement rural (2021). "Bilan des feux

de forêts en Algérie en 2021". [online]

Available at:

https://www.minagriculture.gov.dz/index.php/

communication/actualites/1413-bilan-des-

feux-de-forets-en-algerie-en-2021 [Accessed

27 Feb. 2023].

[2] Adam, F., Ahmed, M.U., Hasan, M.R. and

Hasan, M.K., 2020. Forest fire detection and

monitoring system using wireless sensor

network and internet of things. In 2020 11th

International Conference on Computing,

Communication and Networking Technologies

(ICCCNT) (pp. 1-6). IEEE.

[3] Borah, J. and Sarma, K.K., 2019. Real-time

forest fire detection and alert system using

wireless sensor network. Journal of Ambient

Intelligence and Humanized Computing,

10(3), pp.1099-1114.

[4] Zhang, X., Li, W., Li, W. and Li, X., 2018. A

review of forest fire monitoring techniques

based on wireless sensor networks. Sensors,

18(5), p.1639.

[5] Masoud, M., Jaradat, Y., Manasrah, A. and

Jannoud, I., 2019. Sensors of smart devices in

the internet of everything (IoE) era: big

opportunities and massive doubts. Journal of

Sensors, 2019.

[6] Liyang Yu, Neng Wang and Xiaoqiao Meng,

"Real-time forest fire detection with wireless

sensor networks," Proceedings. 2005

International Conference on Wireless

Communications, Networking and Mobile

Computing, 2005. 2005, pp. 1214-1217, doi:

10.1109/WCNM.2005.1 544272.

[7] Zhang, H., Chen, J., & Zhang, C. (2018). A

wireless sensor network-based forest fire

detection system. IEEE Access, 6, 51157-

51166.

[8] Liu, Z., Zhang, H., & Xue, X. (2020). IoT-

based forest fire detection and tracking

system. IEEE Transactions on Industrial

Informatics, 16(9), 5745-5753.

[9] Alkhatib, A.A., 2014. A review on forest fire

detection techniques. International Journal of

Distributed Sensor Networks, 10(3),

p.597368.

[10] Alkhatib, A.A., 2013. Smart and low cost

technique for forest fire detection using

wireless sensor network. International

Journal of Computer Applications, 81(11).

[11] Alkhatib, A.A., 2013. Wireless sensor

network for forest fire detection and decision

making. Int. J. Adv. Eng. Sci. Technol, 2(3),

pp.299-309.

[12] Alkhatib, A.A., Alia, M. and Hnaif, A., 2017.

Smart system for forest fire using sensor

network. International Journal of Security

and Its Applications, 11(7), pp.1-16.

[13] Alkhatib, A.A. and Szmyt, J., 2017. Forest

fire monitoring. In Forest Fire. IntechOpen.

[14] Alkhatib, A.A. and Abed-Al, Q., 2022.

Multivariate Outlier Detection for Forest Fire

Data Aggregation Accuracy. Intelligent

Automation & Soft Computing, 31(2).

[15] Li, Q., Ma, Y., & Sun, H. (2020). Forest fire

detection system based on wireless sensor

network and artificial neural network. IEEE

Access, 8, 234891-234899.

[16] Sharma, D., Verma, S. and Sharma, K.

(2013). Network Topologies in Wireless

Sensor Networks: A Review. IJECT Vol4,

No.3, 2013

[17] Soparia, J., & Bhatt, N. (2014). A survey on

comparative study of wireless sensor network

topologies. International Journal of Computer

Applications, 87(1).

[18] www.espressif.com. (n.d.). Espressif Systems

- Wi-Fi and Bluetooth chipsets and solutions.

[online] Available at:

https://www.espressif.com/en. (Accessed 25

May. 2023).

[19] E. S. Sasmita, M. Rosmiati and M. F. Rizal,

"Integrating Forest Fire Detection with

Wireless Sensor Network Based on Long

Range Radio," 2018 International Conference

on Control, Electronics, Renewable Energy

and Communications (ICCEREC), 2018, pp.

222-225.

[20] Dubey, V., Kumar, P., Chauhan, N. (2019).

Forest Fire Detection System Using IoT and

Artificial Neural Network. In: Bhattacharyya,

S., Hassanien, A., Gupta, D., Khanna, A., Pan,

I. (eds) International Conference on

Innovative Computing and Communications.

Lecture Notes in Networks and Systems, vol

55. Springer, Singapore.

[21] Son, B.; Her, Y.S. A Design and

Implementation of Forest-Fires Surveillance

System based on Wireless Sensor Networks

for South Korea Mountains. IJCSNS Int. J.

Comput. Sci. Netw. Secur. 2005, 6.

[22] Botnroll.com. (n.d.). Flame Sensor with

digital output. [online] Available at:

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2023.19.49

Ali Al-Dahoud, Mohamed Fezari,

Ahmad AA Alkhatib,

Mohamed Nadir Soltani, Ahmed Al-Dahoud

E-ISSN: 2224-3496

512

Volume 19, 2023

https://www.botnroll.com/en/outros/3005-

4duino-relay-flame-sensor.html [Accessed 3

May 2023].

[23] Bosch. BME680—Gas Sensor Measuring

Relative Humidity, Barometric Pressure,

Ambient Temperature and Gas (VOC).

Available online: https://www.bosch-

sensortec.com/products/environmental-

sensors/gassensors-bme680/ (Accessed 27

Feb. 2023).

[24] Anon, (n.d.). Mq-2 (Generale Combustible

Gaz) Shield – Orbit Electronic. [online]

Available at: https://www.orbit-

dz.com/product/mq2/ [Accessed 3 May 2023].

Contribution of Individual Authors to the

Creation of a Scientific Article (Ghostwriting

Policy)

The authors equally contributed to the present

research, at all stages from the formulation of the

problem to the final findings and solution.

Sources of Funding for Research Presented in a

Scientific Article or Scientific Article Itself

No funding was received for conducting this study.

Conflict of Interest

The authors have no conflicts of interest to declare

that are relevant to the content of this article.

Creative Commons Attribution License 4.0

(Attribution 4.0 International, CC BY 4.0)

This article is published under the terms of the

Creative Commons Attribution License 4.0

https://creativecommons.org/licenses/by/4.0/deed.en

_US

WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT

DOI: 10.37394/232015.2023.19.49

Ali Al-Dahoud, Mohamed Fezari,

Ahmad AA Alkhatib,

Mohamed Nadir Soltani, Ahmed Al-Dahoud

E-ISSN: 2224-3496

513

Volume 19, 2023