Fire Detection System in Peatland Area Using LoRa WAN

Communication

Evizal Abdul Kadir

, Hitoshi Irie

, and Sri Listia Rosa

Department of Informatics Engineering, Universitas Islam Riau, Pekanbaru, Indonesia

Center for Environmental Remote Sensing (CEReS) Chiba University Chiba, Japan

Keywords:

Smart Sensor Node, WSNs, Pollution, River Water.

Abstract:

Land and forest ﬁres are one of the threats in a tropical country, especially in Indonesia with forestry land and

additional caused of type of land which peatland that easy to getting ﬁre in the summer season. Currently,

many techniques to detect ﬁre hotspot and land ﬁre but some of the technique can not apply in peatland case.

This research proposes a new technique that can be applied to this case in Riau province, Indonesia which the

land with peat type. Long Range Wide Area Network (LoRa WAN) used in the detection land and forest ﬁre,

with advantages of low power and long-range transmission in LoRA WAN very applicable in this detection

of ﬁre with the distance of fore hotspot very far and large of an area. The simulation result shows good

performance and veriﬁcation used mathematical modeling to check that the system is working and application

to implement. The sensors deployed in the area which indicate for a forest ﬁre in the simulated distance to

detect the potential of ﬁre then the information sent to the monitoring system in the data center. The proposed

LoRa WAN method gives good response and recommended to implement in the peatland area which located

in Riau Province, Indonesia.

1 INTRODUCTION

The signiﬁcant emerging and development of tech-

nology in wireless network has expressively changed

and enhance the natural environment control system

compared to current methods that use satellite ground

detection methods, such as wireless sensor networks.

thread (WSN) (Khajuria and Gupta, 2015). This sys-

tem can provide new data for environmental and po-

tentially fatal warning applications such as land and

forest research and ﬂood detection. The beneﬁts of

ground level detection can be divided into three as-

pects (Chee-Yee and Kumar, 2003; Boubiche et al.,

2018; Jie et al., 2015).

Sensor button; low cost, low power, strong, low

pollution and environmental disturbance; communi-

cation; low data rate, remote detection and correction

and errors; Information processing; nodes, microcon-

trollers and small operating systems for low power

systems. With the advent of IoT technology and Long

Range (LoRa) (Wixted et al., 2016; Lavric and Pe-

trariu, 2018; Carvalho et al., 2018), WSN and connec-

tivity are becoming more reliable, stronger and faster.

With this technology it is possible to develop intelli-

gent monitoring systems to detect forest and land ﬁres

(Lee and Ke, 2018; Kadir, 2017; Kadir et al., 2018).

Therefore, this research focuses on developing in-

telligent ﬁre detection systems, especially in the peat

ﬁeld, based on the detection and monitoring of envi-

ronmental behavior in terms of temperature, humid-

ity and gas. To provide new methods and technolo-

gies for detection and surveillance systems that use

low-power wireless data communications with LoRa

WAN technology. Sensor integration with LoRa

WAN technology affects local communities where

users have access to real-time database information

at any time.

The method for detecting the surface of the earth

will be used in other regions, regions and regions in

Indonesia. This solution is a faster and cheaper al-

ternative to obtaining satellite data currently in use,

which will certainly beneﬁt social welfare and eco-

nomic development. In addition, developing real-

time perception will require some support from pol-

icy makers to understand how the system works and

at the same time to understand the outcome models to

take appropriate action.

130

Kadir, E., Irie, H. and Rosa, S.

Fire Detection System in Peatland Area using LoRa WAN Communication.

DOI: 10.5220/0009145101300134

In Proceedings of the Second International Conference on Science, Engineering and Technology (ICoSET 2019), pages 130-134

ISBN: 978-989-758-463-3

2 LORA WAN DETECTION AND

MONITORING SYSTEM

Detection and monitoring systems are widely used for

objects or parameters that require continuous time.

Currently, many types of monitoring systems are

based on the objectives and parameters to be moni-

tored. Fire detection and environmental monitoring

are carried out in different organizations or organi-

zations to verify the current environmental situation.

Commonly used technology is the use of satellite data

to detect ﬁre points.

This technology has a number of drawbacks and

limitations, including proper ﬁre detection and, in

some cases, no possibility of satellite imagery pass-

ing through the clouds. The new method proposed in

this system is designed to detect smoke, temperature,

particle changes, etc. LoRa sensor. He uses the LoRa

WAN, where he is placed in a high risk ﬁre area to

collect data. Figure 1 shows a series of hot spots in

Riau province.

Figure 1: A map of Riau province with number of ﬁre

hotspots used satellite images.

That information collect by the sensor is sent to

the sensor gateway as base station to transmit data

have been collect by the monitoring system, because

of the distance between the sensor and the base station

of the monitoring system. is far enough in some ar-

eas. For correct data, a large numbers of WAN LoRa

sensors are installed in the area because external sen-

sors can transmit up to 15 km. In addition to the LoRa

WAN sensor, it is mounted on each base station with a

high-resolution camera to analyze aerial imagery be-

fore and after the ﬁre, then train the data to analyze

changes in the image of the environment.

Selected sites likely to cause high forest ﬁres have

been identiﬁed in the peat area as shown in Figure

2, installed systems have been approved by local au-

thorities such as the Local Council of Ria and the

Indonesian Ministry of Agriculture. Browser. Envi-

ronment and Forests. Again, the links between Riau

Islamic University and local authorities should be in-

tegrated into the decision-making process and facil-

itate access to installation, monitoring, data analysis

and communication. Improving forest ﬁre monitor-

ing with intelligent ground detection and LoRa WAN

technology can be an early indicator for better disas-

ter risk reduction decision making. This project ben-

eﬁts from new designs and new developments thanks

to the latest LoRa WAN technology and research on

signal transmission.

Conﬁguring sensor base stations in different areas

to gather information from the WAN LoRA sensor

network being developed in the peat area. The infor-

mation collected by the sensor base station is stored

in an internal database and sent to the data center be-

cause the sensor base station is located in remote rural

areas for more than 100 km. Sensors can be detected

and alerted immediately before a ﬁre occurs with the

responsible agency for preventive measures. The next

step will be to conﬁgure some sensors and base sta-

tions that will cover the entire province of Riau and

build this project as a prototype system in another

Indonesian country. The proposed LoRa sensor sce-

narios also allow analysis of the behavior and envi-

ronment before and after the ﬁre is analyzed through

a new image processing method, particle detection,

sensor data and system. the media. Data Figure 3

shows a proposal to implement a data network scheme

to monitor WAN and environmental sensors.

Figure 2: Actual condition of land and forest burn on the

ﬁeld.

Fire Detection System in Peatland Area using LoRa WAN Communication

131

Figure 3: LoRa WAN block diagram for forest ﬁre monitor-

ing system.

3 LORA WAN SOLUTION FOR

PEATLAND MONITORING

The recommended LoRa WAN solution uses a pow-

erful LoRa module made by Semtech technology for

long-term transactions. It is a standard of the LoRa

alliance that creates mechanisms for formatting, pro-

visioning, access agent, message security and protec-

tion, and device management. Figure 4 shows the

LoR WAN that forms a star topology around the gate-

way, which acts as a packet sender between the ter-

minal and the core Network Server (NS). The NS is

to control for manage the Medium Access Control

(MAC) layer for processing and perform as a gate be-

tween application that running to the end of devices

and application server. The standard of LoRa WAN

can be deﬁne in three classes; so the ﬁnal device can

respond to various scenarios such as network topol-

ogy A, B and C (Abeele et al., 2017).

Class A devices often have their own transceiver

in deep sleep conditions and rarely wake up to send

data to NS. Wireless media access in the WAN LoRa

adheres to the ALOHA method, does not use listening

before speaking, and is therefore limited to most parts

of the world when used. In Europe, in sample, the 868

MHz band contains different subbands, with the CDR

between 0.1% and 10% and 1% most common (Zainal

et al., 2017).

3.1 LoRa WAN Networking

Recent technology for sensing system and network

technology is introduced by LoRa WAN with capa-

bility to send in long distance. Furthermore, the low

power transmission make power long life and good

for the maintenance device. As mention in the pre-

vious section, the recommended scenario for network

architecture in this WAN loop is a level network of

architecture that meets traditional internet standards,

such as Internet Protocol version 6 (IPv6). Expect

rapid integration of all LoRa WAN systems and sys-

tems. One-way nodes of the WAN ecosystem are fast

and heavy. However, the transmission capacity of the

WAN LoRa technology is very limited, which means

a limited throughput and a small package size. There-

fore, it is not easy to integrate an IPv6 datagram di-

rectly into LoRa-WAN packages and a compression

mechanism is required. The proposed solution pro-

vides an IPv6 connection to the LoRa node using

the LoRa WAN connection, while a multi-computer

(MEC) -based architecture is used to achieve this in-

tegration: network access, MEC node, bidirectional

ﬂow can be created between LoRa and LoRa, as

shown in FIG. IPv6 is responsible for translating com-

pressed or uncompressed packages into network seg-

ments. WAN and IPv6 (Sanchez-Iborra et al., 2018).

The proposed solution can contribute to the exten-

sive LoRa network, including:

• A true extension of IPR6 in LoRa has been devel-

oped and tested.

• LoRa WAN button LoRa is used in bank testing to

deliver environmental data via IPv6 links to WAN

links.

• The LoR WAN environment for smart environ-

ment detection is conﬁgured to be ready for users.

3.2 LoRa WAN Physical Error Model

Physical error is one of parameter have to check, after

changing the Physical Signal Error Model (PHY), the

output data is cleared to increase the entropy of the

source. Note that in a small simulation the Bit Error

Rate (BER) information is obtained from a balanced

distribution, and therefore the entropy of the source

information is within the maximum limit. Before the

bleached current is sent to the modulator, the modu-

lator is mapped. It generates a whole sequence that

is sent to the LoRa WAN sensor button. At the LoRa

sensor node, the N number of the complex base se-

quence sample is changed to N with the time created

by the phase accumulators indicated by (1), where N

is the sample with a base band symbol equivalent to

2BF (fs. / BW) . An entire entry determines the time

shift (Abeele et al., 2017).

m(i) =



exp(− jπ), if i = 0

m(i − 1)exp( j f (i)), if i = 1, . . . , N − 1

(1)

where the instantaneous of frequency can write as

ICoSET 2019 - The Second International Conference on Science, Engineering and Technology

132

Figure 4: LoRa WAN overview in hierarchical architecture refer to Semtech technology.

Figure 5: The IPv6 Architecture of LoRa WAN networking solution stack.

f (i) is given by

f (i) = −π +

2π, for i = 1, . . . , N − 1 (2)

The number of samples in the WAN LoRa symbol

was then sent via Gaussian White Noise (AWGN) for

a given SNR parameter

c(i) = m(i) +

2SNR

[N (0; 1)+ jN (0;1)]

for i = 0, . . . , N − 1

(3)

where N (0; 1) is the normal of standard distribution

and SNR = 10SNRdB/10. Take note that the energy in

each symbol is one for the LoRa WAN sensor button.

Finally, the LoRa WAN decoding uses a demodule-

based relationship in decoding in which the receive

symbol as associated with all known LoRa symbols.

Symbolic decisions are shown by choosing the LoR

icon with the highest correlation value. The value

of correlation physical error to the model of LoRA

WAN application in land and forest ﬁre detection can

Fire Detection System in Peatland Area using LoRa WAN Communication

133

be write as the number of area going to detect com-

pare to the number of sensor nodes in LoRA WAN

deploy.

4 CONCLUSIONS

A system for detection of land and forest ﬁre use

LoRa WAN technology is proposed. Results show

the simulation and mathematical modeling based on

calculation gives good response and the system ap-

plicable to apply for the alert system in the detection

of the forest ﬁre. LoRa WAN system can send in-

formation in long-distance over than 10 miles, thus

very applicable in the detection of forest ﬁre in large

of an area. The system can be integrating to the sev-

eral sensing systems and collect the information to be-

come a group of information to send to the data center

for monitoring system.

ACKNOWLEDGEMENTS

Authors would like to say thank you very much to

KEMENRISTEKDIKTI Indonesia for funding this

research and Universitas Islam Riau as well as Chiba

University to support the facilities.

REFERENCES

Abeele, F. V. D., Haxhibeqiri, J., Moerman, I., and Hoe-

beke, J. (2017). Scalability analysis of large-scale lo-

rawan networks in ns-3. IEEE Internet of Things Jour-

nal, 4:2186–2198.

Boubiche, D. E., Pathan, A. S. K., Lloret, J., ZHOU, H.,

Hong, S., Amin, S. O., and Feki, M. A. (2018). Ad-

vanced industrial wireless sensor networks and intelli-

gent iot. IEEE Communications Magazine, 56:14–15.

Carvalho, D. F., Depari, A., Ferrari, P., Flammini, A., Ri-

naldi, S., and Sisinni, E. (2018). On the feasibility

of mobile sensing and tracking applications based on

lpwan. IEEE Sensors Applications Symposium (SAS),

pages 1–6.

Chee-Yee, C. and Kumar, S. P. (2003). Sensor networks:

evolution, opportunities, and challenges. In Proceed-

ings of the IEEE, volume 91, pages 1247–1256.

Jie, L., Ghayvat, H., and MUKHOPADHYAY, S. C. (2015).

Introducing intel galileo as a development platform of

smart sensor: Evolution, opportunities and challenges.

In 2015 IEEE 10th Conference on Industrial Electron-

ics and Applications (ICIEA), pages 1797–1802.

Kadir, E. A. (2017). A reconﬁgurable mimo antenna sys-

tem for wireless communications. In 2017 4th Inter-

national Conference on Electrical Engineering, Com-

puter Science and Informatics (EECSI), pages 1–4.

Kadir, E. A., Irie, H., Rahim, S. K. A., Arta, Y., and Rosa,

S. L. (2018). Reconﬁgurable mimo antenna for wire-

less communication based on arduino microcontroller.

In 2018 IEEE International RF and Microwave Con-

ference (RFM), pages 119–122. IEEE.

Kakhandki, A. L., Hublikar, S., and Kumar, P. (2017).

An efﬁcient hop selection model to enhance lifetime

of wireless sensor network. In 2017 Innovations

in Power and Advanced Computing Technologies (i-

PACT), pages 1–5. IEEE.

Khajuria, R. and Gupta, S. (2015). Energy optimization and

lifetime enhancement techniques in wireless sensor

networks: A survey. In International Conference on

Computing, Communication and Automation, pages

396–402.

Lavric, A. and Petrariu, A. I. (2018). Lorawan communi-

cation protocol: The new era of iot. In 2018 Inter-

national Conference on Development and Application

Systems (DAS), pages 74–77.

Lee, H. C. and Ke, K. H. (2018). Monitoring of large-area

iot sensors using a lora wireless mesh network sys-

tem: Design and evaluation. In IEEE Transactions on

Instrumentation and Measurement, pages 1–11.

Sanchez-Iborra, R., Sanchez-Gomez, J., Santa, J., Fernan-

dez, P. J., and Skarmeta, A. F. (2018). Ipv6 commu-

nications over lora for future iov services. In 2018

IEEE 4th World Forum on Internet of Things (WF-

IoT), pages 92–97.

Wixted, A. J., Kinnaird, P., L. H., Tait, A., Ahmdinia, A.,

and Strachan, N. (2016). Evaluation of lora and lo-

rawan for wireless sensor networks. In 2016 IEEE

SENSORS, pages 1–3.

Zainal, N. A. B., H. M. H., Chowdhury, I., and Islam, M. R.

(2017). Sensor node clutter distribution in lora lpwan.

In 2017 IEEE 4th International Conference on Smart

Instrumentation, Measurement and Application (IC-

SIMA), pages 1–6.

ICoSET 2019 - The Second International Conference on Science, Engineering and Technology

134