IRJET- Smart Agriculture Solution using Lora and IoT

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 07 | July 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 1872
“Smart Agriculture Solution using LoRa and IoT”
Shamali K.Makode1, Radhika R. Harne2
1(P.G. Scholar), Electronics System and Communication Engineering GCOEA, Amravati Maharashtra, India.
2(Assistant Professor), Electronics System and Communication Engineering GCOEA, Amravati Maharashtra, India.
---------------------------------------------------------------------***--------------------------------------------------------------------------
Abstract- Traditional agriculture is converting into smart
agriculture due to the bulge of the IoT (Internet of things).
An agriculture service platform is developed to support
environmental monitoring and to improve the efficiency of
agricultural management. In agricultural field like
Greenhouses, various Climatic Condition Parameters are
essential to monitored for regulation of crop production,
and climate change has already shown its negative impact
that has on agriculture and food production, particularly
because of rainfall deficit or excess of temperature changes
and other associated environmental variations. Wireless
sensor networks are gaining greater attention from the
research community and industrial professionals. LoRa is a
low power and enables very long range transmission (upto
15km) with low power consumption (in terms of mili-watt)
as compare to Wi-Fi module, mobile communication,
Bluetooth an other wireless techniques. In this work, the
technical differences of Wifi, LoRa, Bluetooth and mobile
communication are present and compared in terms of
physical or communication features. And finally explain
LoRa is the best technology in all wireless sensor network.
This system is based on a high-performance micro-
controller, integrated temperature and humidity sensor on
real-time data analysis. In our approach K-means
unsupervised clustering algorithm used for data mining and
IoT for real time data analysis. Also find the fault of the
particular sensor. The use of Data collection, pattern
classification and apply strategic analysis then control
execution for the final result.
Keywords: Agriculture, LoRa techniques, IoTs (Internet
of Things), Wireless Sensor Network, k-means
clustering algorithm.
I. INTRODUCTION
The rapid development of the Internet and the Internet of
Things (IoTs), various convenient service applications are
being used in various fields. In recent years, a new
agricultural information and communication technology,
called smart agriculture, which meets the needs of farmers
for information collection, signal processing, data analysis
and equipment control, has been developed. This work
proposes an agricultural service platform that is based on
a wireless sensor network and LoRa communication
technology. Work uses LoRa as a network transmission
interface to solve problem of communication failure and
save energy. An smart agriculture service platform is
developed to support environmental monitoring and to
improve the efficiency of agricultural management.
The goal of this work is to integrate IoTs awareness and
communication technology into an smart agriculture
platform. The accuracies of sensors of various types are
measured and these sensors are integrated into multi-
function sensor component. Then, multi-functional sensor
components are integrated with LoRa wireless network
components. In this work, an smart sensor network
platform for agricultural applications is designed and
constructed.
Environment Sensors - Natural changes in temperature,
humidity, Soil moisture, and direction may cause poor crop
growth or disease, reducing crop yields. Therefore, in this
work, an environmental sensor is used to reminds changes
in the physical atmosphere. The information is collected
and sent to an ESP32 controller platform for
preprocessing. ESP32 controller uses a LoRa
communication component to send data to a server for
analysis [6] Humidity and Temperature Sensor.- The
temperature sensor uses Thermally Sensitive Resistance
(TSR). TSR is very sensitive to temperature changes, and
obtained changes in resistance values can be converted
into temperature values. The humidity sensor is used to
define the quantity of water vapours in the atmosphere
[4,5]. The Soil moisture sensor measures the water level of
the environment. It helps for the growth of plants in
agriculture [3]. The light sensor measures the brightness
of the environment using a camera IC. It the light value
from a voltage and directly outputs a digital signal
[9].Temperature Control- To ensure the +accuracy of
sensed data, the sensor must be placed in a suitable
environment. The inaccuracy of measurement as a result
of long-term operation is eliminated by using a
temperature controlled fan to ensure that the sensor is
properly ventilated.
Internet of Things -
For the connection with remote location MQTT protocol is
used. MQTT is a machine-to-machine, Internet of Things
connectivity system development protocol. MQTT is a
telemetry protocol based on the publish-subscribe

communication model. MQTT is based on publisher, server
and clients. Likewise, the server is the guy who is
responsible for handling the client’s requests of receiving
or sending data between each other.
MQTT server is called a broker and the clients are simply
the connected devices. When a device (a client) wants to
send data to the broker, this operation is called a
“publisher”. When a device (a client) wants to receive data
from the broker, this operation is called a “subscriber” So
in this system, node MCU esp32 is the publisher which will
publish the acquired data of sensors GSM and GPS module.
It is further sent to the MQTT broker. And those whoever
are subscribed on that channel will get that data on their
device. So this is how caregiver will acquire obtained data.
II. LITERATURE SURVEY
Yi-Wei Ma1 and Jiann-Liang in 2018 which represents
proposed architecture of agriculture With the rapid
development of the Internet and the Internet of Things
(IoTs), various convenient service applications are being
used in various fields. In recent years, a new agricultural
information and communication technology (ICT), called
intelligent agriculture, which meets the needs of farmers
for information collection, signal processing, data analysis
and equipment control, has been developed. This work
proposes an intelligent agricultural service platform that is
based on a wireless sensor network and LoRa
communication technology [1]. A multi-sensor component
and an integrated communications network are
established. Wireless sensor networks and network
communication technology are used to support intelligent
agricultural data collection and equipment control [1].
Vaibhavraj S. Roham Kopargaon, Ganesh A. Pawar
Kopargaon, Abhijeet S. Patil Kopargaon, Prasad R. Rupnar
in 2015, that In farming Temperature, Humidity and CO2
are the most essential parameters. The growth of crops is
mainly depending on these three parameters. Currently
farmers don't have any system which will show real-time
levels of these parameters. Even farmer don't know when
humidity is increased or CO2 level increased in his green
house, because of it crop production gets affected. The
proposed system is going to monitor these changes
periodically and take an action automatically or pretend
the required action to the farmer. System will have a
provision to visualize the graphical representation of all
the streaming data from the green house. Later on farmer
can operate the devices from remote location by using its
smart phone.
The author Manikandan .S.V1, Jayapriya represented that
the Wireless Sensor Networks (WSNs) have concerned
much attention in recent years. In 2016, Deployment of
sensor networks still is a problem and subject of wide
range researches and developments. Prototype of WSN
built in framework of current research shows that small
networks are more or less functional while large scale
WSNs with long range nodes are issue. Despite of
problems our WSN prototypes gave possibility to gather
valuable data for field weather monitoring. The sensor
network technology will help the farmers to know the
exact values of the requirements that they need to improve
the crop productivity. It will help them in taking better
decisions at the right time. This will save their time and
labour also. The basic aim here is to transport the Indian
farmer from prediction to the exact values which are
beneficial for their farms[11].
Authors Aloÿs Augustin1, Jiazi Yi 1,*, Thomas Clausen 1
and William Mark Townsley 2 explained In this techniques
to LoRa is a long-range and low-power telecommunication
systems for the “Internet of Things”. The physical layer
uses the LoRa modulation, a proprietary technology with a
MAC protocol. LoRaWAN is an open standard with the
speciﬁcation available free of charge [23]. This paper gives
a comprehensive analysis of the LoRa modulation,
including the data rate, frame format, spreading factor,
receiver sensitivity, etc. A testbed has been built, to
experimentally study the network performance,
documented in this paper . The results show that LoRa
modulation, thanks to the chirp spread spectrum
modulation and high receiver sensitivity, offers good
resistance to interference. Field tests show that LoRa can
offer satisfactory network coverage up to 3 km in a
suburban area with dense residential dwellings. The
spreading factor has signiﬁcant impact on the network
coverage, as does the data rate. LoRa is thus well suited to
low-power, low-throughput and long-range networks. This
paper has also shown that LoRaWAN is an LPWAN
protocol very similar to ALOHA. Its performance thus
degrades quickly when the load on the link increases [13].
Author Orestis Georgiou and Usman Raza described that
Low Power Wide Area (LPWA) networks. We have
investigated the effects of interference in a single gateway
LoRa network, a LPWA technology with promising IoT

applications. Unlike other wireless networks, LoRa
employs an adaptive CSS modulation scheme, thus
extending the communication range in the absence of any
interference. Interference is however present when signals
simultaneously collide in time, frequency, and spreading
factor. Leveraging tools from stochastic geometry, we have
formulated and solved two link-outage conditions, one
based on SNR, and the other on co-spreading sequence
interference. Each displays interesting behaviours, unique
to LoRa, with the latter causing performance to decay
exponentially with the number of end-devices, despite
various interference mitigation measures available to
LoRa, thus limiting its scalability. It is interesting that LoRa
networks appear to be impervious to cumulative
interference effects (typically modelled as shot-noise [8]).
If this assumption is invalid, then our qualitative results
are simply optimistic upper bounds towards network
scalability. Going beyond this ﬁrst foray into the modelling
of LoRa, it would be interesting to understand the effets of
multiple gateways [6], and spatially inhomogeneous
deployments. Finally, we point towards recently developed
packet-level simulators [13] which can further shed light
into the performance of LoRa networks.
In 2016 designed by R. Fatchurrahman, H. Putra, I.
Joyokusumo, M. H. Habib, I. Imawati, S. P. Hadi,paper
presents a technique for selecting the light sensor
candidates according to its effectivity and sensitivity
criteria. the light sensor candidates, such as light
dependent resistor, photodiode, and photo transistor, is
taken due to its good performance in daylight sensing.
Analytic Hierarchy Process (AHP) which provides a
structured technique for organizing and analyzing complex
decisions based on mathematics was chosen. It offers
simple ways in matrix based-operations to rank the best
light sensor candidates over those criteria. In expecting the
best work of Wi-MoLS for controlling building illuminance
by adjusting the incoming natural light coming to the
building, the best light sensors must be selected over
expected criteria. The effectivity and sensitivity test had
been performed to evaluate the light sensors
characteristic. AHP is used to select the best light sensor
candidates. The result shows that light dependent resistor
is the most effective and sensitive light sensor by 62.16 %
in score instead of phototransistor (19.72%) and
photodiode (18.12%)[3].
III. METHODOLOGY-
Currently, various types of sensors are used to collect
agricultural information, such as the moisture of soil, light
intensity and temperature and humidity in air. However,
such sensors can only perform environmental monitoring
and data collection; they lack an integrated computing
platform, so obtained data cannot be effectively used or
analyzed. To solve this problem, an agricultural
intelligence platform is researched and developed here.
This work develops an smart agricultural platform that
integrates communication and multi-functional sensing
components. The intelligent agriculture platform can be
used to collect environmental information in agricultural
areas and transmit that data to remote computers through
the LoRa and in another approach used of IoTs i.e. sends
over the internet. for analysis, to make decisions
concerning equipment control. For example, ambient
temperature sensing can yield information about the
current temperature and changes in humidity in the field.
If the temperature is too high, a fan is turned ON and if the
moisture level is less then DC pump turns ON. The sensor
is placed outside of sensor platform so that sensor probe is
in direct contact with the environment for sensing. The
light sensor horizontally on the upper of a sensor platform.
The placement of air temperature, air humidity under
eaves. And soil moisture sensor deep at agricultural
platform. The wind speed and direction sensors outside.
Block Diagram:
Fig 1. Methodology in Smart Agriculture Platform
Reference [1]
Finally, LoRa is placed in the sensor house and
transmitting antenna are located on its roof, as Collected
data are transmitted to the LoRa base station; instantly
uploaded to a server, and presented on a web page for
analysis. This work is carried out by using k-means
unsupervised clustering algorithm for data analysis.
IV. RESULT AND DISCUSSION
Another approach is based to compare differences and
conditions between a single sensor and a calibration
sensor. A curve that reveals a linear correlation indicates
that the former sensor has a small deviation and is suitable
for use. In our dissertations work, several readings has
been taken for 'n' number of days. We had taken readings
of 28 days for analysis. To avoid complexity we consider

reading of five days are taken for plotting the accurate
graphs. Below figure shows that readings of five days at
different environmental conditions over surrounding.
Similarly the graph of humidity, soil moisture, and light
intensity etc.
EXPERIMENTAL SETUP –
(a) Setup of Smart Agriculture Platform
(b) LoRa Receiver
(c) Output on blynk app
It concludes, whether the sensors working or not. And it
contains all non-zero readings i.e., the temperature sensor
works continuously. According to expert readings the
equivalent reading of evaluated individual sensor, now
sensor is compared with centroid value and then compare
faulty and normal readings. Below figures shows graph of
measurement of sensors. The Y-axis show the different
values of temperature and X-axis show different readings
with respect to time in seconds at different days. The
different colors indicates that the different readings for
each day. Graph is non-linear according to time variation
due to temperature in day scenario. Similarly the graph of
humidity, soil moisture, and light intensity etc.
(a)
(a)Temperature Dependence Coefficient of Sensor.
(A)
(A) Graph of Temperature Measurement Sensor.
In this case the reading of sensors contains zero values so
there is some fault in that sensor. After that find the fault
of particular sensor and then gives the linear or Non-linear
graph between all individual sensor. The faulty sensor
indicates on X-axis as shown in graphs. And the figure
shows the Faulty temperature dependence coefficient
comparison.
In our dissertations work, several readings has been taken
for 'n' number of days. To avoid complexity we consider
reading of five days are taken for plotting the accurate

graphs. The five days analysis at different environmental
conditions over surrounding. If it contains zero readings
then it concludes that there is some fault in sensors.
Figures show that days has the faulty sensors. Similarly the
graph of humidity, soil moisture, and light intensity etc.
(1)
(1)Graph of Faulty Coefficient of Temperature Sensor.
The below figures represent the mean error of sensors
which indicates the smallest error. And it shows that the
mean error comparison of sensors. The results show that
day has smallest mean error in sensors. The Y-axis shows
the different values of sensor and X-axis shows different
readings with respect to different days. Similarly the graph
of humidity, soil moisture, and light intensity etc.
(2)
(2)Graph of Mean Error Graph of Temperature.
V. CONCLUSION AND FUTURE WORK
Wireless sensor networks have a huge potential for smart
monitoring systems. The ability to develop low cost and
simple end devices for quickly deploying in the field makes
WSNs very attractive for smart agricultural systems. There
is a gap in the literature about the engineering design and
application of such systems for a real-world application.
Moreover, most of the work focuses on LoRa-WAN as the
networking protocol of choice for deploying LoRa based
sensor networks, specifically for the applications.
Dissertation work focused on design, development, and
application of LoRa and IoT technology using low-cost
solution available for agriculture platform. The experiment
measures to determine sensor accuracy, select sensor and
design a long-range (up to 500m) low power (66 mW per
hour) communication platform. To reduce communication
failure and save energy to enhance the efficiency of
agricultural management by designing the LoRa network
transmission interface. Experimental work aids remote
monitoring of fields to farmers as well as assists to
increase in yield. In another approach use of IoTs for data
analysis of the sensor network is recommended with the
help of a smartphone. In our approach K-means
unsupervised clustering algorithm used for data analysis
and also we can find the fault of a particular sensor
network.
LoRa module has further increases range by improving
antenna design. The customize web-page design for real
time data analysis for the specific wireless sensor network
design. In another approach the number of parameter with
n number of sensors extends so get real time data with
high accuracy.
In future this work can be extended as large number of
data collected then gives more accurate and precise results
by the use of algorithm.
AKNOWLEDGEMENT
It gives me a better experience on bringing out the
dissertation report entitled ‘Smart Agriculture Solution
Using LoRa and IoT’. I express my sense of gratitude and
sincere regards to my dissertation guide R. R. Harne,
Assistant Professor, Department of Electronics
Engineering. I thank her for extending necessary help,
providing facilities and time to time valuable guidance in
deciding dissertation topic, related hardware and software
selection and thesis writing. I would also like to thank the
entire Electronics Engineering Department for being co-
operative with me during this whole session.
REFERENCES
1. Yi-Wei Ma1 and Jiann-Liang Chen, “Toward Intelligent
Agriculture Service Platform with LoRa-base Wireless
Sensor Network” IEEE International Conference in
2018.
2. P. P. Ray, M. Mukherjee and L. Shu, "Internet of Things
for Disaster Management State- one of the-Art and

Prospects," IEEE Access, vol. 5, pp. 18818-18835,
2017.
3. Christopher Brewster, Ioanna Roussaki, Nikos
Kalatzis, Kevin Doolin, and Keith Ellis "IoT in
Agriculture: Designing a Europe-Wide Large-Scale
Pilot" IEEE Communications Magazine September
2017.
4. Liwei Geng!!", Tingting Dong Hebei, Agricultural
University, baoding, China “An agricultural monitoring
system based on wireless sensor and depth learning
algorithm” iJOE journal Conference in 2017.
5. S. Persia, C. Carciofi and M. Faccioli, "NB-IoT and LoRA
connectivity analysis for M2M/IoTs smart grids
applications,“ Proceedings of the AEIT International
Annual Conference, pp, 2017.
6. O. Georgiou and U. Raza, "Low Power Wide Area
Network Analysis: Can LoRa Scale?," IEEE Wireless
Communications Letters, 2017.
7. J. Haxhibeqiri, A. Karaagac, F. Van den Abeele, W.
Joseph, I. Moerman and J. on Hoebeke,"LoRa indoor
coverage and performance in an industrial
environment: Case study,"Proceedings of the IEEE
International Conference Emerging Technologies and
Factory Automation, 2017.
8. Gaurav Jadhav, Kunal Jadhav, Kavita Nadlamani,
“Environment Monitoring System using Raspberry-Pi”
International Research Journal of Engineering and
Technology (IRJET),2016.
9. Manikandan .S.V1, Jayapriya .P2, “Precision
Agriculture Using Wireless Sensor Network System:
Opportunities and Challenges” International Journal Of
Engineering And Computer Science ISSN: 2319-7242,
2016.
10. R. Fatchurrahman, H. Putra, I. Joyokusumo, M. H.
Habib, I. Imawati and S. P. Hadi, "Light sensor selection
of Wi-MoLS (wireless modern light sensor) based on
analytic hierarchy process (AHP)," Proceedings of the
International Conference of Industrial, Mechanical,
Electrical, and Chemical Engineering, pp. 152-155,
2016.
11. Aloÿs Augustin1, Jiazi Yi 1, Thomas Clausen 1 and
William Mark Townsley, “A Study of LoRa: Long Range
& Low Power Networks for the Internet of Things” in
2016.
12. Murat Denera, “Smart technology with Wireless
Sensor Networks” World Conference on Technology,
Innovation and Entrepreneurship, 2015.
13. Vaibhavraj S. Roham, Ganesh A. Pawar Student at
Sanjivani College of Engineering, “Smart Farm using
Wireless Sensor Network”, International Journal of
Computer Applications (0975 – 8887) National
Conference on Advances in Computing (NCAC) 2015.
14. Zhuo-fu Liu, Lin Wang, Guang-sheng Xi, Zhongming
Luo, Yongbo Li , “Temperature and Humidity Sensor
Location Optimization Based on BP Neural Network”
Proceedings of the International Conference on
Instrumentation and Measurement, Computer,
Communication and Control, pp. 283-287, 2015.
15. Z. Liu, Z. Luo, T. Su, C. Liu and W. Jiang, "Problem-
Oriented Pedagogy Using Temperature and Humidity
Sensor Network Optimization, "Proceedings of the
International Conference on Instrumentation and
Measurement, Computer, Communication and Control,
pp. 652-657 2014.

IRJET- Smart Agriculture Solution using Lora and IoT

More Related Content

What's hot (20)

Similar to IRJET- Smart Agriculture Solution using Lora and IoT (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- Smart Agriculture Solution using Lora and IoT