Proceedings of the International Conference on Electronics and Sustainable Communication Systems (ICESC 2020)

IEEE Xplore Part Number: CFP20V66-ART; ISBN: 978-1-7281-4108-4

IoT based Smart System for Enhanced

Irrigation in Agriculture
Bhanu K.N. 1, M ahadevaswamy H.S.2, Jasmine H.J.3
Department of Computer Science, Amrita School of Arts & Sciences, M ysuru
Amrita Vishwa Vidyapeetham, India
E-mail: [email protected] 1, [email protected] 2, [email protected] 3

Abstract — Internet of Things (IoT) is an interconnection Implementing IoT in agriculture utilizes sensors and
of devices that can transfer information over the internet and microcontrollers to make use of the system efficiently,
to control operations without human interference. Agriculture
provides a rich source of parameters for data analysis which where sensors have the potential to get a large amount of
helps in better yielding of crops. The usage of IoT devices in field information. By incorporating a distinctive sensor’s
agriculture helps in the modernizing of information and data it is possible to decide on suitability and fertility of the
communication in smart farming. The key parameters that can
soil. IoT helps in providing a better yield of the crop which
be considered for better growth of crops are soil types, soil
moisture), mineral nutrients, temperature, light, oxygen and so improves productivity.
on. Various sensors have been used to sense these parameters To monitor the crop in the agriculture field with
and communicate the same to the cloud. This paper considers a improved productivity it is necessary to know the condition
few of these parameters for data analysis that helps in
proposing the users to take better agricultural decisions using of the soil in the field. IoT helps in monitoring the field from
IoT. The proposed system performed better and is any geographic location whenever necessary. Moisture,
implemented atThingS peak IoT cloud platform. temperature, weather, and soil PH are considered for
Keywords— Internet of Things (IoT), Aurdino Uno, monitoring soil conditions. Global System for Mobile
Wireless Sensor Networks (WSN), Wi-Fi, Cloud Computing. [GSM] technology is used to send the condition of the farm
to the user as discussed in [1]. Automation in agriculture
I. INTRODUCTION helps to adopt new technology by replacing old traditional
IoT is the interconnection of computing gadgets that technology with less human interaction which helps in
communicates the data to the cloud. Data such obtained by increasing the production rate. In [2] the authors have used
IoT is processed to perform necessary actions. The main Arduino Uno with a moisture sensor to measure soil
occupation in developing countries like India is agriculture. moisture; data is updated to the web page using GSM-GPRS
Around 70% of the Indian population is reliant [directly or SIM900A, using which farmers can check the sprinkler
indirectly] upon agriculture. Agriculture contributes 15 to status. It concentrates on maintaining and monitoring the
16% to the country’s GDP [Gross Domestic Product]. soil moisture with an automated watering system. The
Agriculture is affected by numerous temporary changes such system considers three different parameters for better
as climate, soil erosion by flood, growing the same crop productivity in crops. Due to global warming, the water
again and again without scientific farming, etc. Adapting level is reducing so it is difficult in finding the variations in
smart farming gives results like less wastage of water, an soil moisture and fertility where different sensors are used
increase in the quality of crops and reduced usage of with Arduino Uno to measure temperature, humidity, and
resources. This helps to get a good yield of the crop at a moisture content and displayed on the LCD display is
faster rate compared to previous farming methodology. explained in [3]. The work proposed in [4], for the use of
Nowadays agriculture requires various parameters, such as Arduino Uno in the irrigation system by considering water
the crop type suitable for the present soil condition, type of flow sensors, moisture and temperature sensor. Sensed data
irrigation, fertilizer requirement, etc. The soil monitoring is sent to the website. Users can control the pump from a far
system for agriculture is a new system that uses parameters place using the website. It helps to maximize the crop with
such as moisture, temperature, humidity, ground less interaction of human errors. As explained in [5], water
temperature, soil PH, etc. is the primary factor of irrigation it is distributed only when
it is necessary which in turn helps to avoid water wastage

978-1-7281-4108-4/20/$31.00 ©2020 IEEE 760

Authorized licensed use limited to: University of New South Wales. Downloaded on August 08,2020 at 12:37:43 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the International Conference on Electronics and Sustainable Communication Systems (ICESC 2020)
IEEE Xplore Part Number: CFP20V66-ART; ISBN: 978-1-7281-4108-4

and soil degradation. Farmers need to know about the probes. Soil conducts more electricity (less resistance) when
parameters such as temperature, moisture, humidity and the presence of moisture content is more in the soil. If soil is
water consumption. Arduino Uno and Sensors are used with dry then it conducts less electricity (more resistance). The
the Wi-Fi module to send the data into the cloud. output of the sensor relies upon electric resistance of dry or
The objectives of smart irrigation system for IoT are: wet soil.
(i) To communicate various parameters of soil sensed
B. DHT11 Sensor:
through the sensors to the cloud using Arduino; (ii) To
apply machine learning to the live data received from the
sensors to classify based on the threshold values; (iii)
Based on the classification take action to send e-mail to
user; (iv) Send e-mail to the user to take appropriate
decision.

This paper is organized as follows: IoT devices for

Smart Agriculture are discussed in section II. Section III
explains the IoT Based Smart agriculture system. Section IV Fig. 3: DHT 11 T emperature and Humidity Sensor

explains the proposed methodology and algorithm. Section Fig.3 shows the DHT11 sensor, utilized to gauge the
V gives the results and discussion and Section VI concludes humidity and temperature in the environment. It measures
the paper. the temperature and water vapors present in the surrounding
air which is in gaseous state.
II. IOT DEVICES FOR SMART AGRICULT URE
C. DS18B20 Sensor:

Fig. 1: Arduino Uno

Fig.1 shows the Arduino Uno microcontroller

Fig. 4: DS18B20 Waterproof T emperature Sensor
(Atmega328). Open-source Arduino Software is utilized to
execute, instructions provided by the user to get the readings
Fig.4 shows the DS18B20 sensor, utilized to measure
from the sensors. It has 0 to 13 digital input-output pins and
the temperature in the soil. It is encapsulated with stainless
analog input-output pins A0 to A5 with the operating
steel tube to measures the ground temperature. The
voltage being 7 to 20V. It is a microcontroller board with
temperature of the soil varies before and after irrigation.
inputs and outputs and it needs to be connected to the
computer using USB cable or battery to get started.

A. YL-38 Sensor: III. SMART SYSTEM FOR IRRIGATION USING IOT

IoT based smart agriculture system contains sensors,
microcontroller as hardware gadgets and Arduino as
software. Where combinations of both hardware and
software devices are used to determine the suitability of the
soil for agriculture purposes. Data that is sensed by the
sensors are sent to the cloud where live data is taken from
the sensor. The microcontroller is connected together with
sensors is showed in Fig. 5.
Fig. 2: YL 38Soil Moisture Sensor

Fig.2 shows the YL 38 sensor, is used to calculate the

percentage of water content present in the soil. It has two

978-1-7281-4108-4/20/$31.00 ©2020 IEEE 761

Authorized licensed use limited to: University of New South Wales. Downloaded on August 08,2020 at 12:37:43 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the International Conference on Electronics and Sustainable Communication Systems (ICESC 2020)
IEEE Xplore Part Number: CFP20V66-ART; ISBN: 978-1-7281-4108-4

Fig.6 circuit diagram shows the Aurdino Uno and

various sensors are used to gauge the soil data. Tools used
and their functionalities are given below:
 Arduino Uno: Microcontroller
 YL-38 Sensor: It is utilized to gauge the soil water
content.
 DHT11 Sensor: It is utilized to gauge the
temperature and humidity
 DS18B20 Sensor: Waterproof sensor utilized to
gauge the temperature of the soil
Wi-Fi module is utilized for transmitting the data to the
Fig. 5: System Architecture
cloud for further analysis.

 Input data are taken by the sens or from the external

environment.
 Sensors are controlled by arduino microcontroller.
 Every data is sensed discretely by sensors and
stores same to the cloud for further analysis.
 The threshold value of soil parameters is
considered to predict suitability.
 Message is sent to the user to take further actions
based on prediction.

T able 1: T hreshold values

Soil parameters Threshold value
Fig. 7: Experimental Setup
Moisture 30-80%
Temperature 18-40°C Fig.7 shows the experimental setup and various soil
(atmosphere) parameters that are sensed using a microcontroller and
Humidity 45-70% different sensors.
Soil temperature 20-40°C IV. PROPOSED MET HODOLOGY AND ALGORIT HM
IoT helps in the identification of soil suitability in
The parameter for the soil varies from place to
agriculture. Plants are dependent on environmental factors
place and time to time. Few parameters are considered in
that frequently change from time to time and place to place.
this work to predict the soil status. Table1 shows the various
Heterogeneous IoT devices such as sensors and Arduino
threshold values for different parameters of the soil.
Uno help to identify the parameters of soil. The proposed
LCD 16*2 Display DS18B20 Sensor DHT11Sensor scheme works as follows: 1) Every sensor senses the related
soil parameters and sends it to the Arduino. 2) The sensed
information is sent to ThingSpeak Cloud through arduino. 3)
The live data received from sensors are stored in cloud. 4)
Stored data is classified and analyzed using classification
algorithm at ThingSpeak cloud.5) The data such classified
which fall beyond threshold values are analyzed and an
email is sent to the user to perform necessary actions.

A. Proposed Algorithm

Nomenclature:
n: Number of data, Sm: Soil Moisture,
Sgt : Soil ground Temperature, M th : Soil Moisture threshold,
Arduino Uno Wi-Fi module YL-38 Sensor GTth : Ground Temperature threshold, DS: Data Store.
Fig. 6: Circuit Design

978-1-7281-4108-4/20/$31.00 ©2020 IEEE 762

Authorized licensed use limited to: University of New South Wales. Downloaded on August 08,2020 at 12:37:43 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the International Conference on Electronics and Sustainable Communication Systems (ICESC 2020)
IEEE Xplore Part Number: CFP20V66-ART; ISBN: 978-1-7281-4108-4

The data obtained from YL-38 Sensor, DHT11 Sensor,

Algorithm for Smart Irrigation System in Agriculture and DS18B20 Sensor help to take the necessary decision for
the farming land. Table.2 gives the sample data gauged by
1: Begin the different sensors.
2: Sensors senses data Sm and Sgt
3: Node MCU sends the sensed data to ThingSpeak
cloud a) Classification using KNN
4: Data processing in ThingSpeak
5: for i=1 to n
6: {
7: if (Sm >= M th ) && (Sgt >=GTth )
8: Classified as HIGH;
9: else
10: Classified as LOW;
11: end if
12 :}
13: if (Classification = HIGH)
14: Send e-mail to switch on motor
15: else
16: Store values in DS
17: End

Fig. 8: Accuracy of KNN Classifier

Data’s sensed by the sensors is sent to ThingSpeak
IoT platform cloud. Where classification algorithm is The above Fig.8 shows the accuracy of KNN classification
preferred on the data obtained discreetly and classified algorithm as 70.8% for the data set obtained by the sensors.
based on threshold value. All values falling above threshold
b) Classification using Naïve Bayes algorithm
value should be considered for action and for values falling
below threshold value can be stored in data store for further
analysis is explained in the above algorithm.

V. RESULT S AND DISCUSSION

Sensors discreetly send the data for analysis . It indicates
the variations in soil temperature, soil moisture content,
humidity and temperature of surroundings.

T able 2: Sample data received by sensors

Temp Humidit Soil temperature

Moist
eratur y (°C)
ure
e (%)
(%)
( °C) Fig. 9: Accuracy of Naïve Bayes Classifier
49 26 78 25.63
74 26 78 25.5 Fig. 9 shows the accuracy of Naïve Bayes classifier as
76.4% for the data set obtained from the sensors in the
76 28 75 29.4 proposed system.
72 24 82 23.45
c) Classification using SVM Algorithm
69 29 63 27.82
68 32 55 28.32
68 29 60 28.1

70 26 68 25.2

978-1-7281-4108-4/20/$31.00 ©2020 IEEE 763

Authorized licensed use limited to: University of New South Wales. Downloaded on August 08,2020 at 12:37:43 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the International Conference on Electronics and Sustainable Communication Systems (ICESC 2020)
IEEE Xplore Part Number: CFP20V66-ART; ISBN: 978-1-7281-4108-4

Fig. 10: Classification using SVM Fig. 13: Smart System for Irrigation in T hingSpeak Cloud

Fig.10 signifies the representation of moisture with Fig. 13 shows the Smart Irrigation System in ThingSpeak
ground temperature data. White dots indicate the values’ cloud where the information for the action to be performed
that are within the threshold value. And green dots represent regarding the data beyond threshold value is being sent
the values above the threshold value. For all data falling through e-mail.
above threshold value decision is taken and an email is sent
to the user.

Fig. 14: e-mail to user from Smart System for Irrigation

Fig. 14 shows the e-mail being received by the user for the
Fig. 11: Accuracy of SVM Classifier values classified above the threshold.

Fig. 11 shows the accuracy of Support Vector Machine

VI. CONCLUSION
(SVM) classifier as 87.5% for the data set obtained from the
sensors in the proposed system. The proposed work provides the information on various soil
parameters that includes soil temperature, soil moisture and
d) Comparison of classification Algorithms
atmospheric temperature to predict irrigation suitability.
This system helps to analyze the soil parameters thereby
ensuring a better system of irrigation for agriculture. The
data collected from the sensors are made to learn using
machine learning techniques to ensure a fully automated
system. Implementing an IoT based smart agriculture
system helps in obtaining quality crops and it also reduces
the human involvement in agricultural activities. As seen in
results the Support Vector Machine (SVM) classification
algorithm suits better for the data set obtained from the
Fig. 12: Comparison graph
proposed method by providing 87.5 percent accuracy and
Fig.12 shows the comparison of KNN, Naïve Bayes, and performed better than KNN Classifier and Naive Bayes
SVM classifier for accuracy. The above figure is clearly Classifier. In the present work correlation of Soil
understood that SVM performs better for the proposed temperature and soil moisture is considered for
system by giving 87.5% accuracy whereas KNN is giving classification. In future work classification based on the
70.8% and Naïve Bayes provides 76.4% accuracy for the correlation of all the four parameters can be considered; also
same data set obtained from the sensors. the system can be enhanced in future to predict the fertility
of the soil by considering soil fertility parameters.

978-1-7281-4108-4/20/$31.00 ©2020 IEEE 764

Authorized licensed use limited to: University of New South Wales. Downloaded on August 08,2020 at 12:37:43 UTC from IEEE Xplore. Restrictions apply.
 Proceedings of the International Conference on Electronics and Sustainable Communication Systems (ICESC 2020)
IEEE Xplore Part Number: CFP20V66-ART; ISBN: 978-1-7281-4108-4

Humanitarian Technology Conference (GHT C 2014). IEEE,

REFERENCES
2014.
[18] Mampentzidou, Ioanna, Eirini Karapistoli, and Anastasios A.
[1] Balaji, G. Naveen, et al. "Iot based smart crop monitoring in
Economides. "Basic guidelines for deploying wireless sensor
farm land." Imperial Journal of Interdisciplinary Research (IJIR)
networks in agriculture." 2012 IV International Congress on
Vol 4 (2018): 88-92.Rawal, S. (2017). IOT based smart
Ultra Modern T elecommunications and Control Systems. IEEE,
irrigation system. International Journal of Computer
2012.
Applications, 159(8), 7- 11.
[2] Bhadani, Prahlad, and Vasudha Vashisht. "Soil Moisture,
[19] Goap, Amarendra, et al. "An IoT based smart irrigation
management system using Machine learning and open source
T emperature and Humidity Measurement Using Arduino." 2019
technologies." Computers and electronics in agriculture 155
9th International Conference on Cloud Computing, Data Science
(2018): 41-49.
& Engineering (Confluence). IEEE, 2019.
[20] Das, Rahul, V. Pooja, and V. Kanchana. "Detection of diseases
[3] Lokhande, Kalyani, et al. "IOT based Automatic Farm
on visible part of plant—A review." 2017 IEEE T echnological
Monitoring." (2019).
Innovations in ICT for Agriculture and Rural Development
[4] Guchhait, Prasun, Pranav Sehgal, and Vidyadhar J. Aski.
(T IAR). IEEE, 2017.
"Sensoponics: IoT-Enabled Automated Smart Irrigation and Soil
[21] Machaiah, M. D., and S. Akshay. "IOT Based Human Search
Composition Monitoring System." Information and
and Rescue Robot using Swarm Robotics."
Communication T echnology for Sustainable Development .
[22] Cp, Prathibhamol, et al. "ICACCI-02 (A): Artificial Intelligence
Springer, Singapore, 2020. 93-101.
and Machine Learning/Data Engineering/Biocomputing
[5] Shah, Kritika, et al. "Proposed Automated Plant Watering
(Regular Papers).
System Using IoT ." Available at SSRN 3360353 (2019).
[23] Pundir, Yogita, Nancy Sharma, and Yaduvir Singh. "Internet of
[6] More, Abhishek, et al. "Soil Analysis Using Iot." Available at
things (IoT): Challenges and future directions." International
SSRN 3366756 (2019).
Journal of Advanced Research in Computer and Communication
[7] Rajalakshmi, P., and S. Devi Mahalakshmi. "IOT based crop -
Engineering 5.3 (2016): 960-964.
field monitoring and irrigation automation." 2016 10th
[24] Muangprathub, Jirapond, et al. "IoT and agriculture data
International Conference on Intelligent Systems and Control
analysis for smart farm." Computers and electronics in
(ISCO). IEEE, 2016.
agriculture 156 (2019): 467-474.
[8] Bhanu, K. N., T . Bhaskara Reddy, and M. Hanumanthappa.
"Multi-agent based context aware information gathering for
agriculture using wireless multimedia sensor
networks." Egyptian Informatics Journal 20.1 (2019): 33-44.
[9] Gondchawar, Nikesh, and R. S. Kawitkar. "IoT based smart
agriculture." International Journal of advanced research in
Computer and Communication Engineering 5.6 (2016): 838-
842.
[10] Ramesh, Maneesha V. "Real-time wireless sensor network for
landslide detection." 2009 T hird International Conference on
Sensor T echnologies and Applications. IEEE, 2009.
[11] Jaishetty, Shruti A., and Rekha Patil. "IoT sensor network based
approach for agricultural field monitoring and control." IJRET :
International Journal of Research in Engineering and
T echnology 5.06 (2016).
[12] Gayatri, M. K., J. Jayasakthi, and GS Anandha Mala. "Providing
Smart Agricultural solutions to farmers for better yielding using
IoT ." 2015 IEEE T echnological Innovation in ICT for
Agriculture and Rural Development (T IAR). IEEE, 2015.
[13] Patil, V. C., et al. "Internet of things (Iot) and cloud computing
for agriculture: An overview." Proceedings of agro-informatics
and precision agriculture (AIPA 2012), India (2012): 292-296.
[14] Ray, Partha Pratim. "Internet of things for smart agriculture:
T echnologies, practices and future direction." Journal of
Ambient Intelligence and Smart Environments 9.4 (2017): 395-
420.
[15] Mekala, Mahammad Shareef, and P. Viswanathan. "A Survey:
Smart agriculture IoT with cloud computing." 2017 international
conference on microelectronic devices, circuits and sy stems
(ICMDCS). IEEE, 2017.
[16] Nikolidakis, Stefanos A., et al. "Energy efficient automated
control of irrigation in agriculture by using wireless sensor
networks." Computers and Electronics in Agriculture 113
(2015): 154-163.
[17] Kumar, Akash, et al. "Smart irrigation using low-cost moisture
sensors and XBee-based communication." IEEE Global

978-1-7281-4108-4/20/$31.00 ©2020 IEEE 765

Authorized licensed use limited to: University of New South Wales. Downloaded on August 08,2020 at 12:37:43 UTC from IEEE Xplore. Restrictions apply.