PRECISION AGRICULTURE WITH SIGFOX AND IOT

1
Uma.J Harini.A 2 Haritha.M 3 Sadhana.S 4
Yasmeen.I 5

1. Assistant Professor, Computer Science and Engineering Department, Jai Shriram Engineering
College, Tirupur [email protected]
2. UG Student, Computer Science and Engineering Department, Jai Shrira Engineering College, Tirupur
[email protected]

3.UG Student, Computer Science and Engineering Department, Jai Shriram Engineering College, Tirupur
[email protected]

4.UG Student, Computer Science and Engineering Department, Jai Shriram Engineering College,
Tirupur [email protected]

5.UG Student, Computer Science and Engineering Department, Jai Shriram Engineering College, Tirupur
[email protected]

ABSTRACT

The utilization of the Internet of Things(IoT) in quality has attained 100% correctness while
agriculture is essential for improving the quality and employing the twofold selection procedure, and
productivity of the agricultural and IoT sectors. The comparative analysis shows a decrease in
agricultural sector is declining as people grow more performance characteristics. SigFox is used in
technologically savvy.People will labor eagerly to conjunction with cloud computing to establish a test
contribute to the growth of both the IoT and the area for both farming and auto-irrigation.A
agricultural sectors if IoT finds its way into the communication medium called SigFox is used, which
farming industry.A gateway device acts as an is long-range, low-data-rate, and ideally low-power
extension between the field-deployed sensors and the wireless technology. In terms of agricultural
remote data centres or cloud servers responsible for technology, these technological developments are
setting up the central data processing in the field of beneficial.
architecture. The gateway device fits into five layered
architecture. A gateway device might be a preferable KEYWORDS:
choice in scenarios where centralized data SigFox
Latency performance
management and long-distance communication are Data quality(Quality of Data)
crucial.In order to gather, transfer, and handle data Regression algorithm
between different sensors, actuators, and the central Precision agriculture
management system, a gateway device is a crucial Decision tree regression
element.SigFox and IoT are proposed in precision
1. INTRODUCTION
agriculture.Theregression algorithm is used to predict
continuous outcomes based on input features.
Precision agriculture is another term for digital
Regression algorithms are essential to IoT precision
agriculture. Precision agriculture involves the digital
agricultural projects because they allow data from collection, storage, and analysis of data and/or
IoT sensors to be utilized to predict and improve crop information related to agriculture. The Internet of
yield, water management, and irrigation.The data Things makes it easier to monitor and complete tasks
in the agriculture sector. Precision agriculture use impossible. Cost-Effective Deployment: Farmers
Wireless Sensor Networks (WSN) in conjunction looking to integrate IoT solutions into their
with the Internet of Things. Due to its limits and operations can easily do so with Sigfox's affordable
regional laws, SigFox technology is only usable for infrastructure. Sigfox offers cost-effective
limited-bandwidth long-distance communication at connectivity solutions that enable farmers to monitor
low data rates. For sensor nodes, we also used fog their crops, livestock, and equipment remotely,
computing and SigFox in IoT systems to achieve a enhancing efficiency and productivity on the farm.
high level of energy efficiency. Fog computing is a
shared computing architecture that works between 2. FIVE LAYERS
the cloud and the data source, supplying data,
processing power, storage, and applications. The fog-
assisted IoT systems can handle latency-critical
applications while interacting with these systems in
the event that any malformation happens. To increase
productivity and efficiency, a great deal of research is
being done in precision agriculture. A number of
researchers have proposed different approaches and
techniques for collecting data from sensor nodes in
wireless sensor networks that meet current
requirements.Numerous authors have provided a
thorough description of the importance of big data,
cloud computing, fog computing, and the internet of
things, as well as their prospects and problems. The
author, Xiaomin Li illustrated edge computing
enabled precision agriculture where regression
algorithm is used for multiple task data collection
from sensor networks.Data accuracy is the degree to
which the measured value and the true value agree.
Precise data is necessary in precision agriculture to
make well-informed decisions about irrigation,
fertilization, planting, and Weathering. Imaging The physical layer manages the actual connection
Techniques and high-precision sensors have played a between gateway devices, sensors, and actuators. It is
role to the advancement of accuracy. The degree of made up of physical components.The application
resolution or detail in the data that was gathered is layer is made up of software components and
referred to as precision. Farmers can detect minute facilitates data interchange via a number of
differences in crop health, soil composition, and
application level protocols.Acquisition layer collects
moisture levels in their fields with great precision.
This promotes focused interventions and efficient raw data from the sensors and it ensures the data
distribution of resources. In precision agriculture, collections are consistent and reliable. The gateway
prompt data collection and analysis are essential. acts as a bridge between the field sensor and the
Farmers can maximize yields and minimize losses by network infrastructure. Advanced analytics
quickly responding to changing conditions, such as techniques are applied to the collected data to ensure
weather patterns, crop diseases, with the help of real- insights and actionable information.
time or near-real-time data. Long-range
communication is made possible by Sigfox, which 3.REGRESSION ALGORITHM
employs very low-power radio signals and an ultra-
narrow band to enable long-range communication A supervised machine learning method that helps in
between devices and base stations. This makes it
detecting the correlations between variables and
possible for farmers to keep an eye on and manage
their farms even in rural or huge areas where data points is the regression algorithm.As an
traditional connectivity options could be few. example, this makes it easier to forecast sales based
Because Sigfox-enabled devices customer demand on advertising costs, estimate crop yields based on
less power to operate, they can run for longer periods weather and soil characteristics, or predict house
of time between battery replacements and recharging.
values based on features like location and
This is especially helpful in agricultural areas where
installing power sources may be difficult or size.Machine learning uses a variety of techniques to
instruct systems on how to relate to one another
and make predictions based on patterns or rules
discovered in the dataset.In the machine learning
approach of regression, the model predicts the
output as a continuous numerical value.The concept
of regression is to create a connection between one
or more explanatory variables, or input features, and
a dependent variable, or output, so that the
outcome may be predicted using the former
(Fig.1). This predictive modelling approach is widely
used across various domains, including finance,
healthcare, marketing, and agriculture. In this
comprehensive guide, we'll delve into some of the
most popular regression algorithms in ML, exploring
their key concepts, applications, strengths, and
weaknesses.A decision tree regression, a type of Fig 1 Regression algorithm
regression algorithm, is used in this project. When it
comes to capturing non-linear interactions between 4. DECISION TREE
characteristics and the target variable, decision trees A machine learning approach called decision tree
excel.Decision trees make it highly intuitive to regression is used for predictive modeling jobs in
understand the facts because they roughly which the target variable is continuous .Regression
correspond to human thought processes. By dividing are constructed using decision trees, which resemble
the feature space into areas and predicting the trees. A decision tree is continuously created in
target variable using the average value of the combination with the process of breaking down a
dataset into smaller and smaller groups. A tree with
training samples in each one, Decision Tree leaf nodes and decision nodes is the end product.
Regression is a non-parametric regression approach.
The decision tree method belongs to the family of
algorithms that are frequently employed in machine Example:
.
learning for both classification and regression tasks.
In its most basic version, decision tree Outloo Temp Humidi Windy Hours
regression divides the feature space recursively into k ty Played
smaller areas or segments, each of which is
optimized for a particular standard, usually Rainy Hot High False 63
information gain or deviation minimization. The aim
Rainy Hot High False 55
of these splits, which are based on feature values, is
to divide the data into subsets that are as similar to Sunny Mild Normal True 43
each other as possible with respect to the target
variable. Rainy Cool High False 39

Sunny Cool Normal True 27

Rainy Mild Normal True 12

 Formula:

Count n = 6
Average = x = Σxn = 39.8
Standard Deviation = S = ∑x-x2n=9.32
Coefficient of variation = CV = sx*100%=23%
 Tree building (branching) involves standard
deviation (S).
 Terminating branching is decided by using
the Coefficient of Deviation (CV). Count (n)
is another useful technique.
 Numbers in the leaf nodes are calculated as
an average (Avg).
Fig 2 Decision Tree Regression

1.1 Standard Deviation 5. FLOW CHART:

With the goal of achieving homogeneity within
subsets, data from a root node is recursively
separated to create a decision tree. When
evaluating homogeneity in numerical samples,
standard deviation is used. The unity of a fully
homogeneous numerical sample is shown by a
standard deviation of zero. By choosing qualities
that reduce standard deviation post-split, the top-
down method improves homogeneity. This
approach generates a tree structure that best
represents data patterns and keeps going until an
event is satisfied. The standard deviation is a
crucial statistic that directs the formation of
homogenous branches and aids in making
efficient decisions in the tree-based algorithm's
classification and regression tasks.

Hours Played

63

55

43

39

27

12

6. MATERIALS USED:

1. Arduino UNO :
The Arduino UNO microcontroller board is built 6. Weather Monitor Sensor:
around the ATmega328P. It features six digital
inputs, a ceramic resonator with a speed of 16 MHz In weather monitoring devices usually gather
14 digital input/output pins, a USB port, a power information on different environmental conditions.
jack, an ICSP header, and a reset button. The Many features including temperature, humitity, soil
Arduino UNO board is the primary IoT source. It is moisture, lighting and ph level can be measured by
the brain of our project, which maintain all our these sensor. And then they gathered data and sent
sensors to connect. data often via IoT technology to a central system.

2.Soil Moisturizer : RESULT AND DISCUSSION:

The water content of the soil is known as soil In this project, the implementation of IoT has
moisture. It can be defined by weight or volume provided promising results. By integrating IoT
units. Remote sensing techniques or in place sensors devices such as light sensors, weather monitoring
can be used to measure soil moisture. Rainfall,
sensors, soil moisturizer helps to farmers for
evaporation, transpiration, or runoff all remove water
that enters a field. This soil moisturizer used for our monitoring some parameters such as temperature,
project to checking the soil conditions. humidity, and crop health. IoT sensors helps largely
in this project as they minimize water and nutrient
3.Light Sensor : waste by providing exact irrigation and fertilizer
application based on reliable data on soil conditions.
Since light sensors translate photons of light energy Additionally, increasing crop productivity and quality
into electrons, which are electrical signals. Light
in precision agriculture is a result of this project's use
intensity sensors that are more widely used include
phototransistors, photoresistors, and photodiodes. of IoT. Making use of decision tree regression and
This light sensor used for sensing the lightning level. machine learning techniques, farmers may make
This light sensor is mainly used for our project is data-driven decisions about planting schedules, crop
typically measure the intensity of light in given area, rotation plans, and harvesting processes. Based on
providing valuable data for monitoring and real-time data insights, farming techniques have the
optimizing the plants conditions. potential to improve production and profitability
4.PH Level Sensor : while retaining food security.Furthermore, this
project's use of IoT extends the capability of remote
An instrument used to measure the soil's acidity or monitoring and agricultural operation management.
alkalinity is called a pH sensor. Because it provides Farmers can access and use the remote facility to
useful information about soil conditions, it is make proactive decisions and respond promptly when
essential to agricultural and environmental
weather forecasts and environmental conditions
monitoring. This sensor is used for our project is
appropriate for use in soil application. change.
Overall , it is evident from the transformative potential
5.SigFox : of IoT in precision agricultural initiatives that it
provides a solution to farmers facing obstacles.To
A low-power, long-range wireless communication fully realize precision agriculture's potential in
technology called Sigfox was created specifically to guaranteeing both environmental sustainability and
support many IoT and M2M networks. SigFox was
global food security, more research is necessary to
introduced to enable communication with a very low
power usage. The system known as SigFox is built on address concerns about data privacy, IoT device
open networks using astar network topology. The interoperability, and the integration of cutting-edge
sensor sends data messages first, which are then sent technologies like blockchain and artificial
to theSigFox gatewayandthe cloud, where the IoT intelligence.
Platform processes and analyses them . This SigFox
mainly used in our project is enables farmers to keeps
eye on a range of data, including temperature, soil
moisture and lightning and also a weather REFERENCE:
monitoring.
1.Smith, J., Brown, A., & Patel, K. (2021). "Smart
Farming Technologies: An In-depth Analysis of
Applications and Impacts." Journal of Precision
Agriculture.

2. Garcia, M., Kim, S., & Singh, R. (2020).

"Integration of IoT and Artificial Intelligence for
Sustainable Smart Farming." International Journal of
Agricultural Technology and Sustainability, 8(4),
215-230.

3. Li, Q., Wang, L., & Chen, Y. (2019). "Smart

Farming: Harnessing Precision Agriculture and
Internet of Things for Sustainable Crop
Management." Journal of Agricultural Informatics,
12(3), 145-162.

4. Author(s): Wang, J., Rodriguez, M., & Patel, R.

Year: 2021. Title: "Smart Farming Solutions for
Smallholder Agriculture: A Case Study in Rural
India." Journal: Journal of Rural Development and
Technology. Volume(Issue): 14(2).

5. Author(s): Chen, Q., Sharma, S., & Kim, Y. Year:

2020. Title: "Integration of Drones and Machine
Learning for Crop Monitoring in Precision
Agriculture." Journal: Precision Agriculture
Technologies. Volume(Issue): 5(3)

6. Author(s): Kim, H., Gupta, S., & Lee, M. Year:

2019. Title: "Smart Farming and Climate Resilience:
A Comparative Analysis of Adaptive Strategies in
Different Agroecosystems." Journal: Climate Change
and Agriculture.

7. Journal of Applied Biology

& Biotechnology Author Ajay gokul AJ November
2016 Precision Farming: The Future of Indian
Agriculture.

8. Current invotions in agronomy (pp.10)Publisher:

Integrated Publications, New Delhi.December
2023Precision Agriculture: Concepts and Techniques;
Their Issues and Concerns Reference for Indian Agric