International Journal of Computer Applications (0975 – 8887)

Volume 179 – No.21, February 2018

Improving Agricultural Production using Internet of

Things (IoT) and Open Source Technologies

Benjamin Odoi-Lartey Edward Danso Ansong

Kwame Nkrumah University of Science and Kwame Nkrumah University of Science and
Technology Technology
Computer Science Department Computer Science Department
PMB University Post Office, Kumasi, Ghana PMB University Post Office, Kumasi, Ghana

ABSTRACT 2. LITERATURE REVIEW

This article proposes a sensor network framework built using This article presents the literature review on IoT and more
Internet of Thing (IoT) and Opens Source tools to automate specifically its implementation in the agricultural sector. It
farm fields monitoring in order to have the farm communicate provides an overview of previous research on the use of
real time with the necessary stakeholders. The study reviewed technology in Agriculture specifically Internet of Things
a couple of Internet of Things technology in Agriculture and technologies. It looks at a couple of Internet of Things
used Open source software and hardware technologies to technologies that are been researched into and currently
build-up a prototype that is cheaper and easy to deploy making agriculture less difficult and increasing farm yield.
compared to existing proprietary systems. The designed
system uses Raspberry Pi model 3 that serves as base station Within less than half a decade, much has been researched into
for the sensor network and connects with an Arduino Uno so far as IoT in the field of Agriculture is concerned. It is very
serving as an intermediary between various sensors and the essential to set the context of the literature review work by
Raspberry Pi. Among the sensors integrated in the system are looking into previous research works by authors in the field of
DHT11 Soil Sensor, Humidity and Temperature Sensor, sensor networks, sensor boards, data monitoring, security of
Passive Infra-Red (PIR) Sensor and Raspberry Pi Camera IoT and finally reviewing on the field of IoT in Agriculture.
Module. The study deployed the prototype and tested it over IoT is consistently developing and is a hot research point
Open IoT API platforms like KAA and OpenIOT for where openings are unending. Creative impulses are vast
automation, monitoring and data collection which have put it nearly reshaping the present type of web
into a changed and coordinated rendition. With the persistent
Keywords headways in innovation this potential IoT development is
Internet of Things, IoT, Raspberry Pi, Arduino, Sensors, descending the street which is prospering as a pervasive
iFarm, Open Source, Agriculture. worldwide figuring system where everybody and everything
will be associated with the Web [5].
1. INTRODUCTION
New inventive Internet of Things systems are inclining to
The Internet has been the greatest invention in the 21st
issues of Agriculture production and escalating the worth,
century improving various facets of life and advancing society
supportability and cost effectiveness of rural agricultural. The
in health, education, transportation, and communication.
wide-ranging ranches we have today can, for example, use
Agriculture is no exception from these improvements so far as
these Internet of Things technologies to screen sensors that
the Internet is concerned [8].
can identify soil fertility and nutrients readiness remotely to
Many primary experts in the field emphasised agriculture due help remotely oversee and control their reaping and water
to abundant resources associated with it and its ability to system hardware, and use computerized reasoning based
transfer surpluses to the more important industrial sector. investigation to rapidly break down operational information
Stringer [15] stated that “the primary role of Agriculture in the consolidated with outsider data, for example, climate
transformation of a developing economy was seen as administrations, to give new bits of knowledge and enhance
subordinate to the central strategy of accelerating the pace of decision making [17].
industrialisation” [15] [16].
According to FAO [3] [4], World population is estimated to
Enu [2] postulates that field of agriculture still remains one of increase by 33% in the year 2050 and rural Agricultural
the most essential sectors for the economies of most countries production will have to increase by 60% to feed this
across the globe. Ghana for instance has Agriculture population. Farming must along these lines change itself on
contributing about 40% to its gross domestic product (GDP). the off chance that it is to bolster a developing worldwide
populace and give the premise to financial development and
Research has shown that the focus of earlier works to improve neediness lessening. Environmental change will make this
Agriculture has been with the mechanical machineries and assignment more troublesome under the same old thing
improvement of products such as varieties of seedlings and situation, because of unfavorable impacts on horticulture,
breeds of animals [6]. requiring spiraling adjustment and related expenses.
The Internet of things has taken over the world and businesses FuBing [7] proposed a system structure based on Internet of
that seek to integrate it in their setup are usually on top of Things that integrates wireless sensor network, 4G network
their competitors or predicted to be on top in the coming and the Internet to realize the farmland air and soil
years. [12] environment data collection and analysis. The proposed

36
 International Journal of Computer Applications (0975 – 8887)
Volume 179 – No.21, February 2018

system design has three layers thus the farmland environment 2.2 Review of Related Architecture
information collection layer, the network transport layer and
the system application layer. In between the network layer and
the sensory layer are Mobile devices that are used to collect
data such as photographs, and description voices and
transmitted to the cloud platform with accessing Internet by
4G wireless communication network.
Ma et al [10] report in their paper that via sensory nodes,
agriculture and other fields would be connected to Internet of
Things thus allowing us to establish a common relation
among agronomists, farmers farm fields and other stake
holders irrespective of their geographical locations and this
will aid the stake holders to have a better considerate of crop
growth models and improve on methodologies in farming.

Ma et al [10] made a comparison between two different

available boards thus MSP430F1611 and
ATmega1281/AT86RF230. Out of these boards this article
considered using an open version of similar boards to build
another architecture called the iFarm Architecture.

2.1 IoT APIs (Application Programming

Interfaces)
Jayaraman and Palmer [9] in their research proposed a
semantically improved digital agriculture use case Phenonet
developed with OpenloT platform (Phenonet-OpenIoT). The
platform demonstrated how the semantic interoperability of
the OpenloT platform helps in addressing the challenges
posed by Phenonet applications. Their work extended the Fig 1: Fubing Architecture [7]
OpenloT stage to bolster the Phenonet utilize case by building
up an area particular Phenonet philosophy in view of W3C
SSN standards to connection sensor assets with space ideas
and joining novel semantic thinking capacities and future
work incorporates augmenting the utilization of OpenloT
stage to more assorted administrations in computerized
farming.
Among other Open Source platform that this journal paper
revealed included but not limited to; Kaa, ThingsBoard,
OpenIoT, Predix, Oracle Internet of Things, Carriot,
SalesForce, Cisco IoT Platform, ThingWorx etc.

Fig 2: Ma et al Architecture [10]

3. METHODOLOGY – DESIGN AND

PROTOTYPE BUILDUP
According to Rankin and Rankin any agriculture technology
should address practical problems that are agreed upon by
farmers and technology providers and the technology should
fit into the five criteria for judging sustainable agricultural
technology [13].

37
 International Journal of Computer Applications (0975 – 8887)
Volume 179 – No.21, February 2018

The five criteria for judging sustainable agricultural

technology includes; Localization, Holistic, Usefulness,
Affordability and Democratic. They research adopted the
IDEO Human-centered Design approach as a guideline for the
Prototype buildup.

Fig 3: IDEO Human-Centered Design Process

3.1 Schematics for the various Sensing

Modules
Soils Sensor Schematic
Aronson [1] in his Farmbot white paper shows soil sensor has
three key connectors that links it to the Arduino
Microcontroller. It has a VCC for voltage with maximum
volts of 5V with another connector for ground and finally a
single connection for data transmission. The connector for
data or signal transmission can be Digital or analog and the
Arduino microcontroller has connection for all the two. The
Arduino is programmed using Arduino IDE Software. The
program is uploaded and output is displayed on the Console.
Fig 5: Humidity and Temperature Sensor Schematic

Motion Detection and Image Capture

Module: the motion detection and Image Capture module
is designed on the Raspberry Pi 3 system on a chip (SOC)
board which has the capabilities of both embedded systems
and a full Input/output computer system. A motion detection
sensor is interfaced with the GPIO of the Raspberry Pi and the
camera module is connected to the normal display connector
for Camera module. The logic behind is simplified caption of
the camera anytime an unusual image is seen on the field.
Aside that, timed captions are made from the day one of setup
through to harvesting and post harvesting.

Fig 4: Soil Sensor Schematic The various module can work independently on their own but
integrating these modules brings about a more advance system
Temperature/Humidity Module (DHT11) that works cohesively producing a better results for field
The DHT11 sensor has four connectors out of which three monitoring and accumulating field data over a period of time.
links to the Arduino microcontroller for sensed data The system is designed by integrating these various modules
interpretation and communication as shown in Figure 3.18. It to have a more advance system. To implement the system
has a VCC for voltage with maximum volts of 5V, another for there is the need to consider each farmers preference. [11]
ground and finally a single connection for data transmission.
The connector for data or signal transmission can be Digital or
analog and per the choice made, the Arduino will work
interpret accordingly. The program is uploaded and output is
displayed on the Console or pushed to the cloud API platform
for remote access.

38
 International Journal of Computer Applications (0975 – 8887)
Volume 179 – No.21, February 2018

3.2 Overview and Architecture of the

iFarm Architecture
The proposed system is made up of both open source
hardware and software components as shown in Figure 3.2.
Out of the brainstorming with the experts on the field and
identified quality attributes shown in Table 3.1: System
quality attributes. The bottom most layer of the architecture
has the various sensor that are connected to the base station
(BS) which comprises Raspberry Pi Arduino and GSM
Module. The BS is interfaced with the cloud through services
provided by telecommunication companies for Internet
connectivity sake since the third party cloud service run on the
Internet. Among sample API platforms chosen as testbed
Fig 6: Motion Detection and Image Capture Schematic KAA, Allseen Alliance and OpenIoT which sits in the cloud
as both application and database server to the user [14].

System Architecture Overview

Fig 7: System Architecture Overview

3.3 Sub layers of the iFarm Architecture sensors which senses the changes of voltages and feeds the
changes of measured voltage to an ADC (Analog to Digital
Sensory Layer Converter), the digital converted signals is sent to the CPU
The bottommost layer is called the Sensory Layer which is (Central Processing Unit) for further processing. The sensing
basically made up of collection of sensors that reads sensory subsystem in the proposed prototype here includes, Camera
data from the farm fields and get it logged to the base station. module, Soil sensor, Humidity and temperature sensor.
By automation, data is generated at periodic intervals as the The sensory layer interfaces with a processing subsystem which
system has been programmed. The generation of the data occurs has an external memory subsystem and gives the system the
per assigned periods of activeness meaning when the nodes are capability of reading and storing data onsite.
not in active mode, they are deemed to be asleep or in a passive
mode. Network Layer

Fig 9: Data Logging Unit

Fig 8: Sensing Unit
The GSM/GPRS module serves the purpose of creating
Among the sensors used in the prototype build up are Soil
connectivity to the Internet and giving GPRS functionalities to
Sensor, Humidity/Temperature Sensor (DHT11), and Motion
the system. It has a transceiver having a wireless antenna,
Sensor etc. A collection of these sensors based on farmers’
responsible for transmitting the radio signals using the optical
preference makes up the sensory subsystem. It consist of tiny
signals or infrared waves as a medium, operates in ISM

39
 International Journal of Computer Applications (0975 – 8887)
Volume 179 – No.21, February 2018

(Industrial, Medical and Medical purposes) band. It uses a free station, sensor readings are continually uploaded to the cloud
radio frequencies for communication in most countries, platform and this information is made available for access from
allocated in the international table of frequency allocation, any web browser with Internet connectivity. The readings can
contained in Article s5 of the radio regulations (volume 1). always be read numerically or graphically through the API
(Wireless Communication, Jean Paul Linnartz Reference dashboard at the user end. The readings are kept in the Database
website). The Data Logging unit is built around the Network Server which is stored in History log from the initial setup of
layer and as such giving it the capability to easily log data to the the system to Current time.
online repository for further analysis. This unit comprises
Arduino and Raspberry Pi and the main Network capability is
dependent on the GSM module as it provides network
connectivity to the raspberry pi.

Application Layer
The system is designed with the application layer in mind to
make data access easier and readily available for users. The
uploaded sensor data sits on the cloud and users can access the
data through personal computers or smart phones with internet
connectivity. In order to make data presentation so meaningful
to users, the sensor readings should be displayed with
Fig 10: Data Access Unit
visualizations like graphs to give a brief or summary of data at a
glance. The GSM/GPRS module attached to the system makes
location of each base station (BS) easily identified with base
coordinates. Per the specified time interval programmed on base

Fig 11: iFarm System Architecture

40
 International Journal of Computer Applications (0975 – 8887)
Volume 179 – No.21, February 2018

4. SYSTEM IMPLEMENTATION, before the three are integrated and tested as a whole unit or
system.
RESULTS AND ANALYSIS
4.1 Implementation After a successful testing of the three modules, integration is
To ensure a successful, affordable experimentation and made to have a complete system interlaced with the cloud
prototype built, open source modules of embedded hardware platform for remote monitoring and control. There are
and software systems were used out of which the prototype numerous cloud platforms both Open Source and Proprietary
will also be used to demonstrate how data is sensed or that supports and integrates Internet of Things. In this work,
captured and sent to the cloud platform to be assessed by the the chosen test bed platforms included KAA, Cayenne,
farmers, agronomist and other stakeholders Three different Allseen Alliance and OpenIOT which gave tremendous result.
modules thus Moisture Module, Temperature Module and Data could be represented in both text and graphical format
Motion detection module were tested. The different modules which makes it very easy to understand and interpret the
were tested separately for individual or unit testing purposes results at a glance

4.2 Complete System Prototype

Fig 12: System Prototypes

4.3 RESULTS AND CONCLUSION [3] FAO. (2009). FAO - News Article: 2050: A third more
Upon integrating the three different modules into a system, all mouths to feed. Rome. Retrieved from
the three modules could easily be integrated and the various http://www.fao.org/news/story/en/item/35571/icode/
information expected to be accessed remotely could be [4] FAO. (2013). Climate-Smart Agriculture Sourcebook.
achieved at any point in time over existing Open Source APIs. Sourcebook on Climate-Smart Agriculture, Forestry and
It can be concluded that with the numerous availabilities of Fisheries. Retrieved from
sensors and open source technologies, any kind of information http://www.fao.org/docrep/018/i3325e/i3325e00.htm
needed by a farmer or agriculturist can easily be obtained [5] Farooq, M. U., & Waseem, M. (2015). A Review on
remotely with the click of a button or swipe on a smartphone. Internet of Things ( IoT ). International Journal of
Computer Applications (0975 8887), 113(1), 1–7.
5. REFERENCES https://doi.org/10.5120/19787-1571
[1] Aronson, R. L. (2013). FarmBot: Humanity’s open-
source automated precision farming machine. Retrieved [6] Floros, J. D., Newsome, R., Fisher, W., Barbosa-
from https://farmbot.io/whitepaper/ C??novas, G. V., Chen, H., Dunne, C. P., … Ziegler, G.
R. (2010). Feeding the world today and tomorrow: The
[2] Enu, P. (2014). Analysis of the Agricultural Sector of importance of food science and technology.
Ghana and Its Economic Impact on Economic Growth, Comprehensive Reviews in Food Science and Food
5(July). Safety, 9(5), 572–599. https://doi.org/10.1111/j.1541-

41
 International Journal of Computer Applications (0975 – 8887)
Volume 179 – No.21, February 2018

4337.2010.00127.x Forum, (January), 40. Retrieved from

http://www3.weforum.org/docs/WEFUSA_IndustrialInte
[7] FuBing. (2015). The Research of Agricultural Expert rnet_Report2015.pdf
System based on IOT (pp. 1400–1403). Atlantis Press.
[13]Rankin, D., & Rankin, D. (2015). Opening Agriculture :
[8] Goldin, I. (2014). FUTURE OPPORTUNITIES , aKey Alternative Technological Strategies for Sustainable
Trends Shaping the Global Economy and Society, Farming. Sustainability. Retrieved from
(October). http://scholar.colorado.edu/cgi/viewcontent.cgi?article=2
[9] Jayaraman, P. P., & Palmer, D. (2015). Do-it-Yourself 124&context=honr_theses
Digital Agriculture Applications with Semantically [14] Raul. (2016). IoT Applications With Examples - Internet
Enhanced loT Platform. 2015 IEEE Tenth International Of Things Wiki. Retrieved April 26, 2017, from
Conference on Intelligent Sensors, Sensor Networks and http://internetofthingswiki.com/iot-applications-
Information Processing (ISSNIP) Singapore, 7-9 April examples/541/
2015, (April), 7–9.
[15] Stringer, R. (2001). How Important are the Non-
[10] Ma, J., Zhou, X., Li, S., & Li, Z. (2011). Connecting traditional Economic Roles of Agriculture in
Agriculture to the Internet of Things through Sensor Development? (No. 118). Most. South Australia.
Networks. 2011 IEEE International Conferences on Retrieved from
Internet of Things, and Cyber, Physical and Social http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.
Computing. 1.29.3767&rep=rep1&type=pdf
https://doi.org/10.1109/iThings/CPSCom.2011.32
[16]Stringer, R., & Pingali, P. (2004). Agriculture ’ s
[11] Mohanraj, I., Ashokumar, K., & Naren, J. (2016). Field contributions to economic and social development.
Monitoring and Automation using IOT in Agriculture Electronic Journal of Agricultural and Development
Domain. Procedia - Procedia Computer Science, Economics, 1(1), 1–5.
93(September), 931–939.
https://doi.org/10.1016/j.procs.2016.07.275 [17] The World Bank. (2005). Module 4: Investments in
sustainable agricultural intensification. Agriculture
[12] O’Halloran, D., & Kvochko, E. (2015). Industrial investment sourcebook. https://doi.org/10.1596/0-8213-
Internet of Things : Unleashing the Potential of 6085-X
Connected Products and Services. World Economic

IJCATM : www.ijcaonline.org 42