Sensor Networks Data Acquisition and Task Management For Decision Support of Smart Farming

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

2016 International Conference on Information Technology Systems and Innovation (ICITSI)

Bandung – Bali, October 24 – 27, 2016


ISBN: 978-1-5090-2449-0

Sensor Networks Data Acquisition and Task


Management for Decision Support of Smart Farming
Sinung Suakanto#1, Ventje J. L. Engel#2, Maclaurin Hutagalung#3, Dina Angela#4
#1,2,3,4
Information Technology, Institut Teknologi Harapan Bangsa
Jl. Dipati Ukur 80-84 Bandung, West Java, Indonesia
Email: [email protected], [email protected], [email protected], [email protected]

Abstract—Agricultural sector has been facing great


challenges in order to feed the increasing number of population
living in the world. In the future, it will be very difficult to rely on
traditional farming to produce food. Some researchers and
industry experts have begun developing smart farming by
utilizing information technology. This paper presents the
conceptual model and system design for decision support of
smart farming with network sensor applications in order to
perform necessary tasks required for farmers. We propose a
comprehensive model using Internet of Things (IoT) approach
which will be applied to agriculture. Data acquisition via sensors,
control and tasks management, and data analysis are considered
in the development of model and system design. In this system,
we propose a solution to help farmers facing problems of tasks Figure 1. Global Population and food supply [3]
management and planning, environment factors measurements,
and information distribution. Farmers are starting to use smart devices to turn on and
switch off sprinklers, yield monitors to measure harvest
Keywords—sensor networks, data acquisition, smart farming, volumes, soil moisture sensors, soil density sensors, infrared
task management, decision support, internet of things crop health sensors and a lot more [4]. This sensing technology
makes farm more intelligent and more connected then so-called
I. INTRODUCTION smart farming. The other term for smart farming is precision
Food needs are one of the basic needs for human beings. agriculture.
With the increasing number of population living in the world,
Smart farming and precision agriculture are alike but
food becomes a necessity that needs to be supplied in big
actually rather different. Precision agriculture is a farming
quantities and good quality. One of the sectors that need to be
management concept based on observing, measuring and
improved to the food supply is agriculture and livestock.
responding to inter and intra-field variability in crops. The goal
The agricultural sector is going to face great challenges in of precision agriculture research is to define a decision support
order to feed the 9.6 billion people that the FAO predicts are system (DSS) for whole farm management with the goal of
going to inhabit the planet by 2050 [1]. Food production must optimizing returns on inputs while preserving resources [5],
increase by 70% by 2050. This has to be achieved in spite of [6]. It is clear now that smart farming focuses in getting and
the limited availability of lands and the increasing need for monitoring the farmland environment. Meanwhile, precision
fresh water (agriculture consumes 70 per cent of the world’s agriculture focuses in data and information analysis to support
fresh water supply) [1]. Figure 1 depicts growth comparison farmers and stakeholders in making decisions [4]. Smart
between global population and food supply. In addition to a farming technologies, coupled with data-driven precision
growing scarcity of natural resources, farmers and farming agriculture, will enable farmers to reduce costs as well as
industries have to cope with climate change effects, such as maximize yields and profits [7]. In this paper, the two terms
notably greater rainfall variability and more frequent extreme will be used interchangeably.
weather events [2]. So it becomes important for farming
The most important things for smart farming are
stakeholders to understand about the changes in farmland
environmental conditions monitoring and water resources
environments more than ever.
management. The measurements of environmental conditions
Information technology and communication (ICT) can be and water management pair with farming market data should
empowered to help to manage agriculture. ICT can be become the data driver of precision agriculture. Authors have
empowered to monitor and record what happens at some crop developed and used various devices for detecting and
land to be reported and acted upon. Agriculture is going collecting environmental conditions [8] and realized the
through radical technological changes. Farmers are starting to opportunity to implement the custom systems for smart
use guidance systems to determine the position of their tractors. farming. The customization has to consider how to keep tracks

978-1-5090-2449-0/16/$31.00 ©2016 IEEE


of environmental factors in farmland and does communication a. Planting Plan (seeding, planting, fertilizing,
with external systems, such as irrigation system to regulate harvesting)
water and gateway system to notify the stakeholders [9]. b. Trucking plan
Moreover, the system must be able to do data analysis with
cheap computing resources. 4. Information Distribution

This paper presents the conceptual model and system a. Access and notification (SMS, app-based alert)
design for smart farming with network sensor applications in b. Dashboard
order to perform necessary tasks required for farmers. Data 5. Decision Support System (DSS)
acquisition, tasks control, and data analysis are also considered.
In this system, we developed the solution to help farmers DSS is used to determine critical decision from the
facing problems of tasks management and planning, soil result of the monitoring part. DSS will use statistical
factors measurements, and information distribution. analysis and basic machine learning to help farmers to
make decision in real-time.
II. CONCEPTUAL MODEL 6. Control Action
The main problems addressed in this paper are of tasks
management and planning, soil factors measurements, and Control action is used to follow up decision made by
information distribution. From these main problems, there has DSS automatically.
to be requirements analysis to define the needs and landscape
for building the conceptual model.
A. Requirement Analysis
In order to build the model, it’s important to understand the
constraints from sensor networks and signal access in rural area
[8]. Second, the amount of data processed have to be
considered also. These are the requirements for the model:
• Easy to be implemented is very important because
some farming stakeholders already have their own
system
• Easy to configure is the next important requirement.
• Able to extract and filter sensor data
• Support low-medium data traffic
• Able to connect to internet for reaching farming
industry stakeholders and getting market data
• Able to adapt with change and adversary that may Figure 2. Smart Farming Conceptual Model
happen
It’s important that modules built on top of this model have
B. Smart Farming Model to be light and easy enough to integrate with other system.
In the previous subsection, the requirements have been Some researchers have implemented a smart way for irrigation
defined and from those requirements the model now can be control [10]. Therefore, the already running irrigation control
built. Smart farming model can be categorized into building can be integrated within this model.
blocks that can be modeled as shown in Figure 2.
1. Monitoring
a. Soil characteristics (pH, macro mineral, such as N,
P, and K)
b. Air quality characteristics (temperature, humidity,
Ozone)
c. Water characteristics (pH, temperature, Dissolved
Oxygen, Fluoride)
d. Market monitoring (seed price, fertilizer price,
crops price)
e. Weather API
2. Management
a. Tracking System (Tractors, Farming Vehicle,
Farming Equipment)
b. Irrigation System
3. Planning
Figure 3. Example of Irrigation Control [10]
There are three main processes from data analysis users easier to understand the data. Users can find information
perspective: sensing, preprocessing, and inference [11]. about current condition from this dashboard application. The
Sensing is the acquisition of data from object and results from sensor reading also can also be displayed on
environments. Preprocessing is an important step in data dashboard application.
analytics. Preprocessing concerns with cleaning and filtering
the data and then extracting features that are needed for B. Gateway
inference. Inference is the core of data analysis. Inference Gateway stores the data locally and then send the data into
focuses in mathematical reasoning for decision support. These internet. Gateway consists of two modules: data collector and
three processes are included in the decision support part in the Remote Terminal Unit (RTU) [12]. Data collector is
model. responsible for translating environmental conditions into data
understood by computing machine. RTU module has main
III. SYSTEM DESIGN function to capture data from sensor both in digital or analog
At this section, we describe about the system design. We input. Package of specific sensors with Remote Terminal Unit
begin with an overview of systematic architecture. We also (RTU) will be placed in some places in the farm to monitor the
describe about the integration network sensor into TCP/IP environment. RTU architecture can be seen in Figure 5.
network with protocol communication [8].
A. Architecture
Proposed architecture for smart farming could be seen in
Figure. 4. The system consists of network sensors, server and
network. The sensors have main function to collect data.
Network sensors are collection of sensors placed in specific
area called swathe. Swathe is 1x1 or 2x2 meters of farmland.
Sensors will be placed in farm swathe and communication
between note and gateway is via IEEE 802.15.4 protocol. The
data from sensor are sent to server via Remote Terminal Unit
Figure 5 Remote Terminal Unit (RTU) Architecture
(RTU). Data transmission can use the radio access technologies
such as GPRS or EDGE. Sensors in farming have different and various
The server can be equipped with a database or GIS characteristics, some are digital sensors and the others are
(Geographical Information System) features. On the server analog sensors. Therefore, RTU uses analog and digital sensor
side, it can use cloud computing infrastructure. The in the form of voltage change [12]. To change signal from
applications will be provided for users. Applications will sensors become voltage change, it requires signal conditioning
retrieve data from sensor, so it will not directly access data circuit [12].
from sensors. The summaries of environment and market
conditions will be displayed in dashboard in order to make

Figure 4 Network Architecture of Smart Farming


C. Integrating Network into Internet amount of sensors will gradually increase. Some researchers
The connection between network sensor with the server and predict that the growth will be exponentially [11].
the users can use internet or TCP/IP network. Internet can be The smart farming data will sometimes need real-time
used as a basic underlying network because mostly network responses. There are primarily two modes of smart farming
hardware now supported with IP-Based. data: periodic observations or monitoring and abnormal event
Cloud computing could be proposed as central of data reporting. Periodic observations present demands due to their
processing to run service like service listener. The advantage of high volumes and storage overheads. Events on the other hand
using cloud is with rapid growing of sensors amount and are one-off but need a rapid response.
requirements, cloud can be expanded easily. It has functions to Because the amount of data will be huge despite the traffic
capture and store information sent from the remote client. management, we need key strategy to data nature constraints,
Otherwise, it could be used for the central data storage and as follows:
application server to display the processed results to the users.
• Availability. It will ensure that the sensors will read the
IV. DISCUSSIONS data farmers need and also manage authorization
The system design above absolutely has constraints. Some regarding who will be able to access the data.
Internet of Things (IoT) research had been conducted to • Usability. It will ensure that the data will help farmers
address the constraints, advantages, and weaknesses of IoT and stakeholders to accomplish the goals and also
design and implementation [13]–[15]. At this section, we will define format and data storage.
discuss constraints regarding traffic management and data • Reliability. It will ensure the consistency and integrity
of data and help in preprocessing of data in decision
nature. These two constraints are unique to implementation in
support part in the conceptual model.
farmland or rural area and connected to data analysis.
If information is incorrect or missing, decisions cannot be
A. Traffic Management made or - even worse - can be wrong, which may lead to
Traffic management will be discussed from front-end to undesired results. Since information is built on data we must
back-end. Sensors will generate traffic to gateway or RTU, so have insight in the quality of our data sources. The above
they are implemented in specific ways, as follows: strategy will ensure that the system will have good quality data
in order to generate needed information.
• Capable to active or die automatically when power is
cut or power supply is empty V. CONCLUSION AND FUTURE WORK
• Able to transmit data independently or autonomously
without user intervention This study has developed conceptual model and system
design that consist of network sensor, server, application and
• Able to coordinate with other node to avoid a flood of
also communication protocol for smart farming. The
data traffic resulting in transmission delay
conceptual model includes six parts responsible for monitoring,
Each sensor will send data to the nearest sensor and the management, planning, information distribution, decision
sensor will send the data to next nearest sensor and so on until support, and control action. This model considers not only
the data arrived at gateway. This will ensure that gateway will aspects in sensor networks and control but also data analysis
not be flooded by data simultaneously from all sensors. It will for decision support. The model concerns about data analysis,
make traffic from sensor to gateway more manageable. because sensor networks alone cannot help the farmers.
At gateway, traffic will be managed, as follows: Authors have developed a complete system design for
smart farming so farming industry stakeholders can implement
• Readings from sensors are not carried out continuously the system in easy way. The design focuses about sensor
but in a given period implementation in farm swathes, communication and
• Communication to server via internet is not carried out transmission protocol, and server infrastructure. Clearly, the
continuously but in a given period system design has weakness in form of constraints. The
• There is a certain period when system will not do constraints discussed are traffic management and data nature.
reading from sensors at all. This period is called force Authors provide the stakeholders with some strategies to
sleep period. overcome the two constraints.
Thus, the gateway will have three periods, namely sensors Further research is needed to implement the system design
reading period, data transmission period, and force sleep in a way to make sure that the constraint strategies will be
period. This strategy not only to make traffic management carried out accordingly. Two main ideas for further research
easier, but also to reserve power resources which are limited at with different perspective are: 1) business and market-oriented
sensor nodes and gateway. research regarding data quality to make sure decision support
B. Data Nature system can really support the stakeholders; 2) geospatial-like
research to ensure the signal reach from one node to other node
This data nature constraint occurs naturally when the and gateway.
amount of implemented sensors increase by time. With sensors
price getting cheaper and implementation becoming easier, the REFERENCES
[1] F. Guerrini, “The Future Of Agriculture? Smart Farming,” 2015.
[Online]. Available:
http://www.forbes.com/sites/federicoguerrini/2015/02/18/the-
future-of-agriculture-smart-farming/#338e9d6e337c.
[2] FAO, “Agriculture to 2050 – the challenges ahead,” FAO News.
[Online]. Available:
http://www.fao.org/news/story/en/item/36193/icode/.
[3] “FAOSTAT,” Food and Agriculture Organization of the United
Nations (FAO). [Online]. Available:
http://faostat.fao.org/site/291/default.aspx.
[4] R. van Hooijdonk, “Smart farming: the new agricultural
benchmark - Trendwatchter en Futurist Richard van Hooijdonk,”
2015. [Online]. Available:
https://en.richardvanhooijdonk.com/smart-farming-new-
benchmark/.
[5] A. McBratney, B. Whelan, T. Ancev, and J. Bouma, “Future
Directions of Precision Agriculture,” Precis. Agric., vol. 6, no.
July 2004, pp. 7–23, 2005.
[6] B. Whelan and A. McBratney, “Definition and interpretation of
potential management zones in Australia,” Proc. 11th Aust.
Agron. Conf., pp. 2–6, 2003.
[7] M. Ryu, J. Yun, T. Miao, I. Y. Ahn, S. C. Choi, and J. Kim,
“Design and implementation of a connected farm for smart
farming system,” 2015 IEEE SENSORS - Proc., pp. 1–4, 2015.
[8] S. Suakanto, S. H. Supangkat, R. Saragih, T. A. Nugroho, I. G.
Bagus, and B. Nugraha, “Environmental and disaster sensing
using cloud computing infrastructure,” Int. Conf. Cloud Comput.
Soc. Netw., pp. 1–6, 2012.
[9] C. D. M, G. R. R, J. S, and R. Priyatharshini, “Smart farming
system using sensors for agricultural task automation,” in
Technological Innovation in ICT for Agriculture and Rural
Development (TIAR), 2015 IEEE, 2015, pp. 49–53.
[10] “Smart Farms,” Smart Farms. [Online]. Available: http://smart-
farms.net/.
[11] C. Perera, A. Zaslavsky, P. Christen, and D. Georgakopoulos,
“CA4IOT: Context awareness for Internet of Things,” Proc. -
2012 IEEE Int. Conf. Green Comput. Commun. GreenCom
2012, Conf. Internet Things, iThings 2012 Conf. Cyber, Phys.
Soc. Comput. CPSCom 2012, pp. 775–782, 2012.
[12] S. Suakanto, T. A. Nugroho, S. H. Supangkat, and Suhardi,
“Smart sensor untuk pertanian,” in e-Indonesia Initiatives (eII)
Forum 2013 - ICT for Smart Society, 2013.
[13] D. Singh, G. Tripathi, and A. J. Jara, “A survey of Internet-of-
Things: Future vision, architecture, challenges and services,”
2014 IEEE World Forum Internet Things, WF-IoT 2014, pp.
287–292, 2014.
[14] J. Stankovic, “Research directions for the internet of things,”
IEEE Internet Things J., vol. 1, no. 1, pp. 3–9, 2014.
[15] D. Bandyopadhyay and J. Sen, “Internet of things: Applications
and challenges in technology and standardization,” Wirel. Pers.
Commun., vol. 58, no. 1, pp. 49–69, 2011.

You might also like