eBooks

Smart Agriculture: An Approach

Towards Better Agriculture
Management
Chapter: Sensor Network based Automatic Irrigation Management System

for Agricultural Crops

Edited by: Dr. Aqeel-ur-Rehman
Published Date: December 2014

Published by OMICS Group eBooks

731 Gull Ave, Foster City, CA 94404, USA

Copyright 2014 OMICS Group

All book chapters are Open Access distributed under the Creative Commons Attribution
3.0 license, which allows users to download, copy and build upon published articles
even for commercial purposes, as long as the author and publisher are properly credited,
which ensures maximum dissemination and a wider impact of our publications. However,
users who aim to disseminate and distribute copies of this book as a whole must not seek
monetary compensation for such service (excluded OMICS Group representatives and
agreed collaborations). After this work has been published by OMICS Group, authors have
the right to republish it, in whole or part, in any publication of which they are the author,
and to make other personal use of the work. Any republication, referencing or personal use
of the work must explicitly identify the original source.

Notice:
Statements and opinions expressed in the book are those of the individual
contributors and not necessarily those of the editors or publisher. No responsibility
is accepted for the accuracy of information contained in the published chapters.
The publisher assumes no responsibility for any damage or injury to persons or
property arising out of the use of any materials, instructions, methods or ideas
contained in the book.
A free online edition of this book is available at www.esciencecentral.org/ebooks
Additional hard copies can be obtained from orders @ www.esciencecentral.org/ebooks

eBooks

Sensor Network based Automatic Irrigation

Management System for Agricultural Crops
Mushtaq Ahmad, Rahim Khan,
Ghulam Ishaq Khan Institute of Engineering Sciences and Technology Topi. KPK,
Pakistan
*Corresponding author: Ghulam Ishaq Khan Institute of Engineering Sciences and
Technology Topi. KPK, Pakistan

During the last decade, in Asian countries including Pakistan, water resources were
depleted due to severe environmental conditions. Agriculture sector is the major source to
fulfill the food requirements of continuously growing population. The vital requirement of
sufficient and on time availability of water is vital through a an advanced irrigation system
for good yield of agricultural crops.. Throughout the world, irrigation schedules are based on
farmers experience and changes according to weather fluctuation. With the passage of time,
the importance of an optimized, environment friendly and intelligent irrigation management
system are increasing where irrigation schedules are based on soil water contents and
environmental parameters. In this chapter, Sensor Network based Automatic Irrigation
Management System (SNAIMS) is presented that consists of agriculture sensor boards,
gateway and a computer. In SNAIMS, the irrigation activity is need based and decisions are
based on soil moisture content, temperature, air humidity and leaf wetness data obtained
through sensor nodes deployed at different locations. SNAIMS investigates collected data
and identifies water deficient locations. In case of vulnerable situations, SNAIMS activates
alarming unit and sends a message over LAN. It is capable of detecting noisy data using
a simplified procedure and avoids it while taking irrigation related decisions. Likewise,
detected noisy data is prohibited from storing in database. SNAIMS provides GUI based
solution for showing data and enables farmers to monitor and control irrigation operation of
their crops remotely. The proposed system is deployed over an area of 6050 square meters
inside GIK Institute orange orchard.

Introduction
Throughout the world, water is considered as an important resource as far as living
creatures are concerned because without water life on this earth is almost impossible. As
we know that every living creature uses water according to its needs. Due to its importance
in our lives, it is necessary that water resource should be used in an efficient manner and
a single drop of water should not be wasted. Almost every sector uses water but the main
consumer of water is the agriculture sector. This sector consumes approximately 70% of
the world water resources for irrigating farmlands and is liable to fulfill about 40 percent
of the worlds food requirement [1]. As the world population is increasing day by day and
so is the Foods demand. In order to fulfill foods requirement of the growing population,
farmers needs to grow more crops and orchards and hence uses more water. The problem of
water scarcity is controlled by utilizing modern technology in irrigation procedures namely
DNA based breeding, sprinkling of gardens, drip irrigation, and other advanced irrigation
1

technologies. However, these procedures are in experimental stages and farmers, scientists,
research funding organizations and investor cooperation are required to commercialize
them [1]. In addition, farmers are needed to be convinced on using modern technologies
in irrigation and to avoid the traditional irrigation procedures especially flooding. Modern
technology is extremely useful in controlling over irrigation as well as under irrigation
and determines when and where irrigation is desperately needed. Thus an efficient and
intelligent irrigation management system based on modern technology is needed that caters
to produce maximum crop yields even for low supply of water and utilizes available water
resources in an efficient manner [2].
The efficiency of an irrigation management system is highly dependent on irrigation
methods and schedules utilized such as surface irrigation or drip irrigation. Each method
has its own merits and demerits and its usefulness is totally dependent on suitable
circumstances. Likewise, irrigation procedures of various crops are completely different
from one another such as the irrigation procedures and schedule applies for orchards are
different from that of wheat crops or other crops. As we know that proper irrigation schedule
is necessary for maximizing crops yields and quality because both under irrigation and
over irrigation are extremely harmful and proper mechanism is needed to avoid them. Both
have severe effects over crops yield, quality, increase vulnerability to diseases, and most
importantly wastage of water resources [3]. In order to manage under and over irrigation,
different methods are proposed in literature for automating, optimizing and organizing
irrigation activities [4]. But most of the farmers especially in under developed countries
including Pakistan are still sticking to the traditional manual irrigation procedures which
causes wastage of water.
During the last decade, reduction in the cost of technology especially sensors and
actuators attracted researchers organizations, scientists and investors to focus on WSNs
technology and its usage in different applications from experimental phase to commercial
levels [5,6]. Although, WSN technology has some limitations associated with it but most
were addressed by adopting new routing protocols, preprocessing of data, sensors sleeping
modes etc. [7-9]. In agriculture sector, initial experiments of WSN technology were carried
out in controlled environment namely green houses and most of these experiments were
successful. Similarly, a number of experimental studies were carried out in open fields as
well but still requires more work to be done in this area [10]. As far as irrigation operation
is concerned, irrigation schedules depend upon weather conditions, soil properties, crops
varieties. For example, irrigation schedules of wheat crops in hot areas are completely
different from that of cold areas that is irrigation schedules adopted in Lahore area is different
from the irrigation schedules of district Bunir. Taking into account the complexities and
problems associated with traditional irrigation procedures automatic irrigation controlling
mechanism was introduced, which would not only result in time saving but also utilizes
different resources (water) efficiently. Among other automation mechanism WSN is one of
them which is used to collect environmental parameters and soil properties continuously
and take necessary action or assist the farmer in this regard [11].
In order to overcome the problems associated with traditional irrigation procedures,
WSNs based irrigation controlling mechanisms were presented in literature. Although, these
methods were useful and solved most of the problems but each of these methods carries
some limitations as well. Most of these methods were developed to work efficiently in special
but for the changing circumstances the results are not convincing. For example, WSNs
developed functions of tested and implemented in vineyard monitoring system which may
work fine for vineyard but the same network cannot work for Maize crop field, because
of different crop and circumstances. Although, these methods are useful in a sense that
it reduces labors overhead, save time, reduce load on farmer, and regulate irrigation but
these systems do not solve over irrigation, under irrigation, need based irrigation and water
wastage. Similarly, noisy data is rarely considered in these projects. In order to solve these
2

problems, an efficient irrigation management system is needed that not only automates
crops irrigation mechanism but also considers environmental parameters, soil properties
and noisy data removal while taking various decisions. Over irrigation and under irrigation
is solved if the system is able to collect soil properties regularly and utilize it before making
irrigation decisions.
In response to the problem stated, this chapter presents a Sensor Network Based
Automated Irrigation Management System (SNAIMS) for oranges orchards. SNAIMS is
capable of collecting soil properties and environmental parameters continuously through
various sensors integrated with agriculture pro boards deployed at different locations in
oranges orchard. Irrigation schedules of different plots are automatically controlled and
are based on soil moisture contents, air temperature, air humidity and leaf wetness. These
parameters are sensed after a defined time interval and investigated by decision support
system installed at central location. WSNs nodes have to rely on it are on board batteries.
In SNAIMS, we consider efficient utilization of batteries and data is preprocessed by the
processing unit before it is handed over to the transceiver. A season dependent threshold
value is defined for soil moisture contents along with temperature, air humidity and leaf
wetness threshold values. During analysis of data if the soil moisture value is less than the
defined threshold values then irrigation activity are crucial. Once the vulnerable condition
is detected, SNAIMS delays its decision until upcoming reading from similar sensor node.
After collecting next reading, three readings are compared and if the vulnerable value is
found correct then it activates alarming unit and send a message over LAN. However, if
the vulnerable data value is identified as noise, malfunctioning of sensor nodes then it is
discarded and thus increasing system accuracy in terms of decisions. It enables farmers to
identify locations where irrigation is needed by viewing different graphs plotted by SNAIMS
automatically. In order to inform farmers about vulnerable conditions, it activates the
alarming unit and sends an alarm activation message over LAN. In subsequent section
of this chapter, a brief overview of WSNs technology in the context of agriculture sector is
presented.

Wireless Sensor Networks in Agriculture

The reduced size of a sensor node makes it one of the prominent and contested among
other electronic devices to automate various agriculture related activities. Apart from the
limitation associated with sensor nodes, this technology is extremely useful for different
application areas including agriculture sector. In agriculture, sensor nodes are used for
collecting environmental parameters and soil properties which have drastic effects on crops
yields, quality, cost and irrigation schedules. Irrigation schedules, pesticides sprays and
fungicides sprays are completely dependent over environmental parameters and sensor
nodes are proficient to provide this information regularly. Soil parameters are extremely
useful to determine various crops suitability and also depicts whether crops need water
or not [12]. Additionally, WSNs are used in different agricultural projects ranging from
greenhouses automation to open fields deployment as shown in Figure 1 [13]. Some of
experimental studies are presented here briefly.
In greenhouses environments, dew condensation problem requires major attention
because most of the diseases occur due to this. WSNs are used to automatically control
or prevent dew condensation problem inside greenhouses [14]. Three different types of
sensor nodes were used namely sensing nodes for gather data, computation over gather
data is performed by base nodes and relay nodes are used to adjust environment inside
greenhouse. Additionally, a server is used for applying Bahrenburg formula to calculate dew
condensation, processing saving gathered data. Bangladesh University of engineering and
technology, MIT and CENS (a small unit of University of California log Angles) has started a
joint project to comprehend the ratio of arsenic compound presence in groundwater
3

Figure1: Soil Conductivity and pH sensor nodes

In order to achieve their goal, WSN comprised of 48 sensor nodes were deployed in rice
fields for short period of time (ten days). Sensor nodes were used to measure arsenic, soil
moisture, temperature, calcium, carbonate, chloride, nitrate, and PH etc. [14]. WSN was
used for collecting data which is analyzed at laboratory. In Malawi, a WSN based project
was carried out by Zennaro et al., [15] to understand water quality. WSN was comprised of
four nodes relay on single hop communication and each node was measuring pH, turbidity
and redox (reduction/oxidation). Initial focus of the project was on automatic fault recovery
mechanism because the system has to be deployed in remote area.
In Spain, WSNs comprising of three different types of nodes were successfully deployed
and tested for monitoring salt, soil water contents, and temperature [16]. Looking to the
importance of water in life of living creatures, a WSN was applied to identify different means
of saving water. By thoroughly investigating the relationship between the capacity of shower
nozzle and sensor node radius, an expression is identified that describe how area under
consideration will be covered by least number of sensor nodes [17]. A conceptual model
of WSN based automated irrigation management system was proposed by Feliciano et al.,
[18] but only realized a prototype model of their presented system by utilizing soil moisture
sensor, temperature sensor, electro mechanical sensor and capacitance sensors along with
computer controlled decision support system. In 2011, a WSN based architecture for precision
agriculture is presented by Anurage et al., [19]. Their proposed architecture comprised of
three distinct components. First component of their architecture is intelligent nodes capable
of sensing different parameters necessary for an automatic agricultural system. Second
component is a mesh network of wireless nodes used for communication with other nodes
with energy efficient routing protocols. Third activation part is used for activating various
units automatically. In rural areas, currently farmers experiences are used for determining
and managing seedling periods of water melons. Likewise, environmental temperature
has a dominant role in its seedling periods as well but presently thermometers are used
to measures temperature values. Zuol et al., [20] presented a WSN based monitoring
system deployed in plastic greenhouse of watermelon. Three different parameters namely
temperature, air humidity and light are sensed periodically using different sensors and then
a thorough investigation is performed at centralized location. Due to the valuable role of
water in growth and yields of different crops, water level monitoring is considered seriously
4

by research community. In this regard, a design of water level monitoring system based
on wireless sensor network in presented in [21]. In addition, an algorithm is presented for
delivering maximum packets to destination.
A process control strategy based management system for the production of an efficient
water melon seedling process is realized by S. Ke et al., [22]. WSNs along with decision support
system, RFID based method, and queuing theory based algorithm for seedling in nurseries
is presented. In Lofar project, a pilot study was carried out to understand and describe the
potentials of WSNs in detecting various diseases. Phythophthora is a fungal disease and was
used as a case study. Environmental parameters along with soil properties are sensed after
a predefined time interval and investigated to identity this disease. The results of this pilot
study were convening but the complete projects results are not reported yet [23]. Vineyard
monitoring system [24] utilized wireless sensor network for disease prediction, pest controls,
identifying disease locations and helps managers in handling different activities smoothly.
Kotamaki et al., [25] deployed sensor nodes at different location for perceiving river basin
water quality and investigated soil parameters effects over crops yield. In order to prevent
the occurrence of tomato disease favorable conditions inside greenhouses, sensor nodes are
deployed for periodically sending information to disease prevention system which analyzes
received data by comparing it with previously stored data for taking necessary actions [24].
Various studies on automatic controlling of greenhouses ranging from automatic control of
a single parameter to overall management are reported in [13-16]. In Common-sense net
project, sensor network based decision support tool was designed for marginal agriculture
in Indian state of Karnataka. This system helps in conserving rainy water for long term
usage and sensor nodes were deployed to collect environmental data for controlling different
crop diseases [26]. Flow AID presented a controlled irrigation mechanism useful in situation
where water quality and quantity is limited [27]. Automatic drip irrigation management
system for cherry trees was proposed by Dursun et al., [28]. Soil moisture sensors were
deployed at different locations for report to base station after certain time interval. A.
Pardosi et al., [29] investigated the possible integration of rote zone sensors into wireless
sensor network and its utilization in water resource limited conditions. X. Ding et al.,
presented underground sensor pivot system for managing irrigation activities automatically
[30].Different site specific automatic irrigation management systems based on wireless
sensor networks were investigated for enhancing crop yield and efficient utilization of water
resources [31,32,33].

Sensor Network Based Automated Irrigation Management System

(SNAIMS)
Considering the potentials of WSNs in different application areas especially agriculture
sector, Sensor Network Based Automated Irrigation Management System (SNAIMS) is
presented and realized to automate the irrigation activity of oranges orchard. SNAIMS
irrigation related decisions are based on temperature, air humidity, leaf wetness and soil
moisture values collected from agricultural sensor boards positioned at different locations in
oranges orchard of GIK Institute. Water deficient locations are identified by comprehensively
investigating collected environmental and soil data. The proposed system is comprised of
two different parts namely hardware unit and software unit. Hardware unit is utilized for
collecting and storing information whereas software unit examines the collected information
to decide whether irrigation is necessary or not. Both of these units are briefly explained in
subsequent section.

Hardware Unit
The hardware unit of SNAIMS comprised of agriculture boards and pro-board, soil
moisture sensors, temperature sensors, humidity sensors, leaf wetness sensors, WaspMote gateway, and a computer used as a sever. Wasp-mote agriculture boards are used
5

in this experimental setup due to distinguished feature (long range capabilities, easy
programming, and designed especially for agriculture sector). Likewise, fourteen different
sensors are integrated with it at the same time and to prolong batteries lifetime, its switches
to different operational modes. The schematic diagram of wasp-mote agriculture sensor
board is shown in Figure 2. It has two types: simple and pro boards. A pro board has
the capacity of connecting four extra sensors compared to a simple board. Each board is
equipped with 3.7 volts lithium-ion battery having lifetime between 6-12 months depending
heavily on sampling rate of different sensor and wireless communication range. In order
to recharge batteries, it has a 12 volts solar penal unit as well [33]. The board has AT
mega 1281 microprocessor and 2 GB micro-SD card port. In Figure 3 [33] depicts soil
moisture sensors that are integrated with wasp-mote board in group of three. For precise
and accurate analysis of soil moisture, three soil moisture sensors are connected to one
board at the same time and are placed at different depth levels. XBee-802.15.4 is used for
wireless communication as pointed out by dotted arrow in Figure 2 [33]. It communicates
with microcontroller at baud rate of 38400bps speed through UART_0. Its transmission
range is about 500meters. Figure 4a shows SHT75 modules (sensirion) used for sensing
temperature and air humidity [33]. Figure 4b shows leaf wetness sensor integrated with
wasp-mote agriculture board during this experimental study.

XBee
Communication
Module

Figure 2: Wasp-mote Agriculture Board with Integrated Communication Module.

Figure 3: Soil Moisture Sensor Unit.

Figure 4: a. Sensirion temperature and humidity sensor b. Leaf Wetness Sensor

Figure 5: Wasp-mote Gateway

Gateway is an important part of wireless sensor networks and acts as a bridge between
sensor nodes and server. Figure 5 [25] shows wasp mote gateway. It communicates wirelessly
with sensor nodes and serially communicates with computer must set to on before deploying
it in experimental fields. In order to perform processing over sensed data, the data needs to
be stored in allotted memory location. Xbee802 is the communication module name followed
by different functions such send () mean broadcasting data depending over embedded mac
address of destination module.

Software Unit
Prior to the actual deployment of SMAINS in open field, agriculture boards are programmed
using wasp-mote integrated development environment. Keeping in view the importance of
batteries lifetime in wireless sensor networks scenario, boards programming was done in a
manner to efficiently utilize available power. In this regards, every sensor unit was switched off
after performing its intended task of probing environment. A sample of wasp-mote working code
is presented below. The status of an agriculture board [Starting Wasp-Mote Module]

Sample of working code of Wasp-Mote Agriculture Board

Wasp-Mote.Status : ON
1. Communication.Module.initi()
2. [Setting Frequency Range]
7

Communication_Module.Frequency (Status)
3. [Setting API of Communication Module]
Comm.Module.API(Status)
4. [Setting Communication Module Status]
Comm.Status : ON
5. [Setting Destination Address in Communication Module]
Mac address of Receiver
6. [Repeat Step to ]
Start Loop
7. [Setting Temperature Sensor Status to ON]
SensorAgr.TempSensorMode : ON
8. [Storing Value of Temperature Sensor]
Store.Temp (Value)
9. [Conversion of Temperature Sensor Reading in Compatible Format]
Convert.FloattoString(Temp)
10. [Setting Air Humidity Sensor Status to ON]
SensorAgr.AHumidSensorMode : ON
11. [Storing Value of Air Humidity Sensor]
Store.AHumid (Value)
12. [Conversion of Air Humidity Sensor Reading in Compatible Format]
Convert.FloattoString(AHumid)
13. [Setting Soil Moisture Sensor Status to ON]
SensorAgr.SMoistureSensorMode : ON
14. [Storing Value of Soil Moisture Sensor]
Store.SMoisture (Value)
15. [Conversion of Soil Moisture Sensor Reading in Compatible Format]
Convert.FloattoString(SMoiture)
16. [Setting Leaf Wetness Sensor Status to ON]
SensorAgr.LWetnessSensorMode : ON
17. [Storing Value of LWetness Sensor]
Store.LWetness (Value)
18. [Conversion of Leaf Wetness Sensor Reading in Compatible Format]
Convert.FloattoString(LWetness)
19. [Packet Formation]
Msg : (Destination Add, Sender ID ,String Data)
20. [Sending Sensors Data]
Comm_Module.Send(Msg)
21. [Setting Communication Module Status OFF]
22. Comm_Module.Status : OFF
23. [Free Temporary Memory Assigned to hold Data]
Temp.Memory: Free
AHumid.Memory : Free
SMoisture.Memory : Free
8

LWetness.Memory : Free
24. [Setting Temperature Sensor Status to OFF]
SensorAgr.TemperatureSensorMode : OFF
25. [Setting Air Humidity Sensor Status to OFF]
SensorAgr.AHumidSensorMode : OFF
26. [Setting Soil Moisture Sensor Status to OFF]
SensorAgr.SMoistureSensorMode : OFF
27. [Setting Leaf Wetness Sensor Status to OFF]
SensorAgr.LWetnessSensorMode : OFF
28. [Time Interval between Next Reading]
AgrB.Dely(Time)
29. [Exit]
Stop
Graphical User Interface (GUI) design of system is an important measure for nave users
i.e. farmers to understand and efficiently use it. Bearing in mind GUIs importance, special
attention is given to an easily understandable GUI design in SNAIMS. It is designed in Lab
View which is a system design software tool by national instruments [34]. Lab view is used
due to its integration of graphical, text based, and other programming paradigm within a
single programming environment. A schematic diagram of SNAIMS programming carried
out in Lab view is presented in Figure 6.

Figure 6: SNAIMS programming in Lab view [120].

SNAIMS GUI is shown in Figure 7 representing different tabs for navigation among different
parts. It allows user to fine-tune SNAIMS file reading according to the sampling rate of sensor
nodes because sampling rate of sensor nodes is adjusted during experimental setup.

Figure 7: SNAIMS GUI with adjustable delay iteration mode.

This software presents received data in more user friendly way that is textual data
representation as well as graphical data representation. An example of textual data representation
is shown in Figure 8. Textual data from sensor nodes is represented separately with selected
line number and wasp-mote identifiers. Similarly, data is represented collectively as shown in
data file view of Figure 8 where data is represented as starting with name of agriculture board,
temperature, air humidity, soil moisture and lastly leaf wetness senors data.

Figure 8: SNAIMS GUIs textual data representation.

10

Graphical representation is easier to understand than textual data representation.

Therefore, SNAIMS provides a consistent graphical data representation as well. As received
data is in packets format and needed to be separated before its plotting. Hence, the data is
divided into its constituent parts and forwarded to concerned module to plot it separately.
Data is plotted continuously by appending newly arrived data with existing graph as shown
in Figure 9 and Figure 10 for 10 and 15 consecutive

Figure 9: SNAIMS temperature values plotted graph.

Figure10: SNAIMS fifteen temperature values plotted graph.

Temperature readings from different sensor nodes. Separate data plotting of each
individual sensors reading is valuable and beneficial in identifying water deficient locations.
Wasp-mote agriculture boards sensed data is viewed in serial monitor provided with waspmote programming environment. Before realizing SNAIMS, these boards were programmed
using wasp-mote programming environment, an open source software platform, which
supports java based programming. In WSNs scenario, energy is a significant resource and
its efficient utilization is desperately needed. Hence, the boards programming was carried
out by considering energy efficiency. In this regard, sensors and boards work in operational
modes for shorter time interval and then switched to hibernate mode. Likewise, before
transmission data is preprocessed and merge into a single packet. Serial monitor of wasp11

mote Integrated Development Environment (IDE) allows user to view data as received from
boards but does not let it to store or copy to another location. However, data storage is a
major requirement in any decision support system, therefore, Cool Term software [35-37]
was utilized that is capable of storing data automatically in form of a text file.
The decision support system of SNAIMS is programmed in java by using java socket
programming in two different variants. Its server version is running over server PC where
client version run on other PCs connected with server PC through local area network. The
former version takes Cool Term auto saved text file as input and examine it for vulnerable
condition where vulnerable condition means desperate need of water. If water scarcity
locations are identified then alarming unit is activated an alert message is sent via LAN to
other PCs connected to t Server.

DSS Client Side pseudo Code

1. [Open or Start Server Socket]
Ss: Server Socket
2. [Check Server Status]
Check.Sever (Status)
3. [Read Input File in Text Format]
Read. File (file_name.txt)
4. [Repeat Step 5 to 16 ]
Continue until End of File
5. [Start File Reading Line by Line]
String.Read.Line (File_name.txt)
6. [Split Packets into Its Constituent Parts]
Split.Line.(Parts)
7. [Store Temperature value at Temporary Location]
Store.Temperature.Value(Storage Path)
8. [Store Air Humidity value at Temporary Location]
Store.Humidity.Value(Storage Path)
9. [Store Soil Moisture value at Temporary Location]
Store.SoilMoisture.Value(Storage Path)
10. [Store Leaf Wetness value at Temporary Location]
Store.LeafWetness.Value(Storage Path)
11. [Setting of Threshold Values]
Temp.ThresholdValue(Value)
Humid.ThresholdValue(Value)
SMoisture.ThresholdValue(Value)
LWetness.ThresholdValue(Value)
12. [Compare Most Recent Packet Values with Threshold Values]

12

 CurrentValue.TempTemp.ThresholdValues
CurrentValue.HumidHumid.ThresholdValues
CurrentValue.SMoistureSMoisture.ThresholdValues
CurrentValue.LWetness LWetness.ThresholdValues
13. [If Result of Previous Step is False Then Store Values]
Step-12.Result: False Then
Store.Values(Storage Path)
Display.GUI(Temp.Value)
Display.GUI(Humid.Value)
Display.GUI(SMoisture.Value)
Display.GUI(LWetness.Value)
14. [If Result of Step 12 is True Then Postpone Decision Until Next Packet]
Wait. NextInputStream
Store.Temp(Temporary location)
Store.Humid(Temporary location)
Store.SMoisture(Temporary location)
Store.LWetness(Temporary location)
15. [Match Postponed Packet with Coming Packet and Previously Stored Packet ]
Diff1 (Postponed and Coming Packet) and Diff2 (Postponed and Previous)
16. [if Diff1 Greater than Diff2 then Vulnerable Condition]
Diff1> Diff2
17. [True Result of Previous Step is indication of Vulnerable Condition]
Activate Alarming Unit
18. [Send Message Over LAN]
Msg.Send(LAN)
19. [Store Values]
Store.Temp(Location)
Store.Humid(Location)
Store.SMoisture(Location)
Store.LWetness(Location)
20. [Display Via Continuous Graph]
Display.Temp : True
Display.Humid : True
Display.SMoisture : True
Display.LWetness : True

13

21. [False Result of Step 12is indication of Noisy Data]

Evade.Posponed.Packet()
Remove.Posponed.Packet()
Display.Posponed.Packet : False
22. [if not end of File ]
Go to Step 4
23. [Otherwise]
Exit

Server Side Pseudo code

1. [Open or Start Client Socket]
cc: Client Socket
2. [Check Client Status]
Print Client Status Info
3. [Set IP Address of Server in Client]
C : Socket(192.168.155.86)
4. [Set Port of Server in Client]
C: Socket(8080)
5. [Check Connection Setting]
Connection.Status (Checked)
6. [Store Received Message from Server Module]
Store.Msg(Location)
7. [Display Message Regarding Vulnerable Condition with Location Info]
Msg.show(Vulnerable Condition)
8. [Activation of Alarming Unit if Attached]
Activate.Alarm (Unit)
9. [Stop]
Exit
SNAIMS Design
SNAIMS monitors various environmental parameters and soil properties hourly and on
daily basis during a week. These parameters play a vital role in the design of an efficient
irrigation management system. In order to investigate their effects on irrigation schedule of
oranges orchards, Wasp-mote agriculture boards are deployed at three different locations.
In first phase of our experimental setup, nodes were placed in close vicinity such that every
node was able to communicate directly with gateway. But later on nodes were placed such
that one of them is able to communicate directly with gateway while remaining two nodes
were out of wireless communication range of Xbee module. These nodes communicate
with gateway using multi hop communication. For this purpose, sensor boards placed in
closed proximity of gateway is programmed in a manner to work as a relay node, Node3
in our experimental setup functions as a relay node. Node1 and Node2 broadcast their
14

packets by appending node3 Mac address to its header and destination address as gateway.
When Node3 receives a packet from one of these nodes it examines it and finds destination
address gateway. Then it resends it to destination without modification. The design of
our experimental setup is shown in Figure 11 where dotted arrows represent wireless
communication and solid arrows represent wired communication through a defined USB
port. The wasp-mote gateway communicates serially with computer through a defined USB
port, port-6 in our experimental study. Every node senses temperature, air humidity, leaf
wetness and soil moisture after a defined interval of time that is 30 minutes. The wireless
communication of Xbee module is in a range of about 500 meters. Node3 is placed at about
350 meters from gateway and is able to communicate directly with it. Other nodes are
placed at a distance of about 300 meters from Node3. as shown in Figure 13.

DS
Noise
Detection
Message
over

Alarm
unit

Automatic
Plotting at
Figure 11: WSN for irrigation management system.

15

The distance between nodes and gateway is kept low so that packet lose intensity is
minimized. The gateway module acts as a bridge between sensor nodes and PC. The collected
data is received in a PC via a defined serial port that is COM6 in our case. SNAIMS checks
the received data for any vulnerable condition and informs farmers if the soil moisture,
temperature, air humidity values are below their defined frequency threshold values i.e.
500HZ for soil moisture in our case.

SNAIMS Experimental Setup and Working Mechanism

Soil Moisture Values

In order to check and validate SNAIMS performance and efficiency, wasp-mote

agriculture boards are deployed at three different locations in oranges orchard as shown in
Figure 11. The distance between boards and gateway is adjusted according to their wireless
communication ranges. Four sensors are integrated with each board that is temperature,
air humidity, leaf wetness and soil moisture sensing. Figure 12 presents the internal
structure and working mechanism of SNAIMS. Initially, desired data is collected by sensors
and pass it to agriculture board. Agriculture board creates a packet from it by appending
the destination Mac address and passes it to its Xbee802.15.4 module. The packets are
forwarded to Xbee802.15.4 module of gateway that is connected serially to a computer or
laptop. SNAIMS investigates every packet and compare it against the defined threshold
values. It activates alarming unit, if parameters values are below the threshold values.
However, before taking decision of activation alarming unit, it investigates whether the
vulnerable condition is correct or not. For this purpose, the decisions are delayed until next
data packet is received. This mechanism is carried out to detect noisy data that is likely
to occur in WSNs scenario. Noisy data must be prohibited from storing in data and DSS
should be smart to ignore noisy data because decisions made over noisy data are critical.
In this regards, SNAIMS compares most recent data packet with previously stored packets
but in vulnerable situations this packet is stored in temporary location and waits for the
next packet from similar sensor node. Upon receiving new packet, comparison is made
among three different packets (that is packet received at time T1, T2, and T3). Now if the
data is found noisy, the difference between these three packets are greater than defined
threshold, then packets received at time T2 is discarded while other two data packed are
stored permanently. On the other hand, if the vulnerable packet is found correct then it is
stored and alarming unit is activated. Similarly, a message is send via LAN to client PCs
so that necessary measures are taken. It enables farmers to check the orchard conditions
remotely and fulfill plants water needs according to their requirements. This scenario is
depicted in Figure 12 and Figure 13. The dotted circle in Figure 12 represents noisy data
which is detected and removed by SNAIMS as shown in Figure 13.

9:00:00
9:10:00 Soil 9:20:00
Moisture
Soil Moisture
Soil Moisture
Soil Moisture
Soil Moisture
Soil Moisture
Soil Moisture
Soil Moisture
Node1,
Node1, Node1,
9:30:00
9:40:00
9:50:00
Node1, Node1, Node1, Node1,
Node1,
9:40:00, 9:50:00, 10:10:00,
9:00:00,10:00:00
9:10:00,
500
9:20:00,
500 10:10:00
500
9:30:00,
499.5
10:20:00, 499
10:20:00
499.43 499.31 499.12

Time

Soil Moisture
Node1,
10:00:00, 5

Figure 12: Example scenario of noisy data depicted by dotted circle.

16

Soil Moisture Readings

9:00:00
9:10:00
9:20:00
9:30:00
Soil Moisture
Soil Moisture
Soil Moisture
Node1, Node1, Node1,
9:40:00
9:50:00
10:10:00
10:20:00
9:00:00, 500
9:10:00, 500
9:20:00, 500
Soil Moisture
Soil Moisture
Node1,
Node1,Soil Moisture
9:30:00,
Node1,
9:40:00, Soil
Moisture
499.5
9:50:00,
499.43 Soil
Node1,
Moisture
499.31
10:10:00,
Node1,
499.12499
10:20:00,

Receing Time
Figure 13: Example scenario after noise removal.

Figure 14: SNAIMS processing block diagram.

17

Environmental parameters are important measures to design and implement an

efficient irrigation management system such as SNAIMS. SNAIMS investigates the effects
of temperature and humidity on irrigation schedule of oranges orchard during four
different seasons in Pakistan. Figure 15 shows deployment of wasp-mote agriculture
pro-board. The Sensiron temperature and air humidity sensor is integrated with it as
pointed out by red circle in Figure 13. These sensors are capable of acquiring data
after a defined interval of time and transfer it to board for necessary processing. The
processing unit combines it into one packet and forwards it to its XBee module for
wireless transmission. In order to prolong agriculture boards batteries lifetime, before
transmission sensed data is transferred to the processing unit. It accommodates different
sensors data (temperature, humidity, soil moisture, and leaf wetness in our case) into a
single packet which is then transmitted.

Figure 15: Wasp-mote agriculture board deployment in orange orchard

Leaf wetness sensors are utilized in this study and are placed over the oranges plant
leaf as shown in Figure.-16. It measures the humidity condensed on leaf surface that is
inversely proportional to its output voltage. It is connected with board through a wire as
shown in Figure.-14. Its higher values than the corresponding thresh-hold value means that
the weather is dry and the irrigation schedule must be adjusted accordingly. In SNAIMS,
leaf wetness sensor values are combined with sensiron and soil moisture sensors data and
passed to XBee module for transmitting it to the base station. In order to save energy, we
have programmed these sensors to be ON for small duration of time to sense and pass their
data to processing unit.

18

Figure16: Leaf wetness sensor deployment in orange orchard.

In SNAIMS, irrigation schedule is primarily based on soil moisture sensors data

but temperature, air humidity and leaf wetness sensors data are used to make it more
precise and accurate. The soil moisture sensors are deployed at depth level of 30, 40 and
50 cm underground and 12cm away from the stem of oranges plant as shown in Figure
4-15 and Figure 4-16. Different levels deployment is performed to make the irrigation
schedule more accurate and precise. In order to save energy, soil moisture sensor is
switched off after passing data to agriculture board. The plants are in desperate need of
water if the S3 (Soil Moisture Sensor deployed at the deepest level) value is less than a
defined threshold.

Figure17: Soil moisture sensor deployment near orange plant.

19

30cm
40cm
50cm
Figure18: Soil Moisture sensor deployment at different levels.

Interference among nodes is one of the major issues associated with WSNs and needed to
be handled carefully. Interference at a particular node may be due to an edge that is going
to transmit its data or the node is within the communication range of another node [26].
Interference of a node is directly proportional to the number of nodes in its transmission
range. In proposed experimental setup, interference was avoided by introducing a simplest
mechanism. Sensor nodes were deployed at oranges orchard as shown in Figure 4-11 where
only node3 is capable of directly communicating with base station and no interference at
base station side. The distance between node1 and node2 is greater than 500m and hence no
interference because of their limited transmission rages i.e. 500m. Interference was occurring
at node3 when it was receiving packets from node1 and node2 simultaneously. In order to
reduce interference various algorithms were considered but were not realized in proposed
experiment setup due to their associated complexities. Considering the limited number of
sensor nodes in this study, a simplest interference reduction strategy was implemented.
Interference may occur if the packets from node1 and node2 received at the same time. To
avoid this situation, node1 were scheduled to sends its packets at least five minutes before
node2. In order to make it more accurate and precise, different delay timers were introduced
at node1 and node2 thus reducing the chances of simultaneous packets receiving at node3.
In proposed system, nodes sense their defined parameters after 30 minutes and send it.
So if node1 sends packet at time 10:00:00 then node2 must send packet at 10:05:00 and
thus avoiding interference creating situation. However, this mechanism is applicable in
situations where small numbers of nodes were deployed at experimental fields. Similarly,
different delay timers may also reduce interference up to some extent but not completely
avoids it. An example scenario is node1 sending packets after 30 minutes whereas node2
sends packets after 29minutes and 30seconds, an applicable situation in the proposed
experimental setup.
20

Another issue associated with WSNs is the crops canopy effects over the sensor nodes
communication range. In start of this experiment study the effects of plants canopy
were not considered and nodes were deployed at random locations without considering
line of sight importance. But we were surprised by the receiving packets at the terminal
that was from sensor node3 that was directly communicating with base station. The
distance was same but the only reason of not getting data packets from other nodes
were the unavailability of direct line of sight communication. This experiment inspires
us to check the wasp mote boards communication range in two different phenomenon
i.e. direct line-of-sights and in the presence of plants canopy. These experiments were
fruitful because the communication ranges of these nodes were confirmed. According
to our experiments, wasp mote boards have transmission range of 480m in case where
there is possibility of direct line of sight communication but in presence of obstacles,
oranges plants canopy, their transmission range is about 290m. In light of these two
experiments, wasp mote boards were placed at locations in oranges orchard where direct
line of sight is possible. Therefore, node1 and node2 were redeployed at most suitable
locations.

Results and Evaluation

Wasp-mote agriculture boards transmit data to a central location via gateway which is
serially connected to a computer or laptop as shown in Figure.4-11. For perceiving data,
wasp-motes serial monitor is used but it did not allow data storage or copying. Therefore,
hypo terminal, cool term or other softwares is needed to be used for perceiving its data.
These are simple and useful for researcher or experts but very difficult for a nave user,
farmer or orchard owner, to understand data and use it. In order to make SNAIMS user
friendly, we have proposed and developed software solution that is not only shows data to
its user in more easily understandable format but automatically plots it as well. Initially,
SNAIMS divided packets into its constituent parts and plots every parameter in a separate
chart. Figure 4-17 represents the soil moisture sensor data received from different sensors.
The abrupt changes in data are due to the mal functioning of soil moisture sensor or packet
loss. SNAIMS sampling rate is adjusted according to the sensor nodes sampling rate and is
provided on main page.

Figure19: Soil Moisture sensor data plotting by SNAIMS

21

Figure 18-20 show the plotting of temperature, air humidity and leaf wetness sensors
reading successively. In our experimental setup, Wasp-mote boards send data to a personal
computer via gateway. The data is stored automatically using Cool term software [28] in
a file that acts as a data source for SNAIMS decision support system. It reads every data
entry and start plotting it. It also analyzes each and every entry against the parameters
under consideration threshold values. Normally, it plots data and saves data but if water
deficient condition is detected then it activates the alarming unit connected to it and sends
a message over LAN. In addition to handling irrigation intelligently, manual procedure is
also supported in SNAIMS. Users view and monitor data regularly and if values in graph are
below the defined threshold value, then it is an indication of water deficiency and irrigation
activity is needed to be started immediately. This timely information is not only useful for
plants growth but also saves valuable resources.

Figure 20: Temperature sensor data plotting by SNAIMS.

Figure 21: Air humidity sensor data plotting by SNAIMS.

22

Figure 22: Leaf wetness sensor data plotting by SNAIMS.

The irrigation schedules of oranges orchards and other crops are affected by heavy
rains. It may delays the irrigation activity from three to thirty days depending on crops
water requirements. Hence the irrigation schedules,whether technology based or tradition
mechanisms, must consider this fact. The proposed irrigation management system is a
need based irrigation mechanism and suggests irrigation activity accordig to oranges plants
needs. Therefore, separate case study to be carried out in rainy season is not neede. Similarly,
seasons depending case studies were not investigated due to the fact that SNAIMS suggest
irrigation activity according to pl and DropsMaking the Best use of Water for Agriculture.

References
1. Food and Agriculture Organization of the United Stations (FAO) (2002). Crops andDropsMaking the Best use of Water for
Agriculture. FAO, ROME, Italy.
2. Dursun M and Ozden S (2011) A wireless application of drip irrigation automation supported by soil moisture sensors. Scientific
Research and Essays 6: 15731582.
3. Martnez-Gaitan RB, Gallardo C, Gimnez M, Fernndez C, Thom pson MD (2007) Identification of irrigation and n management
practices that contribute to nitrate leaching loss from n intensive vegetable production system by use of acomprehensive
survey. Agriculture Water Management 89: 261-274.
4. Pardossi A, Incrocci L, Incrocci G, Malorgio F, Battista P, et al. (2009) Root zone sensors for irrigation management in
intensive agriculture. Sensors (Basel) 9: 2809-2835.
5. Feng QY, Li QT (2011) Low-jitter slot assignment algorithm for deadline-aware packet transmission in wireless video
surveillance sensor networks. International Journal of Communication Systems 24: 810-827.
6. Khan R, Ali I, Suryani MA, Ahmad M, Zakaray M (2013) Wireless Sensor Networks based Irrigation Management System for
Container Grown Crops in Pakistan. World Applied Sciences Journal 24: 1111-1118.
7. TaegWon J, Shiguo L, Chongdeuk L (2011) Tournament-based congestion control protocol for multimedia streaming in
ubiquitous sensor networks. International Journal of Communciation Systems 24: 1246-1260.
8. Cristina C, Dandan W, Ufuk T, Qiong S (2011) Cross-layer MAC design for location-aware wireless sensor networks.
International Journal of Communication Systems 24: 872-888.
9. Liang D, Zhong S, Ting C, Yilin C (2013) Analysis and modeling of task scheduling in wireless sensor network based on
divisible load theory. International Journal of Communication Systems 27: 721-731.
10. David S, Antonio B, Jaime DC, ngela R, Claudio R, et al. (2011) An Air-Ground Wireless Sensor Network for Crop Monitoring.
Sensors 6: 6088-6108.
11. Davide DP, Giovanni C, Antonio M, Gianfranco M, Luca B, et al. (2010) Wireless Sensor Networks for On-Field Agricultural
Management Process. In: Geoff VM, Yen K T,Wireless Sensor Networks Application-Centric design (2nd ed.). In Tech Publ,
Italy p: 1-18.

23

12. Park DH, Park JW (2011) Wireless sensor network-based greenhouse environment monitoring and automatic control system
for dew condensation prevention. Sensors (Basel) 11: 3640-3651.
13. Ramanathan N, Balzano L, Estrin D, Hansen M, Harmon T, et al (2006) Designing wireless sensor networks as a shared
resource for sustainable development. In International Conference on Information and Communication Technologiesand
Development p: 1-10.
14. Agriculture.http://www.prismaelectronics.eu/site/index.php?option=com_content&view=article&id=237&Itemid=665&lang=en
15. Zennaro M, Yemane A, Farahbakhshzad N, Pehrson B (2007) wireless sensor net-works for water quality management.
Technical report, Royal Institute of Technology, Stockholm p: 1-7.
16. Feng YB, Zhang RB, Gu GD (2007) Applications of Wireless Sensor Networks in Water Saving Irrigation. China Rural Water
and Hydropower 2: 24-26.
17. Lofez Riquelme JA, Soto F, Suardiaz J, et al (2009) Wireless Sensor Networks for Precision Horticulture in Southern Spain.
Computer and Electronic in Agriculture 68: 25-35.
18. Feliciano Cayanan D, Dixon M, Zheng Y (2008) Development of an Automated Irrigation System Using Wireless Technology
and Root Zone Environment Sensors. Acta Hort. (ISHS) 797:167-172
19. Anurag D, Siuli Roy,Somprakash Bandyopadhyay (2007) Agro-sense:precision agriculture using sensor-based wireless mesh
networks. Indian Institute of Management Calcutta p: 1-5.
20. Zuo X, Gao W, Zhang G, Zhao J, Zhu Y, et al. (2011) Design of Environmental Parameters Monitoring System for Watermelon
Seedlings Based on Wireless Sensor Networks. Applied mathematics and information sciences 5: 243s-250s.
21. Singh I, Bansal M (2011) Monitoring Water Level in Agriculture Using Sensor Networks. International Journal of Soft Computing
and Engineering 1: 202-204.
22. Ke S, Ding M, Li L, Niu Q, Huang D (2012) Grafting watermelon seedling production management system based on process
control strategy. Journal of Shanghai Jiaotong University (Science) 17: 129-134.
23. Mancuso M, Bustaffa F (2006) A wireless sensors network for monitoring environmental variables in a tomato greenhouse. In
IEEE International Workshop on Factory Communication Systems p: 107-110.
24. Burrell J, Brooke T, Beckwith R (2004) Vineyard computing: Sensor networks in agricultural production. Pervasive Computing,
IEEE 3: 38-45.
25. Kotamki N, Thessler S, Koskiaho J, Hannukkala AO, Huitu H, et al. (2009) Wireless in-situ sensor network for agriculture and
water monitoring on a river basin scale in southern Finland: Evaluation from a data users perspective. Sensors 9: 2862-2883.
26. Santiago, Jimenez, Pablo Lozada, Cristhy Jimenez Alejandra Jimenez (2012) Wireless Sensors Network in the Efficient
Management of Greenhouse Crops.In : Proceeding of 2012 Ninth International Conference on Information Technology-New
Generations, Las Vegas p: 680-685.
27. Ahonen T, Virrankoski R, Elmusrati M (2008) Greenhouse moonitoring with wirelesssensor network. In Mechtronic and
Embedded Systems and Applications p: 403-408.
28. Yang X, Zhou Y, Wang L,Guo X, Zhang Q (2007) A wireless solution for greenhouse monitoring and control system based on
ZigBeetechnology. Journal of Zhejiang University-Science A 8:1584-1587.
29. Hemming J,TuijlBA,Pardossi A, Incrocci L, Marzialetti P, et al. (2008) Sensors and wireless sensor networks for irrigation
management under deficit conditions (FLOW-AID). In: Proceedings of the AgEng 2008International Conferuence on
Agricultural Engineering 19: 23-25.
30. Mehmet C, Suat I, Xin D (2012) Autonomous precision agriculture through integration of wireless underground sensor networks
with center pivot irrigation systems. Ad Hoc Networks p: 1-28.
31. Yoder RE, Wilkerson JB, Odhiambo LO, Miranda FR (2005) An autonomous controller for site-specific management of fixed
irrigation systems. Computers and Electronics in Agriculture 48:183-197.
32. Pierce F, Elliott T, Evans R, Chvez J (2011) A Remote Irrigation Monitoring and Control System for continuous move systems
Part A: description and development. Precision Agriculture 11:1-10.
33. Evans RG, Iversen WM, Kim Y (2009) Evaluation of closed-loop site specific. Journal of Irrigation and Drainage Engineering
135-25-31.
34. NIC (2014)NI Lab VIEW-Improving the Productivity of Engineers and Scientists-National Instruments.
35. Bacci L, Battista P, Rapi B, Sabatini F, Checcacci E (2003) Irrigation control of container crops by means of tensiometers. In
International Symposium on Managing Greenhouse Crops in Saline Environment 609: 467-474.
36. Evans RG, Iversen WM, Kim Y (2008) Remote sensing and control of an irrigation system using a distributed wireless sensor
network. IEEE Transaction on Instrumentation and Measurement 57: 1379-1387
37. Libelium(2012) Wireless Sensor Networks-ZigBee-Mesh Networks-Bluetooth.

24