10048-22313-1-PB

An Intelligent Irrigation System Based on Wireless Sensor Network and Fuzzy Control

Liai Gao

Agricultural University of Hebei, College of Mechanical and Electrical Engineering, Baoding, China Email: [email protected]

Meng Zhang

China Agricultural University, College of Information and Electrical Engineering, Beijing, China Email: [email protected]

Geng Chen

Agricultural University of Hebei, College of mechanical and electrical engineering, Baoding, China Email: [email protected];

Abstract—In order to resolve the problems which include loss of soil fertility and waste of water resource in agriculture production, we design an intelligent irrigation system based on wireless sensor networks and fuzzy control. The system mainly consists of wireless sensor networks and the monitoring center. All of the nodes in Monitoring area use solar power, collect the information of soil moisture, together with the growth information of different crops in different periods. Soil moisture content deviation and the rate of change of deviation are taken as input variables of fuzzy controller, and the fuzzy control regular database is established for the fuzzy irrigation control system. The monitoring center receives the data transmission from wireless sensor network node, and output information of irrigation water demands to the relay via a wireless sensor network to control opening and closing time of the valve in crop areas. The experimental results show that the system has a stable and reliable data transmission, which achieve real-time monitoring of soil on crop growth, give a right amount of irrigation based on crops growth information, which has broad application prospects. Index Terms—water-saving irrigation, wireless sensor network, fuzzy control strategy, sensor

I. INTRODUCTION

With the rapid development of agriculture in China, the water consumption was increasing. As the shortage and waste of the low utilization of Chinese agricultural water resources were being existence, the implementations of water-saving irrigation were more and more crucial [1]. Soil moisture content is a prerequisite for the crop growth, while excessive soil moisture would cause the rot of crops’ roots, took away a lot of fertilizer which will cause water pollution. With the development of computer technology and sensor technology, monitoring and control of soil moisture content had made great progress [2], but there remained two main problems: First, most of irrigation control system worked in a wired manner, using serial bus and

field bus technology, therefore it had a complex wiring, installation and maintenance costs. Second, the crop water requirement was a physical quantity with various environmental factors, with a strong coupling and complexity, it was difficult to establish a precise mathematical model, and fuzzy control has good robustness, dynamic response, so it is very suitable for application in irrigation systems and does not depend on accurate mathematical model.

Experts and scholars at home and abroad have applied wireless sensor networks and fuzzy control technology in the irrigation system separately. Literature [3] proposed a soil moisture detection system based on ZigBee wireless network, and all the references only stated monitoring soil moisture content and had no control function; Literature [4] used WSN based on ZigBee technology in the precision – farming, which had a detailed analysis of the deployment of wireless sensor nodes in soil environment and the construction of the gateway and a base station node; Literature [5] set up a small farmland data acquisition platform using ZigBee network, and obtained information of the solar energy, wind and current; Literature [6] achieved a remote monitoring of irrigation system through the distributed WSN and GPRS; Literature[7]put the fuzzy control to the greenhouse monitoring system; Literature [8] proposed a soil moisture monitoring system based on ZigBee, through controlling solenoid valve to control soil moisture rate in irrigation area, but need the support of electric.

In summary, combined with the advantages of wireless sensor networks and fuzzy control technologies, an intelligent irrigation system was designed. First of all, ZigBee wireless network technology was used in this design. Because it had low power consumption, low cost, free wireless communication frequency characteristics, which could replace the wired connections in the traditional system. With solar cell, solar energy was collected and stored in lithium battery to provide power supply for the system. Secondly, the environment factor

1080 JOURNAL OF NETWORKS, VOL. 8, NO. 5, MAY 2013

© 2013 ACADEMY PUBLISHERdoi:10.4304/jnw.8.5.1080-1087

of influence irrigation volume was analyzed, a reasonable irrigation methods for crop water requirement and soil moisture information was selected , as we all known, the soil water environment has the characteristics of big inertia, non-linearity and strong coupling, in this design we introduced fuzzy intelligent control technology to analysis and processing the soil moisture. Adopt MATALB for the generation of the fuzzy controller and the fuzzy control table, using Simulink for the simulation. On the basis of the LabVIEW virtual instrument development platform, a software program for intelligent irrigation monitoring system was designed. Modular design was adopted in this design; it could realize the signal collection for real time monitoring, data display, data processing, control signal output, and other functions. The whole software system operation had friendly interface, convenient development and maintenance. In a word, the system use fuzzy control technology, sensor technology, wireless sensor network (WSN) technology, and agricultural irrigation technology, which can realize the intelligent agricultural irrigation and increase the efficiency of agricultural water.

II. SYSTEM ARCHITECTURE

The whole monitoring system had two parts: a wireless sensor network and monitoring center. Sensor nodes, the controller node, soil moisture sensors, irrigation pipe, spray irrigation and irrigation control valve were deployed in crop-growing regions, the framework of the monitoring system was shown in Fig. 1. ZigBee network was adopted in mesh network topology. In order to meet the network coverage and reduce the node energy consumption and cost at the same time, we selected a small amount of sensor nodes as routers, to complete the data gathering and routing data from other equipment to the coordinator. And most of the sensor nodes act as terminal devices, only collect data and sent to the router or near the coordinator; A control panel was extend to the sensor nodes to make up for controller node, data acquisition can be proceeding at ordinary times, control valve can be opened to realize the irrigation when receiving irrigation command.

Wireless sensor network consisted of sensor nodes, routing nodes and coordinator node, distributed in all regions of the monitoring area [9, 10]. All nodes were powered by solar energy. Nodes used modular design, the three kinds of nodes used common core modules, and different nodes with different extension modules. The temperature and humidity sensor collected temperature and humidity information; routing nodes was responsible for routing communication and forwarding data; the coordinator node received data from routing node and sent it to the host computer monitor center through RS232 serial bus. The monitoring center could record real-time soil moisture content uploading from all nodes, calculate crop irrigation water requirement according to the plant physiology characteristic in different growth period, and output result to relay by wireless sensor network, control opening and closing time of valve, so as

to realize the remote automatic adjustment and control for irrigation.

Figure 1. Framework of the system

In order to solve the node energy supply problem, we proposed a supply system based on solar energy [11], which is composed by solar modules, Energy management controller, and lithium battery.

The solar module can not only charge the lithium battery, but also supply power to the sensor node; similarly, lithium battery can supply power to the sensor node. The microprocessor of the energy management controller collect voltage of solar panels and lithium battery, then transmit the feedback information to the controlling circuit. The controlling circuit executive power supply and charging scheme according to the information of the microprocessor effectively solve the problem of battery charged frequently. Under the condition of continuous power supply, the priority for power supply is solar panels, lithium batteries. If the lithium battery is not saturated, the solar battery charge for lithium batteries; when sunshine is insufficient or no sunshine, added lithium battery power supply or power supply separately.

III. WIRELESS SENSOR NETWORK NODES

There were three kinds of wireless sensor network nodes: sensor nodes, route nodes and the coordinator node. All of the nodes made cc2530 as the core, matched by the different extension modules. The CC2530 is a true System-on-Chip (SoC) solution specifically tailored for IEEE 802.15.4 and ZigBee™ applications. It enables ZigBee™ nodes to be built with very low total bill-of-material costs. The CC2530 combines the excellent performance of the leading RF transceiver with an industry-standard enhanced 8051 MCU, 32/64/128/256 KB flash memory, 32/64/128/256 KB RAM and many other powerful features. The CC2530 has different running mode, makes it particularly suitable for low power requirements of the system. Node hardware structure diagram is shown in Fig. 2.

JOURNAL OF NETWORKS, VOL. 8, NO. 5, MAY 2013 1081

© 2013 ACADEMY PUBLISHER

microprocessor8051 RF module

CC2530

soil humiditysensor

solar cells

solenoid valvedriver circuit

reset circuit

RTC

energy managementcontroller lithium battery

Figure 2 . The hardware structure diagram of node

A. Sensors

The type of soil moisture sensor is TDR-3A, which is produce by Jinjiang Sunshine Technology Co., Ltd .The sensor is an ideal instrument, which can measure soil temperature and humidity and has the features of sealed, waterproof, high precision. The main performance indexes of the sensor are shown in table I.

TABLE I.SENSOR TECHNOLOGY PARAMETERS

Measured parameters

Performance index

humidity

range: 0～100% accuracy: ±2% measuring field: the cylinder which diameter is 3 cm, length is 6 cm around the probe Working voltage : 12V～24V DC Working circuit: 50～70mA, output: 4~20MA

B. Output Control Module

Irrigation actuator output control signal to the relay, thus control irrigation time of the valve open. This paper adopts NPN triode drives relay, the driving circuit is shown in Fig. 3.

VCC(12V)

relay

control signal

R1

R2

8050

4007

Figure 3. The driving circuit of the relay

The type of Solenoid valve is 100-DVF which is produced by Rain Bird Corp. It is normally in closed solenoid valve. The function is opening and closing the water of pipeline according to the control signal sending by the device. The valve is equipped with a TBOS control module, valve start-stop is controlled by positive and

negative pulse, and it is very suitable for the occasions without alternating current.

IV. FUZZY CONTROL ALGORITHM

The soil environment is a large inertia, nonlinear and time delay system, it is very difficult to establish a precise mathematical model, and because of the unified, variability and complexity in the greenhouse field conditions, it is difficult to achieve precise control if we the traditional control method [12]. we select fuzzy control theory of intelligent control method .Fuzzy control theory does not need to establish accurate mathematical model of controlled object, robust, it is suitable for the lag, nonlinear and time-varying system, so it is suitable for using fuzzy control strategy to realize the controlling of the monitoring system.

Whether to construct fuzzy controller reasonably is related to the precision of fuzzy control system, the structure of fuzzy controller is shown in Fig. 4.

fuzzyreasoningfuzzification defuzzi-

cationoutput

E

EC

knowledgebase

humidity

Figure 4. Structure of fuzzy logic controller

E is soil humidity deviation; EC is the rate of change of deviation over time.

The reasoning process is divided into the following steps [13]: first the continuous input, output is converted to a fuzzy subset, its domain is defined, and fuzzy table is set up according to the actual change range of the input output. Establish the knowledge base through the knowledge and experience of experts, and form the fuzzy control rule, use fuzzy table and fuzzy control rule table, the fuzzy control fuzzy control, into the final amount. Calculate the amount of the corresponding fuzzy control; finally, make fuzzy control amount defuzzification for transforming into ultimate control parameters.

(1) The Fuzzification of Precise Variable For the fuzzification operation of precise variable

usually include selecting fuzzy variables, quantifying input data and determining the membership function. Select E as the input variables, the language variables of corresponding deviation are divided into 5 grades, which can reflect the size: NB = negative big, NS =negative small, Z = zero, PS = positive small, PB = positive big. Assume that changes in humidity around 20%, the basic domain of deviation humidity is [ 20, 20]− + , according to the formula:

2 1n km+ = (1) where n is the number of elements; m is classification; the value of k is 2.

So the fuzzy domain is (-4, -3, -2, -1, 0, 1, 2, 3, 4); The basic domain of another input variable ECH is [-5,

+5], fuzzy domain (-4, -3, -2, -1, 0, 1, 2, 3, 4) .



The results of the fuzzy control is the opening time controlled by solenoid valve, we use U as an output variables, fuzzy linguistic variables Z (zero), DS (short-term), ZS (middle-term), CS (Large-time). Set the basic domain is [0,30] , the fuzzy domain (0, 1, 2, 3).

According to the formula / iKET n Z= (2) The quantization factor of respective variable can be

determined. So, the humidity quantitative factor

1 1/ 4 / 20 0.2KET n Z= = = ; Humidity deviation quantization factor

2 2/ 4 / 5 0.8KET n Z= = = ; The output control quantization factor

3 1/ 3 / 30 0.1KET n Z= = = ;

Through above formula, elements of iA , iB , iC in the fuzzy domain can be calculated which is correspond to the accurate digital quantity ieT .

1 1i TA KET e= × (3)

2 2i TB KET e= × (4)

3 3i TC KET e= × (5)

By this method, each variable domain ,fuzzy subset, quantitative grade, fuzzy domain and the quantification factor is completed, the results is shown in table II.

TABLE II. EACH VARIABLE DOMAIN, FUZZY SUBSET, QUANTITATIVE GRADE, FUZZY DOMAIN, QUANTIFICATION FACTOR

variable domain fuzzy subset quantitative grade quantitative grade quantification factor

E [-20%,+20%] Ai (NB, NS, Z, PS, PL )

(- 4, - 3, - 2, - 1, 0, 1, 2, 3, 4) 9 0.2

EC [-5,+5] Bi (NB, NS, Z, PS, PL )

(- 4, - 3, - 2, - 1, 0, 1, 2, 3, 4) 9 0.8

u [0,30] Ci (Z, DS, ZS, CS ) (0, 1, 2, 3) 4 0.1

(2) Determine the membership function

This study use triangle membership function, Fuzzy input variables membership assignment is shown in tableⅢ.

TABLE III. FUZZY INPUT VARIABLES MEMBERSHIP ASSIGNMENT TABLE

fuzzy

variables E/EC

-4 -3 -2 -1 0 1 2 3 4 NB 1 0.5 0 0 0 0 0 0 0 NS 0 0 0.5 1 0 0 0 0 0 Z 0 0 0 0 1 0 0 0 0

PS 0 0 0 0 0 1 0.5 0 0 PB 0 0 0 0 0 0 0 0.5 1

Fuzzy input variables membership assignment is

shown in table IV. TABLE IV. FUZZY OUTPUT VARIABLES MEMBERSHIP ASSIGNMENT

TABLE

Fuzzy variable U 0 1 2 3

Z 1.0 0 0 0 DS 0 1.0 0 0 ZS 0 0 1.0 0 CS 0 0 0 1.0

(3) The fuzzy control rules Formulate rules of fuzzy control principle is make the

static and dynamic characteristics of system output response for the best [14, 15]. That is to say when the output deviation of the system is bigger, we need to select control variable quickly for the purpose of reducing or even eliminating bias; when the system output deviation is small, we must take the system stabilization as the goal when selecting control variable, at the same time pay attention to overshoot phenomenon. For the temperature factor, the fuzzy control rules are as follows:

If E is iA and EC is iB THENU is iC (6)

(4) Fuzzy reasoning The key of fuzzy controller is fuzzy reasoning; it refers

to the process of the simulation of the human brain thinking based on fuzzy concept. In this study, Mamdani fuzzy model is selected as reasoning method, a corresponding membership value is obtained from the current quantized input variables, and then from the fuzzy control rules, we can get output control quantity of fuzzy variable. The fuzzy control rules table is shown in table 4.

TABLE V. THE FUZZY CONTROL RULES TABLE

U E NB NS Z PS PB

EC

NB Z Z Z Z Z NS DS DS Z Z Z Z DS ZS Z Z Z

PS ZS ZS ZS Z Z PB CS CS CS Z Z

(5)Defuzzification The output of fuzzy controller is a fuzzy set, in general,

we often require a relatively precise control signal as the object [16]. Defuzzification is to construct mapping from a set of fuzzy to the normal set, convert the fuzzy output into a precise controlling quantity, it is also called cleanness. Some method are often used, such as the median method, gravity method, gravity method is used in this study, the expression for the

1 1( ) / ( )

n n

i i i ii i

z A Aμ μ μ= =

= (7)

Among them, n is the number of fuzzy variables; iμ fuzzy variable; ( )iA μ fuzzy variable corresponding to



membership. After Defuzzification, we can obtain the output of the exact amount.

B. System Simulation This article use Simulink for the fuzzy controller

simulation. First fuzzy controller is designed based on the theoretical design of the front; the model is shown in Fig.5.

Figure 5. The model of fuzzy control Fuzzy rule base is designed after determining the

membership function, thereby generating fuzzy rule table, as shown in Fig. 6.

Figure 6 .Fuzzy rule table Based on the fuzzy base, we can get a surface plot of

the fuzzy control, as shown in Fig. 7.

Figure 7. Fuzzy control surface Soil moisture content change is a slow and

complicated process; it has a certain relationship with the surrounding environmental factors, it is almost impossible to establish a precise mathematical model. so we select formula (6) as a reference model.

1 1sin( / )y y t xπΔ = (8) Wherein yΔ is humidity change, t is valve conduction

time after conversion of irrigation quantity,

1y and 1x are determined experimentally by the system. Simulation mathematical model are created in Simulink module in accordance with the principle of the fuzzy controller, fuzzy controller simulation model and the PID controller model is shown in Fig. 8.

Figure 8. The simulation model

Set humidity value is 37%, the output curve of the

fuzzy controller simulation and PID controller simulation are shown in Fig. 9. From the simulation graph we can see that the PID control response speed is faster than the fuzzy control, but overshoot ratio is greater; the fuzzy controller system has more stability relatively, it is more suitable for greenhouse irrigation control.

Figure 9 .The simulation curves of PID controller and fuzzy controller

V. SOFTWARE DESIGN

A.Software Diagram

The software design includes the design of the ZigBee node and the monitoring center, the architecture of the overall software system is shown in Fig.10. The development of node is based on the MAC and physical layer offered by the ZigBee protocol stack, mainly relates to the data acquisition and network frame problem [17].

Coordinator node is responsible for configuring the network parameters, initiation network and network maintenance work, reception of data acquisition and transmission threshold of temperature and humidity[18]. The main task of sensor node is data acquisition and transferring, after completing initialization protocol stack and hardware, the terminal node start scanning channel, and then sends the information of joining the network. If confirmed, it began to collect data through the sensors, and then sends to the coordinator node through the



wireless transmission. The flow of the node program is shown in Fig.11.

sensor node coordinator node monitoring platform

acquiredata

startandstopthe

valveform a

network

transmitdata to

hostcomput

er

transmit the

thresholdto

sensor node

displaynetworkstatus

displaythedata

handleand

storedata

Figure 10. Architecture of the overall software system

SRART

Initialization protocalstack and hardware

send device address

collect data

send data via radio

ST ART

Initialization protocalstack and hardware

Start networing

receiving data fromsensor nodes

networing success，form ID

upload databy serial port

join network

wirlessnetwork

coordinator nodes sensor nodes

Figure 11. Flow of the node program

B. Design of the Monitoring Platform

Figure 12.Wireless testing platform

The graphical programming language LabVIEW is used for the design of the monitoring platform [19]. Graphical programming can be more intuitive and effective to complete the test task. Designers judge wireless sensor network state through different buttons, at the same time, the tested results from calculation and analysis of scientific is display through the different

curves and color. The system includes a connection state, data acquisition, data processing and alarm, data storage and other functions. The monitoring platform is shown in Fig. 12.

VI. EXPERIMENTAL RESULT

In order to test the validity of data transmission, we selected an agriculture sightseeing garden located in XuShui, Hebei province as the experimental base. The base had 10 greenhouses with tomato as the main crops. We selected Mar. 5, 2013 as the test date, and that day was in the tomato blossom period. The test system consisted of two sensor nodes and a sink node.

A Wireless Transmission Test

Wireless transmission test is divided into point-to-point and networking communication test. One coordinator and a sensor node are opened when conduct point-to-point communication test, the distance between these two nodes increase from 30 mile to 300 mile, the sensor node send a frame data to the coordinator node at intervals of 1 minute, write down the number of packets which the gateway node receives within one hour (out of a total of 60).Two sensor nodes are opened when conduct the network communication test, the distance still starts from the 60 miles, add the distance between nodes and coordinator node gradually,, until the two nodes are separated by 300 m, the sensor node send a frame data to the coordinator node at intervals of 1 minute, write down the number of packets which the gateway node receives within one hour (out of a total of 60). The number of packets from coordinator node in different methods is shown in table VI.

TABLE VI. THE NUMBER OF PACKETS IN DIFFERENT METHODS

method distance point-to-point networking

30 59 55 60 58 54 90 56 53

120 55 52 150 54 51 180 52 50 210 50 48 240 48 46 270 47 45 300 45 43

From table VI, we can see that the distance of network

transmission is less than point-to-point test, because all nodes in the network send a packet to the coordinator node after networking, resulting in data redundancy and a lost package phenomenon, especially in the distance of more than 200 miles, due to the influence of the obstacle, the packet reception rate is below to 80%.

B．Soil Water Content Monitoring Test

The soil moisture is set in 34% and the probe was embed for the depth of 15 cm The soil moisture is set in 34% and the probe was embed for the depth of 15 cm. When moisture content detected by the sensor is lower



than the set threshold, the control system open the solenoid valve, after 20 minutes (this time can be setting according to different vegetation and soil) closing solenoid valve for water infiltration. Test again after 5 minutes, if the moisture content is still lower than the setting threshold; continue to start the solenoid valve. The humidity test result is shown in table VII.

TABLE VII. THE HUMIDITY TEST RESULTS

Time humidity

Node 1 Node 2 9:58 25.1 24.3 10:08 25.8 24.9 10:29 33.5 32.9 11:35 31.6 30.9 12:12 28.3 28.1 12:58 32.3 32.1 13:27 34.8 34.6 14:15 30.9 30.5 14:48 33.6 33.1 15:22 31.4 31.0 16:06 34.2 33.9 16:54 32.5 32.1

From table VII, we can see that soil moisture content

has significant change after the irrigation for at least 10 minutes, , the system is running well in the later time, the error between the measured soil moisture content and value is around 2%, which indicated the system could regulate greenhouse temperature and humidity value according to the fuzzy control rules. That was to say, when the plants lacked water, the valve could be opened for irrigation. When humidity met the requirement, irrigation would be stopped, and thereby achieved water conservation.

Since node 1 was located in the lower part of the dense foliage of tomato plants, the relative humidity measured by node1 was a little higher than node 2.The measured results basically agreed with the reality, so to achieve the test requirements.

VII. CONCLUSIONS

Aiming at the shortcoming in the irrigation system, this design has combined wireless sensor network with fuzzy control system in the intelligent water-saving irrigation system, realized a remote on-line monitoring and controlling. Nodes in the monitoring area have used devices of solar energy and lithium battery to provide power, which has certain practical significance to solve the problem of limited energy; using the set of multiple hops network protocol to communicate, the nodes have the feature of self-organizing, low power consumption and high reliability. The system has adopts control method of double input and single output fuzzy, used Fuzzy Logic toolbox for modeling and analyzing of the system, and demonstrated the relationship between system input and output, made the control system much scientific. Host computer receives soil water content information collected by the nodes, and transmits

information by setting a threshold of open time of solenoid valve, so as to achieve the goal of water-saving irrigation. The experimental results show that the system can realize automatic real-time monitoring of soil moisture content in crops grow, and combined with crop growth information to irrigate moderately the system. It has not only benefited to crop growth and development, but also avoided the waste of water resources; finally it can achieve better productivity, high efficiency and quality, and water-saving.

ACKNOWLEDGMENT

This work was supported in part by a grant from the high school scientific and technological research guidance project in Hebei province (No.Z2011271) and the science and technology research guidance project in Baoding (No.13ZN022)

REFERENCES

[1] Chen Chao, W.E. Baethgen, W. Enli, Y. Qiang, “Characterizing spatial and temporal variability of crop yield caused by climate and irrigation in the North China Plain”, Theoretical and Applied Climatology, Vol. 106, pp. 365-81, Dec. 2011.

[2] N. Philipcva, O. Nicheva, V. Kazandhiev, M. Chilikcva-Lubcmircva, “A computer program for drip irrigation system design for small plots”, Journal of Theoretical and Applied Mechanics, Vol. 42, pp. 3-18, Dec. 2012

[3] Yunseop K. Evans RG, Iversen W M, “Remote sensing and control of an irrigation system using a distributed wireless sensor network” ,IEEE Trans. on Instrumentation and Measurement, Vol. 57, pp. 1379- 1387, 2008

[4] Jin guangzhao, Peng chenglin, “Zigbee-based soil humidity monitoring system”, Transducer and Micro system Technologies, Vol.27,pp. 92 – 94,2008.

[5] L. pezJ, “A Wireless Sensor Networks for precision horticulture in Southern Spain” ,Computing & electronics in Agriculture Journal, Vol. 68,pp.25-35,2009.

[6] Sun R. M. “wind and water flow as energy supply for small stationary data acquisition platforms”, Computing & Electronics in Agriculture Journal, Vol. 64, pp. 120- 132, 2008.

[7] Christos Goumopoulos, “An Autonomous Wireless Sensor/Actuator Network for Precision Irrigation in Greenhouses, Smart Sensing Technology for Agric. & Environ. Monitor”, Springer-Verlag Berlin Heidelberg, pp. 1-20, 2012.

[8] T. Ahonen, R. Virrankoski M.Elmusrati, “Greenhouse Monitoring with Wireless Sensor Network” , Mechtronic and Embedded Systems and Applications, IEEE Conference Publishing Services, pp. 403-408, 2008

[9] F. Ijaz, A. A. Siddiqui, IM. Byngkwan,“Remote management and control system for LED based plant factory using ZigBee and Internet”, International Conference on Advance Communication Technology, IEEE Conference Publishing Services, pp. 942-946, 2012

[10] N. El RAchkidy, A. Guitton , M. Misson, “Pivot Routing Improves Wireless Sensor Networks Performance”, Journal of Networks, Vol. 7, pp. 962-71, June 2012.

[11] B.D. Vick, L.K. Almas, “Developing Wind and/or Solar Powered Crop Irrigation Systems for the Great Plains”, Applied Engineering in Agriculture, Vol. 27, pp. 235-245, March 2011.



[12] P. Cepeda, P. Ponce, D. Romero, A. Molina, “Fuzzy C-Means Clustering Technique applied for Modeling Parameters of an Intelligent Greenhouse Open Control System”, Electronics, Robots and Automotive Mechanics Conference, IEEE Conference Publishing Services, pp.283-288,2011

[13] F. Harm, “Irrigation fuzzy controller reduces tomato cracking”, International Journal of Advanced Computer Science and Applications, Vol. 2, pp. 106-109, Nov. 2011.

[14] A. Al-Canaan, A.Khoumsi, “Multimedia Web services performance: Analysis and quantification of binary data compression”, Journal of Multimedia,Vol. 6, pp. 447-457, 2011.

[15] J. Segovia, P. Vila,, E. Calle, J.L. Marzo, “Improving the resilience of transport networks to large-scale failures”, Journal of Networks, Vol. 7, pp. 63-72, 2012.

[16] I. Kollia, Y. Kalantidis, K. Rapantzkos, A. Stafylopatis, “Improving Semantic Search in Digital Libraries Using Multimedia Analysis”, Journal of Multimedia, Vol. 7, pp.193-204, April 2012.

[17] Fan Yang, F. W.B. Li, “Active contour projection for multimedia applications”, Journal of Multimedia, Vol. 6, pp. 170-180, April 2011.

[18] Coroama V.C, Hilty L.M, Birtel M,“Effects of Internet-based multiple-site conferences on greenhouse gas emissions”, Telematics and Informatics,Vol.29, pp.362-374. 2012.

[19] Buldu, Ali, “A web-based controlled greenhouse system and monitoring the system's data”, International Review on Computers and Software, vol.6, pp. 856-861, 2011.

Liai Gao received the Bachelor Degree of Engineering in electric engineering and the master degree respectively in Agricultural mechanization engineering from Agriculture University of Hebei, Hebei, Baoding, in2003 and2006.

She is currently with electric Department, the College of mechanical and electrical engineering, Agriculture University of Hebei. Her research

activity mainly focuses on wireless sensor networks, with special interest on the use of wireless sensor networks for intelligent detection and control technology. Now she is a Ph. D Candidate and has published 5 papers in the core journals. Meng Zhang received the Bachelor Degree of Engineering in electric engineering in Agricultural mechanization engineering from Agriculture University of Hebei, Baoding, in 2012. Now she is a graduate student in College of Information and Electrical Engineering, China Agricultural University, Beijing, China. Geng Chen, he is a student in College of mechanical and electrical engineering, Agricultural University of Hebei, Baoding, China



Documents

10048-22313-1-PB