A survey on wireless sensor network infrastructure for ...

69 downloads 1416 Views 658KB Size Report
environment information technology research [1]. Wireless sensor network (WSN) technology has emerged as the technical means to solve the problem.
Computer Standards & Interfaces 35 (2013) 59–64

Contents lists available at SciVerse ScienceDirect

Computer Standards & Interfaces journal homepage: www.elsevier.com/locate/csi

A survey on wireless sensor network infrastructure for agriculture Xiaoqing Yu a,⁎, Pute Wu a, b, c, Wenting Han a, b, c, Zenglin Zhang a, c a b c

Northwest Agriculture and Forestry University, Shaanxi, Yangling, 712100, China National Engineering Research Center for Water Saving Irrigation at Yangling, Institute of Soil and Water Conservation of Chinese Academy of Sciences, Shaanxi, Yangling, 712100, China Research Institute of Water-saving Agriculture of Arid Regions of China, Shaanxi, Yangling, 712100, China

a r t i c l e

i n f o

Article history: Received 26 February 2012 Accepted 11 May 2012 Available online 18 May 2012 Keywords: Hybrid wireless sensor network Information collection Agriculture Wireless underground sensor network Monitoring

a b s t r a c t The hybrid wireless sensor network is a promising application of wireless sensor networking techniques. The main difference between a hybrid WSN and a terrestrial wireless sensor network is the wireless underground sensor network, which communicates in the soil. In this paper, a hybrid wireless sensor network architecture is introduced. The framework to deploy and operate a hybrid WSN is developed. Experiments were conducted using a soil that was 50% sand, 35% silt, and 15% clay; it had a bulk density of 1.5 g/cm3 and a specific density of 2.6 cm − 3. The experiment was conducted for several soil moistures (5, 10, 15, 20 and 25%) and three signal frequencies (433, 868 and 915 MHz). The results show that the radio signal path loss is smallest for low frequency signals and low moisture soils. Furthermore, the node deployment depth affected signal attenuation for the 433 MHz signal. The best node deployment depth for effective transmission in a wireless underground sensor network was determined. © 2012 Elsevier B.V. All rights reserved.

1. Introduction The environmental parameters of soil can be viewed as a type of spatial and three-dimensional network information. Developing methods to gather, process, integrate and apply environmental information is the focus of agricultural environment information technology and contemporary international agricultural science and technology research. Because agricultural regions are scattered and thus the terrains and environmental conditions vary significantly, methods to collect crop growth environment variable information accurately and rapidly are one of the primary problems for agricultural environment information technology research [1]. Wireless sensor network (WSN) technology has emerged as the technical means to solve the problem. A wireless sensor network in an agricultural environment comprises integrated sensors deployed in the area of the farmland. These sensors cooperate with each other to perceive and monitor real-time soil and weather information. The information is processed intelligently by an embedded system. Moreover, the information is transmitted to a diagnosis decision center by a random self-organized wireless communication network, which provides remote monitoring and management of the agricultural environment. Recently, wireless sensor networks have been extensively developed for agricultural environments [2,3]. The networks have been used for irrigation, cultivation, fertilizer management, etc.

⁎ Corresponding author. E-mail address: [email protected] (X. Yu). 0920-5489/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.csi.2012.05.001

Most wireless sensor networks developed for agricultural applications that involve soil monitoring are terrestrial wireless sensor network systems. To avoid transmitting information through the soil, the wireless sensor network is often connected by cables to data access and wireless transceiver devices on the ground. These devices are exposed and thus influence farming activities. In addition, transmission from wireless nodes can be affected by natural geographical and meteorological factors. Wireless underground sensor networks (WUSN) provide a new method for underground monitoring [4–6] and have become a new topic for research in the field of agricultural environment information technology. Wireless underground sensor networks are sensor networks that comprise wireless underground sensor devices, which send and receive functional modules in the soil. These devices are located at specific soil depths and wirelessly transmit data. When the induction module perceives data, the sensor network completes the process of data perception and collection. The sensor network has several merits: strong concealment, ease of deployment, timeliness of data, reliability, and potential for coverage density. In addition to monitoring static parameters of the soil, the wireless underground sensor network can monitor soil motion such as landslides, earthquakes, debris flow, movement of underground ice and volcanic eruptions [3]. Therefore, wireless underground sensor networks have the potential for wide application in fields such as agriculture, military, transportation, structural engineering, and earth science [7,8]. Terrestrial wireless sensor networks in irrigation control systems have been extensively developed and researched. However, the wireless underground sensor network is a new research area without definitive results [9,10]. In a hybrid WSN, a WSN and a WUSN are

60

X. Yu et al. / Computer Standards & Interfaces 35 (2013) 59–64

combined to form a WSN of mixed structure. The application of hybrid WSNs to the monitoring of agriculture information is a novel research topic. The remainder of the paper is organized as follows. We provide an overview of existing wireless sensor network techniques in Section 2. The system architecture for the hybrid wireless sensor network is presented in Section 3. In Section 4, the deployment and connectivity of nodes are discussed. The experimental setup of a WUSN and its analysis are described in Section 5. Finally, the paper is concluded in Section 6. 2. Existing wireless sensor network techniques 2.1. Wireless sensor networks that monitor agricultural environments Recently, wireless sensor networks have been deployed in agricultural environments. Applications include the management of water resources and product storage facilities, the determination of the optimal time for crop harvest, the characterization of crop growth and the prediction of fertilizer requirements. In [11], a wireless sensor network was deployed to monitor the water content, temperature and salinity of soil at a cabbage farm located in Murcia, a semi-arid region of Spain. The wireless sensor network deployed four types of structure nodes: soil, environmental, water and gateway. The software and hardware components of each node were fully described. The management and monitoring of the system were performed by a central processing computer located in the farm's management office. The system was tested in two stages: a laboratory test and a field test. The purpose of the laboratory test was to analyze the functionality of the system devices and to measure the network's performance and energy consumption. The field test assessed the range, robustness and reliability of the system. In [12], the temperatures at various positions in a feed warehouse were monitored using a wireless sensor network. The communication frequency of the sensor nodes was 433 MHz, and the transmission power was 10 mW. The results showed that temperature sensor nodes buried at depths of 25 and 50 cm reliably transmitted temperature signals to the gateway node; between 98.9% and 99.4% of the signals were received by the gateway node. The data obtained from monitoring were used to develop a model of the temperature at different positions in the feed storehouse; the accuracy of the model predictions is between 90.0% and 94.3%. The energy limitations of wireless sensor networks were the focus of [13]. The number of nodes and the characteristics of the regional distribution were used to determine reasonable clusters and the communication's energy model. A theoretical analysis of the optimization of the number of clusters according to the different levels of clusters was performed and experimentally verified. A wireless sensor network was deployed to monitor a greenhouse environment in [14]. The control terminal of the system was designed based on the ARM9 and an embedded Linux operating system, which was used for data receiving, real-time display and data storage. The control terminal communicated with the remote management center using GPRS. The wireless sensor network acquired the greenhouse environment data. The sensor network measured temperature, humidity, CO2 content, light intensity, substrate temperature and the humidity of the greenhouse using 6 channels. A greenhouse wireless monitoring system was designed in [15] using a wireless sensor network based on ZigBee. A dynamic wireless sensor network with a star topology was proposed according to the structure characteristics of the greenhouse. This low cost and low power consumption design shortens the peer-to-peer communication distance by using mobile sink nodes to form a subnet with child nodes based on time and frequency hopping methods. A complex communication network based on the low power radio frequency chip NRF2401A was developed using the method of frame

expansion. The communication algorithms of the sensor, control node and sink node were given. An energy consumption analysis of the network child nodes was conducted using different working states of the sink node. A node system designed for the collection of farmland information using a WSN was presented in [16]. This study combined wireless sensor network nodes and sink nodes with an embedded processor system. The network nodes were distributed regularly over the monitored region and collected soil moisture data. The nodes formed the network and transmitted the information to the sink nodes for dynamic display and storage. Several possible node antenna heights were considered: 0.5, 1.0, 1.5 and 2.0 m. To study the transmission distance of the radio signal in different growth periods, experiments were conducted during three typical growth periods of wheat: seeding, jointing and heading. The relationship between the effective transmission distance of the radio signal in different growth periods of wheat and the optimal antenna height was found. The results of this research provide technical support for the application of wireless sensor networks in agriculture. A wireless sensor network was applied to water-saving irrigation systems in [17]. A two-layer wireless sensor network was designed based on irrigation needs and the characteristics of a fixed pipeline spray irrigation system. An expression for the minimum number of sensors required to cover the field was obtained using the relationship between the sensing radius of each sensor and the range of the shower nozzle. The data transmission method, the network structure and the division of nodes were given. The reliability of the data transmission was ensured using a layered fault diagnosis method.

2.2. Wireless underground sensor networks Wireless underground sensor networks are a new research area. At present, such networks are in the experimental study phase, and no mature products are in the market. Little information has been published on wireless underground sensor networks in agricultural environments. The present work studies the dependence of path loss, bit error rate and maximum transmission distance of the electromagnetic wave on factors such as soil type, volumetric water content of the soil, deployment depth of nodes, internodes distance, the range of frequency, etc. A network system structure for a wireless underground sensor network system designed as an intelligent transportation system for the maintenance of the near surface soil (such as golf courses and football fields) was designed in [13]. The software and hardware systems of the nodes were also designed. The collection nodes used a low performance microcontroller. The receiving nodes on the ground used a high performance microcontroller. However, development and testing of the network system were not carried out. In addition, [13] studied the performance of wireless underground sensor networks that are influenced by the propagation of electromagnetic waves in soil using the underground channel model, electrical characteristics of soil and deployed solutions of wireless underground sensor network nodes. Mathematical simulations were performed using MATLAB for a system given as follows: 400 MHz signals, a sensor deployment depth of 0.5 m, horizontal spacing between sensors of 1 m, conductivity set to 0.1 and a dielectric constant of 10 under. The transmission parameters of electromagnetic waves and energy losses for different volumetric water contents of the soil and different sand and clay soil compositions were studied. In the laboratory of [18], wireless signal attenuation of a ZigBee wireless transceiver module (Soil net) with a 2.44 GHz carrier frequency was researched using soil columns of different soil types and water contents. Experimental results showed that increases in the soil column depth and volumetric water content of the soil increased the signal attenuation. The relationship was expressed using

X. Yu et al. / Computer Standards & Interfaces 35 (2013) 59–64

transmissions are suitable in the WUSN. Soil water content had a significant influence on the performance of the networks. The WUSN must be more robust to changes in water content to be suitable for monitoring soil parameters. The performance of the wireless sensor network was also affected by seasonal environment changes; cross layer communication methods might solve this problem.

3. System architecture An information collection system uses sensors to collect soil information, including temperature and water content. To collect soil information, the WSN uses the CC2430 wireless transceiver module based on the ZigBee agreement. The WUSN uses the nRF905 wireless chip to collect and transmit information. The wireless sensor node was designed using a modular design method. The architecture of the terrestrial WSN is shown in Fig. 1. The WUSN uses a nRF905 wireless chip instead of a CC2430 RF chip. Each node comprises a sensor module, processor module, wireless communication module and energy supply module. The network topology structure is the foundation of the network. A good network topology structure should consider the specific application and have a simple, reliable and effective implementation [22–24]. The proposed design combines the terrestrial and underground wireless sensor network structures. A traditional WSN is adopted above depths of 40 cm, and a WUSN is adopted below depths of 40 cm. The sink node of the WUSN is on the ground. All nodes in the WUSN will transmit data to the terrestrial sink node, which better conceals the network. The location of a WUSN node depends on the specific application. It can be located at the same depth, at a different depth or as a different layer. The sink node can be fixed or movable but must remain in the range of communication. The topological structure is shown in Fig. 2.

4. Deployment and connectivity of nodes To monitor the water content of the soil in real time, signal acquisition nodes are buried in and below the cultivate layer. The hybrid wireless sensor network node and sensor are deployed as shown in Fig. 3. The unique channel characteristics and heterogeneous network architecture of the WUSN complicate the connectivity analysis. In particular, there are three communication channels, based on the locations of the transmitter and the receiver, in a WUSN: undergroundto-underground, underground-to-aboveground, and abovegroundto-underground. As shown in Fig. 4, a WUSN comprises underground sensors deployed in the sensing field, fixed aboveground data sinks set around

Processing Model Communication model

Sensing Model CPU AD/ DA Storage

MAC

Various sensors

Network

a linear model, and the correlation coefficient, R 2, was greater than 0.90. The near surface wireless underground sensor network system used for golf courses was developed and the acquisition nodes, relay nodes and gateway node were designed in [19]. Each underground acquisition node comprises a soil moisture sensor, controller, wireless transceiver (Nordic NRF905, frequency 868 MHz), antenna, memory unit and battery power module. Each collection node can connect with several moisture sensors. Sink nodes are acquisition nodes with no sensors that collect the data from acquisition nodes. Sink nodes can communicate with other sink nodes and gateway nodes in the routing algorithm. Gateway nodes control data storage and the transmission of sink nodes and connect with a computer or GPRS module using a RS232 interface. A gateway node can connect with 31 sink nodes simultaneously, and it can be controlled remotely and visited through DDI as sink nodes. Experimental results indicated that the system operated normally, and soil moisture data at the different depths were transmitted to the central computer stably and accurately. In [20], Agnelo R and Silva studied how the communication performance between terrestrial and underground nodes was impacted by factors such as the antenna bandwidth of the WSN nodes in the 433 MHz frequency, the depth at which nodes are buried in the soil (15 and 35 cm) and the water content of the soil (volumetric water content was 9.5% and 37.3%). The field experiment showed that the ultra-wideband antenna increased the communication range of the original antennas by more than 350%. The transmission distance dropped by 70% when the volumetric water content increased from 9.5% to 37.3%. When the deployment depth of the nodes changed from 35 cm to 15 cm, the transmission distance of the signal for the terrestrial nodes to the underground nodes (downlink transmission) increased three times, and the transmission distance of the signal for the underground nodes to the terrestrial nodes (uplink transmission) only increased 0.4 times. A center pivot sprayer was combined with wireless sensor network nodes with a commercial full-wave 433 MHz magnetic antenna for the precise irrigation of corn in [20]. Eight underground sensor acquisition nodes (at a depth of 35–40 cm) were deployed circularly within the operational range of the sprayer, and a signal receiving node (2.5 m from the ground level) was installed in the sprayer. In this paper, the influence of ground cover, corn canopy and the rotation speed of the sprayer on receiving information was analyzed. In [21], distortion of the acquisition signal of the soil's moisture was greatly impacted by rainfall and stormy weather conditions, soil compactness, soil density, vegetation cover, topology structure parameters of the wireless underground sensor network, sampling time and sampling density. Mehmet C. Vuran studied the channel model of wireless sensor network electromagnetic wave transmission in soil in [21]. He analyzed path loss, bit error ratio, maximum transmission distance and the water content test error of electromagnetic wave transmission under variable conditions, such as the composition of the soil, volumetric water content of the soil (5%–25%), node deployment depth (0.1 m–2 m), internode distance (0.5 m–5 m), and frequency (300 MHz–900 MHz). The signal attenuation and antenna size results showed that frequencies between 300 MHz and 400 MHz were more suitable for wireless underground sensor networks. Because the transmission distance was influenced by the deployment depth, no one frequency is the optimum, and WUSNs based on many frequency cognitive radio techniques are more adaptable to environmental changes. For networks that have shallow nodes, the double path channel model can be used. The single path channel model can be used for networks that have deeply buried nodes. The topology structure of the sensor networks can be designed according to the node deployment depth. The maximum transmission distance was 5 m under the 300 MHz–400 MHz frequency. Therefore, multiple hop

61

Power Model

Fig. 1. Architecture of the wireless sensor network node.

CC2430 /nRF905 RF chip

62

X. Yu et al. / Computer Standards & Interfaces 35 (2013) 59–64

Movable sink

Fixed sink

40 cm Fig. 4. Connectivity in hybrid wireless sensor networks.

Fig. 2. Topology structure of a wireless sensor network.

the sensing field, and mobile data sinks carried by people or machineries inside the sensing field.

5. Experiments and analysis 5.1. Experimental setup of a WUSN Because of the diverse terrain and environmental conditions encountered in agriculture, a permanent WUSN solution must be adaptable and well sheltered from the environment. A wireless underground sensor network that is designed to provide quick and accurate information about the water content of soil at several depths is investigated. This study provides some technical details on the deployment of a remote real-time monitor network for agricultural environments. When WUSN nodes are buried, there are two means by which electromagnetic waves are propagated. One is by direct penetration of the soil, and the other is the transmission method of the communication between WUSN nodes. By modeling, designing and testing the WUSN node, this paper studies the impact of soil parameters, node depth, signal frequency and attenuation on the process of transmission. It is expected that this information will be of great help in the development of a wireless underground sensor network system. The main test model is shown in Fig. 5. When wireless underground sensor network nodes transmit soil information, reflection, scattering and diffraction can occur simultaneously in the soil and at the interface between the soil and air. The frequency of the electromagnetic signals is influenced. In addition, the agricultural environment changes constantly, and the water content of soil has a significant effect on path loss. When underground sensor nodes are deployed, a depth that is both economical and minimizes the signal path loss must be determined.

Fig. 3. Scheme for WUSN nodes and sensor deployment.

The Peplinski principle [25] defines the complex propagation constant of an EM wave in soil as γ = α + jβ with vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 3 u  2 u ε″ uμε′ 4 α ¼ ωt 1þ −15; 2 ε′ vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 3ffi u  2 u ε″ uμε′ 4 β ¼ ωt 1þ þ 15; 2 ε′

where ω = 2πf is the angular frequency, μ is the magnetic permeability, and ε′ and ε″ are the real and imaginary parts of the dielectric constant, respectively. From the above equation, the complex propagation constant of an EM wave in soil depends on the operating frequency, the sand and clay fractions of the soil, the bulk density, and the soil moisture or volumetric water content. Consequently, the path loss also depends on these parameters. In the experiment, we assumed that the soil composition was 15% clay, 35% silt, and 50% sand particles. The bulk density was 1.5 g/cm 3, and the solid soil particle density was 2.6/cm 3 unless otherwise noted. Three different frequencies of RF module nRF905 were considered. The attenuation of signal strength and the bit error rate were measured in soils of different volumetric water contents (VWCs): 5, 10, 15, 20 and 25%. For each frequency, the path loss was measured

Fig. 5. The test model.

X. Yu et al. / Computer Standards & Interfaces 35 (2013) 59–64

63

Fig. 8. The dependence of path loss on the node deployment depth.

Fig. 6. The dependence of path loss on operating frequency and volumetric water content.

at different WUSN deployment depths (h): 0.2, 0.4, 0.6, 0.8, 1, 1.2, 1.4, 1.6, 1.8 and 2 m.

5.2. Result analysis The path loss is the attenuation of the received signal strength as compared to the source signal strength and reflects the efficiency of the wireless electromagnetic signal transmission. MATLAB was used to investigate the relationship between path loss and parameters such as the operating frequency, the deployment depth of nodes, and the volumetric water content. Figs. 6 and 7 show the path loss and bit error rate of the wireless signals as a function of the volumetric water content of soil for signals of different frequencies. Fig. 8 shows the path loss as a function of the deployment depth for a 433 MHz RF signal. These results indicate the following: (1) A wireless underground sensor network developed to acquire environmental information in soil was studied. The underground sensor node was combined with embedded processors to collect, transmit, store and display soil property parameters. The nodes satisfy the requirements of low power consumption and low cost and provide high real-time reliability for soil property monitoring. (2) The path loss and bit error rate of radio signals were determined as a function of volumetric water content for three RF modules, each with a different carrier frequency. The results show that soil attenuation and bit error rate are smallest for low frequency signals and soils with low volumetric water content.

(3) At the 433 MHz operating frequency, the path loss is influenced by the deployment depth of the WUSN node. The results indicate that signal attenuation can be minimized by a suitable choice of deployment depth. (4) In comparison with manual soil property monitoring, an advanced wireless sensor network provides better real-time soil property collection and is the foundation for water-saving agricultural applications. 6. Conclusions In this paper, we introduced a hybrid wireless sensor network architecture for agriculture. This network reduces the intensive human involvement required in current agricultural information collection systems and provides information that is more accurate than the existing sensor networks. This advanced sensor network includes a terrestrial wireless sensor network and a wireless underground sensor network. The hybrid WSN architecture combines the advantages of existing sensor techniques. In particular, the WUSN provides collection functionality when the monitoring area is not in the line-ofsight of the terrestrial sensor networks, and the mobile sink nodes provide an information acquisition capability after collection. The network architecture of the hybrid wireless sensor networks was described, and the deployment strategy of the hybrid sensor networks was discussed. Based on the network architecture and deployment strategies, tests of wireless underground sensor networks were performed. Finally, a test bed will be developed and field experiments will be conducted to test the performance of the hybrid wireless sensor network system in real agricultural applications. Acknowledgments The authors wish to thank the National Engineering Research Center for Water-Saving Irrigation, which partially supported this research through the “National 863 Plan” (2006AA100217), the “National Science and Technology Support Plan” (2007BAD88B10) and the “National Natural Sciences Foundation Project” (40701092). The authors are also grateful to the anonymous reviewers for their valuable feedback. References

Fig. 7. The dependence of the bit error rate on operating frequency and volumetric water content.

[1] A. Sheth, K. Tejaswi, P. Mehta, et al., Senslide: a sensor network based landslide prediction system, Proceedings of Sensys ‘05—the 3rd International Conference on Embedded Networked Sensor Systems, 2005, pp. 280–281. [2] K. Martinez, R. Ong, J. Hart, Glacsweb: a sensor network for hostile environments, IEEESECON l (2004) 81–87. [3] G.W. Allen, K. Lorincz, M. Welsh, O. Marcillo, et al., Deploying a wireless sensor network on an active volcano, IEEE Internet Computing 10 (2) (2006) 18–25. [4] I.F. Akyildiz, W. Su, Y. Sankarasubramaniam, E. Cayirci, Wireless sensor networks: a survey, Computer Networks 38 (4) (2002) 393–422. [5] P. Erich, Stuntebeck, Dario Pompili, Tommaso Melodia, Wireless underground sensor networks using commodity terrestrial motes, IEEE (2006) 112–114.

64

X. Yu et al. / Computer Standards & Interfaces 35 (2013) 59–64

[6] E. Berman, G. Calinescu, C. Shah, A. Zelikovsky, Power efficient monitoring management in sensor networks, Proceedings of IEEE Wireless Communication and Networking Conference (WCNC04), Atlanta, USA, 2004. [7] Z. Abrams, A. Goel, S. Plotkin, Set K-cover algorithms for energy efficient monitoring in wireless sensor networks, Proceedings of the 3rd International Conference on Information Processing in Sensor Networks (IPSN04), Berkeley, California, USA, 2004. [8] J. Carle, D. SimPlot-Ryl, Energy efficient area monitoring by sensor networks, IEEE Computer Magazine 37 (2) (February 2004) 40–46. [9] I.F. Akyildiz, E.P. Stuntebeck, Wireless underground sensor networks: research challenges, Ad Hoc Networks 4 (2006) 669–686. [10] I.F. Akyildiz, M.C. Vuran, Z. Sun, Channel modeling for wireless underground communication in soil, Physical Communication 2 (2009) 167–183. [11] J.A. Lopez, F. Soto, J. Suardiaz, et al., Wireless sensor networks for precision horticulture in Southern Spain, Computers and Electronics in Agriculture 68 (3) (2009) 25–35. [12] O. Green, E.S. Nadimi, V. Blanes, et al., Monitoring and modeling temperature variations inside silage stacks using novel wireless sensor networks, Computers and Electronics in Agriculture 69 (1) (2009) 149–157. [13] L. Li, X.M. Wen, Energy efficient optimization of clustering algorithm in wireless sensor network, Journal of Electronics and Information Technology 30 (4) (2008) 966–969. [14] L. Li, H.X. Li, H. Liu, Greenhouse environment monitoring system based on wireless sensor network, Transactions of the Chinese Society for Agricultural Machinery 9 (40) (2009) 228–231. [15] R.B. Zhang, G.D. Gu, Y.B. Feng, et al., Realization of communication in wireless monitoring system in greenhouse based on IEEE802.15.4, Transactions of the Chinese Society for Agricultural Machinery 39 (8) (2008) 119–122. [16] Y.H. Cai, G. Liu, L. Li, et al., Design and test of nodes for farmland data acquisition based on wireless sensor network, Chinese Society of Agricultural Engineering 25 (4) (2009) 176–178. [17] Y.B. Feng, R.B. Zhang, G.D. Gu, Application of wireless sensor network in water-saving irrigation, China Rural Water and Hydropower (2) (2007) 24–26. [18] H.R. Bogena, J.A. Huismana, H. Meierb, U. Rosenbauma, A. Weuthena, Hybrid wireless underground sensor networks: quantification of signal attenuation in soil, Vadose Zone Journal 8 (3) (August 2009) 755–761. [19] J.R. Coen, K. Henk, K. Leon, et al., A new wireless underground network system for continuous monitoring of soil water contents, Water Resources Research 45 (36) (2009) 36–44. [20] Agnelo R, Silva and Mehmet C. Communication with above devices in wireless underground sensor networks: a empirical study [J]. Communications (ICC), 2010 IEEE International Conference Proceedings, IEEE Communication Society, Cape Town, 23–27 [21] D. Xin, C. Mehmet, Spatio-temporal soil moisture measurement with wireless underground sensor networks, Ad Hoc Networking Workshop 1 (2010) 1–8. [22] L. Li, C. Mehmet, I.F. Akyildizy, Characteristics of underground channel for wireless underground sensor networks, The Sixth Annual Mediterranean Ad Hoc Networking Workshop, 6, 2007, pp. 12–15. [23] S. Slijepeevie, M. Potkonjak, Power efficient organization of wireless sensor networks, Proceedings of IEEE International Conference on Communications (ICC01), Helsinki, Finland, 2001. [24] E. Shih, S. Cho, N. Ickes, R. Min, A. Sinha, A. Wang, A. Chandrakasan, Physical layer driven protocol and algorithm design for energy-efficient wireless sensor networks, Proceedings of 2001 ACM MOBICOM, Rome, Italy, July 2001, pp. 272–286. [25] N. Peplinski, F. Ulaby, M. Dobson, Dielectric properties of soils in the 0.3–1.3 GHz range, IEEE Transactions on Geoscience and Remote Sensing 33 (3) (1995) 803–807.

Xiao Q. Yu received a B.S. degree from the Department of Information Engineering, Lanzhou University of Finance and Economics, Lanzhou, and an M.S. degree from the Department of Mechanical and Electric Engineering, Northwest A & F University, Shaanxi, China in 2006 and 2009, respectively. Currently, she is pursuing a Ph.D. degree from the Department of Water Resources and Architectural Engineering under the supervision of Prof. Pu T. Wu. Her current research interests are in agricultural water-soil engineering and wireless sensor networks.

Pu T. Wu received his B.S. degree from the Department of Water Resources Engineering, College of Northwest Agriculture, Shaanxi, China in 1985. He received his M.S. and Ph.D. degrees from the Water Conservation of Chinese Academy of Sciences, Chinese Academy of Sciences, Beijing, China in 1990 and 1996, respectively. He was an Assistant Researcher at the Water Conservation of Chinese Academy of Sciences from 1991 to 1995. From 1995 to 1997, he served as an associate researcher at the Water Conservation of Chinese Academy of Sciences. He has been a researcher at the Water Conservation of Chinese Academy of Sciences since 1997. Since 1999, Prof. Wu has served as a Director for the Water-saving Irrigation Engineering Technology Research Center in Yangling. In addition, he has been a Vice President for Northwest A & F University, China since 2004. Currently, he is a Vice President at Northwest A & F University. His current research interests are in soil and water conservation and water saving agriculture. Professor Pu T. Wu has contributed engineering technologies for the efficient use of rain in arid regions, key technologies and equipment for water saving irrigation, the efficient water use technology for regional agriculture, and the integration and demonstration of modern water-saving agriculture technologies. He has published more than 150 academic papers, including more than 20 EI articles; 10 published works and hold national invention patents for more than 10 items. Wen T. Han received his B.S. degree from the Department of Mechanical and Electric Engineering, Northwest Agriculture University, Shaanxi, China in 1996. He received his M.S. and Ph.D. degrees from the Department of Mechanical and Electric Engineering, Northwest A & F University, Shaanxi, China in 1999 and 2004, respectively. Working experiences: 2005–present: Assistant researcher, Institute of Soil and Water Conservation of Chinese Academy of Sciences, Northwest A & F University, National Engineering Research Center for Water Saving Irrigation at Yangling.2004–2005: Assistant Professor, Department of Mechanical and Electric Engineering, Northwest A & F University. 2001–2004: A lecturer, Department of Mechanical and Electric Engineering, Northwest A & F University. Research interests: Monitoring of crop and environment information; intelligent control for precise irrigation; water distribution simulation of sprinkler irrigation; development of nozzle.Currently, he has published more than 20 academic papers, including 5 EI articles; 2 ISTP; and he holds 2 national invention patents. Zeng L. Zhang received his B.S. degree from the Department of Mechanical and Electric Engineering, Harbin Institute of Technology, Harbin, and his M.S. degree from the Department of Mechanical and Electric Engineering, Northwest A & F University, Shaanxi, China in 2000 and 2007, respectively. Currently, he is a teacher in the Department of Mechanical and Electric Engineering, Northwest A & F University, Shaanxi. He is pursuing a Ph.D. degree under the supervision of Prof. Pu T. Wu. His current research interests are in agricultural water–soil engineering and wireless sensor networks.