Proceedings of IEEE CCIS2012
EFFICIENT DEPLOYMENT SURFACE AREA FOR
UNDERWATER WIRELESS SENSOR NETWORKS
Shiwei Li1, 2, Wenjing Wang2, Juwei Zhang1, 2
1
School of Electronic and Information Engineering,
Henan University of Science and Technology, Luoyang 471003, China
2
Science and Technology on Electro-optic Control Laboratory, Luoyang 471003, China
jiankelang@[Link], juweizhang@[Link]
Abstract: Underwater wireless sensor networks provide underwater warfare capabilities of the naval forces can
a new remote surveillance way to monitor the be enhanced by using UWSNs, since it can be used for
underwater environment, and have drawn increasing surveillance, submarine detection, mine countermeasure
attention in the past few years from both academic and missions and unmanned operations in the enemy fields
industry. Underwater sensor networks will find [2].
applications in oceanographic data collection, ocean
However, new challenges are introduced in terms of
sampling, pollution and environmental monitoring,
deployment, connectivity, coverage and mobility as the
offshore exploration, disaster prevention, assisted
nodes are deployed in underwater environments. Deploy
navigation, distributed tactical surveillance, mine
nodes in the accuracy place by an efficient way is the
reconnaissance, and submarine surveillance. Plenty of
first thing need to EH done for an underwater network.
challenges have to be faced during the research, since
The other issues, like how to obtain a high level of
the nature of underwater environment variable
coverage while the network can keep high quantity of
temporally and spatially
connectivity, are also important trad-offs need to be
In this paper, the efficient deployment of surface area for consered. In order to deal with these problems, a great
underwater wireless sensor networks was presented. To deal works have been done.
familiarize with underwater wireless sensor networks
Since the 2D and 3D environment is very different,
deployment, this paper started by providing background
majority of studies on the coverage and connectivity
information on the architecture and trajectory of the
about 3D have to be done. In Ref.[3], the Reuleaux
sinking node. Then, a kind of novel inflatable sensor
tetrahedron model to characterize k-coverage of a 3D
nodes was proposed to make sure that they can have a
field is proposed, as they focuse on the connectivity and
given trajectory, according to the analysis sinking
k-coverage issues. Many of the requirements are loosen
process of the sensor nodes which tie to the anchor with
to make it more useful in the real world. Such as 1-
winch. For a given target area, the reducing of surface
coverage is often used instead of k-coverage (here k is
deployment area and numbers of nodes is the most
bigger than 1) to reduce the number of the nodes.
significant advantage. To illustrate the performance, the
compare was made from two aspects. Ref. [4] introduce a Mending Current Mobility model
(MCM) which can impact on the coverage and
Keywords: Underwater wireless sensor networks;
connectivity of the network under different
Deployment; Sensing coverage; Sensor nodes;
deployments. MCM is the ILUVW physically-inspired
Reasonable adjustment of node buoyancy
mobility model which is used in the analysis of mobile
underwater sensor networks. The lifetime of network
1 Introduction was shown can be prolonged with the process of a
Underwater wireless sensor networks (UWSNs) multiple deployment. A novel kind of model is presented
deployed in a given region to perform collaborative and a deployment based on that can perform better, but
monitoring tasks [1], consists of variable number of the use will be limited to the model.
underwater wireless sensor nodes with sensing, A great job has been done by Kemal [5], a kind of
processing, and communication capabilities. A new distributed nodes deployment technique is presented for
remote monitoring and control technology was shown, UWSNs. The technique can improve the coverage of the
and recently, have becoming a promising technology for network and provide a good connectivity with the
underwater environment monitoring. Traditional surface station. 7KH architecture of the UWSNs is shown
technical methods for oceanic exploration have many detailed which include all kinds of sensor nodes, like
disadvantages due to oceanic harsh environments and anchor to the bottom which can adjust the depth, float on
undiscovered working manners [1]. UWSNs, which has the surface with the buoy. To obtain the performance,
attracted rapidly growing research interests, is a new numbers of sensor nodes is needed to be deployed.
paradigm acting as the extension of traditional terrestrial
wireless sensor networks (WSNs). Additionally, the
___________________________________
978-1-4673-1857-0/12/$31.00 ©2012 IEEE
Proceedings of IEEE CCIS2012
Closest to our work is presented in Ref. [6], where a
deployment system with depth adjustment sensor nodes
is proposed. The system show great promise, however,
there are several issues to be addressed before deploying
the system for long periods of time or in deeper water
depth. Apparently, the system only can be used in lower
water environment like Michael [Link] has done in Ref.
[7]. The sensor nodes are deployed and anchor to the
bottom of the ocean at first, and then adjust the length of
the winch which is tied to the nodes, so that they can be
placed in all kinds of depth. Apparently, that’s a time
consuming process. For this reason, some new ideas
come up to us to improve the performance of the system
by analysis the deployment of surface area.
In this paper, the analysis of surface area deployment for
UWSNs was presented. The trajectory of the sinking
sensor node was shown firstly. According to this, the
process of the deployment was clearly to know, so that
Figure 1 Architecture for 3D underwater sensor networks
the sinking way can be changed by some special
measures. With changing the column of the inflatable 2.2 Sensor nodes
buoy, which tie to the anchor with the winch, to have a
given trajectory. The simulation shown that, the time In the architecture for 3D underwater sensor networks, a
that spent on deploy nodes and the numbers of nodes new kind of sensor nodes was used. The sensor node
that need to be deployed for a given target area were all with the shape of sphere and the ability of inflate so that
reduced, by using this new kind of deployment. it can autonomous adjust the column. The radius of the
sensor nodes was assumption to be R. The new kind of
The paper is organized as follows. In the following sensor nodes was going to be designed.
section, a kind of network architecture and a new kind of
nodes which going to be designed, was introduced.
Next, the deployment of the nodes was discussed in
3 Dynamics analysis of sensor node
Section III, especially the trajectory of the sinking In this section, we illustrated how the sensor nodes that
nodes. Following the system description, experiments, would be deployed were working, during the sinking
which characterized the performance of the system in progress. As we all know, the distance is the shortest
water environment, were presented in Section IV. In when the sensor nodes go along with a straight line by
addition, we concluded and presented future directions chang the buoyancy of the sensor.
of this research in Section V.
Firstly, under the presence of ocean current, the
2 Network architectures and nodes trajectory of a sinking sensor node was evaluated by
stress analysis. In particular, we first considered the
In this section, a kind of network architecture for 3D ideal case in which the velocity of the ocean current
underwater sensor networks was described. In particular, from one direction. The dynamics analysed was shown
a new kind of nodes, which was going to be designed in Figure 2.
and could enhance the capabilities of underwater sensor
networks, would be introduced.
2.1 Network architecture
A reference 3D architecture of underwater sensor
networks is shown in Figure 1. A group of sensors are
anchored to seafloor. Each sensor is equipped with a
floating buoy and an acoustic transceiver. The floating
buoy can push the sensor to the ocean surface by inflate
it with a pump. The depth of the sensor can then be
regulated by adjusting the length of the wire that Figure 2 Dynamics analysed
connects the sensor to the anchor, by means of an
The weight force
electronically controlled engine that resides on the
sensor. Each sensor can communicate with the surface & )&
station by the acoustic transceiver. FW m g (1)
Proceedings of IEEE CCIS2012
which depends on the weight m( kg ) of the node, and v v(t0 ) at (7)
on the terrestrial gravitational
Finally, by solving (6) and (7), we can obtain the
acceleration g 9.80m / s 2 ; solution of R and a , which allow us to track the node.
The buoyant force due to the Archimedes' Law
& &
4 Sensor nodes deployment area
FB U wV g (2)
In this section, we discuss the simulation performance of
, which is equal to the weight of the displaced fluid, the proposed algorithm. The surface area will be
where U w 1050kg / m3 represents the average computed when a target area on the ocean bottom need
to be covered. A compare will be made with the
3
density of salty water and volume V ( m ) of the node; algorithm that has been illustrated in Ref.[8]. The
The fluid resistance force sinking trajectories of sensors will be changed for the
& & effect of the currents. Therefore, in order to deploy
FR K U w P AR v (3) enough sensors inside the target area, the surface
deployment should take this into account.
, which is proportional through the constant
K 0.2 Nm 2 s / kg to a parameter P accounting for We consider a square field of side length
the resistance caused by the object shape, to the velocity 500m(l 500m, w 500m)
where the current direction
&
v (m / s ) of the node, to its cross-section AR (m 2 ) ;
is known and just from one direction. The deployment
The force of the current, area is shown in Figure 3, which according to the
algorithm proposed by us and Ref.[8], respectively.
& & & cos D
FC CV AC (vc v ) CV AC v (4)
cos(D E )
which is proportional through the constant
C 721.7 Ns / m3 to the difference between the
&
velocity of the ocean current vc (m / s ) and the object
& 2
velocity v ( m / s ) , to the cross-section AC ( m ) of the
object facing the current, and to an object-dependent
shape factor V (when the shape of the object is sphere
& & &
V 1 ). Apparently, FB , FR and FC will be changed with
changing the column of the sensor node V, as AR equal
to AC will be changed with inflating or deflating.
The time that node will be spent is shortest when the
&
acceleration of the node has the same direction of v .
The coordinate that has made is shown in Figure 2.
Then, project the force on the y-axes, the following Figure 3 Deployment surface area for known current direction
equation is apparently
Figure 3 shows that, the performance of our algorithm is
better than that proposed by Ref.[8], as only a smaller
( FW FB ) sin D FR cos D FC sin D surface area need to be deployed sensors. According the
® (5)
¯( FW FB ) cos D FR sin D FC cos E ma description in section III, the algorithm in this paper can
make sure the sensor nodes go straight to the bottom of
Denote FW , FB , FR , FC , and the cross-section AR = AC the target area. As a result, the distance of the
deployment will be shorter and the time that the sinking
which is equal to S R since the shape of the nodes is
2
progress spent will be shorter, too. That is to say, for the
sphere, into (5). The following equations can be same target area, the fewer sensor nodes will be needed.
obtained
In the other hand, the relationship between the depth and
4S R3 g cos D sin D the side length (here just the w is discussed as the water
°(mg U w )sin D K Uw PS R 2 v cos D CVS R 2 v
° 3 cos(D E ) (6) direction is known) is also should be considered. Figure
® 4 shows the relationship between the depth and the side
°(mg U 4S R 3
g cos D cos E
)cos D K U w PS R 2 v sin D CVS R 2 v ma length w .
°¯ w
3 cos(D E )
Additionally,
Proceedings of IEEE CCIS2012
Acknowledgments
This work was supported by the Aeronautical Science
Foundation of China under grant number 20115142005 and the
National Natural Science Fund project under grant number
61040010.
References
[1] Ian F. Akyildiz, Dario Pompili, Tommaso Melodia,”
Underwater acoustic sensor networks: research
challenges”, Ad Hoc Networks 3 (2005), pp, 257–279
[2] Melike Erol-Kantarci, Hussein T. Mouftah, and Sema
Oktug,” A Survey of Architectures and Localization
Techniques for Underwater Acoustic Sensor Networks”,
IEEE COMMUNICATIONS SURVEYS &
TUTORIALS, VOL. 13, NO. 3, THIRD QUARTER
Figure 4 The relationship between the depth and the side 2011, pp, 487-502.
length w [3] Habib M. Ammari, Sajal [Link], “A Study of k-Coverage
and Measures of Connectivity in 3D Wireless Sensor
Figure 4 shows that the side length w1 will be changed Networks”, VOL.59, NO.2 , FEBRUARY,2010, pp,243-
when the depth is increasing while the side length w2 257.
not. The deeper the longer sinking time will be cost. As [4] Antonio Caruso, Francesco Paparella, Luiz F. M. Vieira,
a result, the horizontal move distance will be longer. Melike Erol, Mario Gerla, “The Meandering Current
Apparently, this will make the surface area of the Mobility Model and its Impact on Underwater Mobile
deployment bigger. So, there are more nodes need to be Sensor Networks”, IEEE INFOCOM 2008 proceedings,
pp, 771-779.
deployed for the bigger surface area.
[5] Kemal Akkaya, Andrew Newell, “Self-deployment of
sensors for maximized coverage in underwater acoustic
5 Conclusions and future work sensor networks”, Computer Communications, 32 (2009),
pp, 1233–1244.
In this paper, the sinking trajectory of sensor nodes was [6] Carrick Detweiler, Marek Doniec, Iuliu Vasilescu, , and
analyzed at first. Then, a deployment analysis was given Daniela Rus, “Autonomous Depth Adjustment for
and guidelines on how to choose the deployment surface Underwater Sensor Networks: Design and Applications”,
area were provided. Thus, the key target of this paper IEEE/ASME TRANSACTIONS ON MECHATRONICS,
was given a novel idea of sensor node which can make 2011, pp, 1-9.
sure achieve better performance for UWSNs [7] Michael L. Incze, “Optimized deployment of autonomous
underwater vehicles for characterization of coastal
deployment. A new way was proposed to choose the
waters”, Journal of Marine Systems, Journal of Marine
surface deployment area and it was proved that the Systems, 78 (2009) S415–S424.
performance of sensor nodes that deploy for the margin [8] Dario Pompili, Tommaso Melodia, Ian F.
area was better. The performance that demonstrated by Akyildiz,”Three-dimensional and two-dimensional
the simulation results was much better and was easy to deployment analysis for underwater acoustic sensor
achieve. networks”, Ad Hoc Networks 7 (2009), pp, 778–790.
[9] C. Detweiler, M. Doniec, M. Jiang, M. Schwager, R.
There were so many benefits of this kind of sensor Chen, and D. Rus,“Adaptive decentralized control of
nodes so that a lot of future works needed to be done. underwater sensor networks for modeling underwater
Particularly on the design of the new kind of sensor phenomena,” in Proc. Conf. Embedded Networked Sensor
nodes which could be used to control the force of the Syst., 2010, pp. 253–266.
sensor more efficient. [10] Peter King, Ramachandran Venkatesan, and Cheng Li,”
An Improved Communications Model for Underwater
Sensor Networks”, IEEE "GLOBECOM" 2008
proceedings, pp, 1-6.
