Underwater optical wireless communication

network
Shlomi Arnon, MEMBER SPIE
Ben-Gurion University of the Negev
Electrical and Computer Engineering Department
Satellite and Wireless Communications
Laboratory
P.O. Box 653
Beer-Sheva, IL-84105, Israel
E-mail: shlomi@ee.bgu.ac.il
Abstract. The growing need for underwater observation and subsea
monitoring systems has stimulated considerable interest in advancing
the enabling technologies of underwater wireless communication and
underwater sensor networks. This communication technology is ex-
pected to play an important role in investigating climate change, in moni-
toring biological, biogeochemical, evolutionary, and ecological changes
in the sea, ocean, and lake environments, and in helping to control and
maintain oil production facilities and harbors using unmanned underwa-
ter vehicles UUVs, submarines, ships, buoys, and divers. However, the
present technology of underwater acoustic communication cannot pro-
vide the high data rate required to investigate and monitor these envi-
ronments and facilities. Optical wireless communication has been pro-
posed as the best alternative to meet this challenge. Models are
presented for three kinds of optical wireless communication links: a a
line-of-sight link, b a modulating retroreector link, and c a reective
link, all of which can provide the required data rate. We analyze the link
performance based on these models. From the analysis, it is clear that
as the water absorption increases, the communication performance de-
creases dramatically for the three link types. However, by using the scat-
tered light it was possible to mitigate this decrease in some cases. It is
concluded from the analysis that a high-data-rate underwater optical
wireless network is a feasible solution for emerging applications such as
UUV-to-UUV links and networks of sensors, and extended ranges in
these applications could be achieved by applying a multi-hop concept.
2010 Society of Photo-Optical Instrumentation Engineers. DOI: 10.1117/1.3280288
Subject terms: optical communication; underwater; subsea; FSO; ocean.
Paper 090580PR received Jul. 30, 2009; revised manuscript received Nov. 8,
2009; accepted for publication Nov. 11, 2009; published online Jan. 15, 2010.
This paper is a revision of a paper presented at the SPIE conference on
Free-Space Laser Communications IX, August 2009, San Diego, California. The
paper presented there appears unrefereed in SPIE Proceedings Vol. 7464.
1 Introduction
The present technology of acoustic underwater communi-
cation is a legacy technology that provides low-data-rate
transmissions for medium-range communication. Data rates
of acoustic communication are restricted to around tens of
thousands of kilobits per second for ranges of a kilometer,
and less than a thousand kilobits per second for ranges up
to 100 km, due to severe, frequency-dependent attenuation
and surface-induced pulse spread.
14
In addition, the speed
of acoustic waves in the ocean is approximately 1500 m/ s,
so that long-range communication involves high latency,
which poses a problem for real-time response, synchroni-
zation, and multiple-access protocols. As a result, the net-
work topology is simple and goodput is low. In addition,
acoustic waves could distress marine mammals such as dol-
phins and whales. As a result, acoustic technology cannot
satisfy emerging applications that require around the clock,
high-data-rate communication networks in real time. Ex-
amples of such applications are networks of sensors for the
investigation of climate change; monitoring biological, bio-
geochemical, evolutionary, and ecological processes in sea,
ocean, and lake environments; and unmanned underwater
vehicles UUVs used to control and maintain oil produc-
tion facilities and harbors Fig. 1. An alternative means of
underwater communication is based on optics, wherein
0091-3286/2010/$25.00 2010 SPIE Fig. 1 The line-of-sight communication scenario.
Optical Engineering 491, 015001 January 2010
Optical Engineering January 2010/Vol. 491 015001-1
Downloaded from SPIE Digital Library on 28 Jan 2010 to 132.72.138.1. Terms of Use: http://spiedl.org/terms
high data rates are possible. However, the distance between
the transmitter and the receiver must be short, due to the
extremely challenging underwater environment, which is
characterized by high multiscattering and absorption. Mul-
tiscattering causes the optical pulse to widen in the spatial,
temporal, angular, and polarization domains.
Although high data rates are threatened by extremely
high absorption and scattering, there is evidence that broad-
band links can be achieved over moderate ranges. Hanson
and Radic
5
demonstrated 1-Gbit / s transmissions in a labo-
ratory experiment with a simulated aquatic medium with
scattering characteristics similar to oceanic waters.
Cochenour, Mullen, and Laux
6
measure both the spatial
and temporal effects of scattering on a laser link in turbid
underwater environments. Using Monte Carlo simulations
and measurement results, they predict longer-range under-
water free-space optical performance with bandwidths
greater than 5 GHz for a range of 64 m in clear ocean
water, dropping to 1 GHz for a range of 8 m in turbid har-
bor water. The authors of Refs. 7 and 8 examine the funda-
mental physics and natural variability of underwater optical
attenuation and discuss the design issues of underwater op-
tical communications associated with oceanic physics and
parameter variability. In Ref. 9 the authors examine the
potential of subsea free-space optics for sensor network ap-
plications, leveraging the emerging technologies of highly
sensitive photon-counting detectors and semiconductor
LED and laser light sources in the solar blind UV. The
authors of Ref. 10 propose to use retroreecting free-space
optical links in water, which allow much of the weight and
power payload of the system to be located at one end. Ar-
non and Kedar
11
propose a novel non-line-of-sight network
concept in which the optical link is implemented by means
of back reection of the propagating optical signal at the
ocean-air interface, which could help to overcome obstruc-
tions. In Ref. 12 the possibility of a wireless sensor network
concept dubbed optical plankton is described and evalu-
ated. The paper by Jaruwatanadilok
13
presents the modeling
of an underwater wireless optical communication channel
using the vector radiative transfer theory. The vector radia-
tive transfer equation captures the multiple scattering in
natural water, and also includes the polarization of light.
I present models of three kinds of optical wireless com-
munication: a a line-of-sight link, b a modulating ret-
roreector link, and c a reective link, all of which can
provide the required data rate. I also present performance
analyses based on these models. From the analyses it is
clear that as the water absorption increases due to changes
in water turbidity, the communication performance de-
creases dramatically for all three link types, but the modu-
lated retroreector link is the most affected. However, the
absorption coefcient increases more moderately than does
the water turbidity. We conclude from the analysis that a
high-data-rate underwater optical wireless network is a fea-
sible solution for emerging applications such as UUV-to-
UUV links and networks of sensors. Extended ranges in
these applications could be achieved by applying a multi-
hop concept.
The remainder of the paper is organized as follows. Sec-
tion 2 describes the properties of the underwater optical
wireless communication channel. Section 3 presents the
communication link models. Section 4 contains a discus-
sion and a numerical example. Finally, Sec. 6 summarizes
our results.
2 The Properties of the Underwater Optical
Wireless Communication Channel
Light pulses propagating in aquatic medium suffer from
attenuation and broadening in the spatial, angular, tempo-
ral, and polarization domains. The attenuation and broad-
ening are wavelength dependent and result from absorption
and multiscattering of light by water molecules and by ma-
rine hydrosols mineral and organic matter.
The extinction coefcient c of the aquatic medium is
governed by the absorption and scattering coefcients
and , respectively, and we have
9
c = + . 1
Figure 2 depicts the absorption, scattering, and extinction
coefcients for four types of waterpure sea water, clean
ocean water, coastal ocean water, and turbid harbor
waterat 520-nm wavelength.
6,10,14
It is clear that an in-
crease in the turbidity dramatically increases the extinction
coefcient, from less than 0.1 m
1
for pure water up to
more than 2 m
1
for turbid harbor water. However, the ab-
sorption coefcient increases more moderately than does
the turbidity.
The propagation loss factor as a function of wavelength
and distance z is given by
L
pr
, z = exp cz. 2
3 Communication Link Models
We now consider three types of communication links: the
line of sight, the modulating retroreector, and the reec-
tive. In addition, we perform a bit error rate BER calcu-
lation.
Turbid Harbor 2.19
Coastal Ocean 0.30
Clean Ocean 0.15
Extinction coefficient
(m
-1
)
Fig. 2 Absorption, scattering, and extinction coefcients for four
types of waterpure sea water, clean ocean water, coastal ocean
water, and turbid harbor waterat 520-nm wavelength.
Arnon: Underwater optical wireless communication network
Optical Engineering January 2010/Vol. 491 015001-2
Downloaded from SPIE Digital Library on 28 Jan 2010 to 132.72.138.1. Terms of Use: http://spiedl.org/terms
3.1 Line-of-Sight Communication Link
The most common link between two points in optical wire-
less communication systems is a line-of-sight LOS link as
illustrated in Fig. 3a. In this scenario, the transmitter di-
rects the light beam in the direction of the receiver. The
optical signal reaching the receiver is obtained by multiply-
ing the transmitter power, telescope gain, and losses and is
given by Ref. 11 as
P
Rlos
= P
T

R
L
pr
,
d
cos

A
Rec
cos
2d
2
1 cos
0

, 3
where P
T
is the average transmitter optical power,
T
is the
optical efciency of the transmitter,
R
is the optical ef-
ciency of the receiver, d is the perpendicular distance be-
tween the transmitter and the receiver plane, is the angle
between the perpendicular to the receiver plane and the
transmitter-receiver trajectory, A
Rec
is the receiver aperture
area, and
0
is the laser beam divergence angle. When the
transmitter beam divergence angle is very narrow

0
/ 20, Eq. 3 can be approximated as
P
Rlos
= P
T

R
L
pr
,
d
cos

A
Rec
cos
d tan
0

2
. 4
3.2 Modulating Retroreector Communication
Link
The modulating retroreector link
10
is used when one party
for example, a submarine has more resources another one
for example, a diver, as in Fig. 3b. In this case, the
submarine has more energy, payload, and lifting capacity
than the diver. Therefore it would be wise to put most of
the complexity and power requirement of the communica-
tion system into the submarine. In a modulating retroreec-
tor link, the interrogator sits at one end in our case, in the
submarine, and a small modulating optical retroreector
sits at the remote end. In operation, the interrogator illumi-
nates the retroreecting end of the link with a continuous-
wave beam. The retroreector inactively reects this beam
back to the interrogator while modulating the information
on it. The received power in this scenario is given by
P
RRetro
= P
T

Rec

Retro
L
pr
,
2d
cos

A
Retro
cos
2d
2
1 cos
0

A
Rec
cos
d tan
0retro

, 5
where
Retro
is the optical efciency of the retroreector,
is the angle between the perpendicular to the receiver plane
and the transmitter-receiver trajectory, A
Retro
is the retrore-
ectors aperture area, and
0retro
is the retroreectors
beam divergence angle.
3.3 Reective Communication Link
In some communication scenarios the line of sight is not
available due to obstructions, misalignment, or random ori-
entation of the transceivers.
11
To address this problem a
reective communication link could be used. In this case,
the laser transmitter emits a cone of light, dened by inner
and outer angles
min
and
max
, in the upward direction
Fig. 3c. Here
i
and
t
are the angles of incidence and of
transmission, respectively. The latter is derived from the
former using Snells law.
The light reaching the ocean-air surface illuminates an
annular area and is partially bounced back in accordance
with the reectivity. Since the refractive index of air is
lower than that of water, total internal reection TIR can
be achieved above a critical incidence angle. When the
transmitter is at depth h, the illuminated annular surface
with equal power density at depth x is given by
A
ann
= 2h + x
2
1 cos
max
1 + cos
min

= 2h + x
2
cos
min
cos
max
. 6
Equation 6 describes an annular area taken from a sphere
of radius h+x, which would have uniform power density in
free space.
If we model the ocean-air surface as smooth, then
=
i
, and we can derive the link budget by using the vari-
ables dened in Eq. 3. Then we can dene the auxiliary
function and calculate the received power as
Fig. 3 a The line-of-sight communication scenario. b The modu-
lating retroreector communication scenario. c The reection com-
munication scenario.
Arnon: Underwater optical wireless communication network
Optical Engineering January 2010/Vol. 491 015001-3
Downloaded from SPIE Digital Library on 28 Jan 2010 to 132.72.138.1. Terms of Use: http://spiedl.org/terms
f
Rref
=
P
T
cos
A
ann

R
L
pr
,
h + x
cos

1
2

tan
t

tan
t
+

2
+

sin
t

sin +
t

,
min

c
,

R
L
pr
,
h + x
cos
,
c

max
.

7
At the plane of the receiving sensor, node coverage is pro-
vided within an annular area bounded by radii h+x
tan
min
and h+xtan
max
. Equation 7 can be simplied
on the assumption that the receiver aperture is small rela-
tive to h+x, yielding the approximate received power as
P
Rref
A
Rec
f
Rref
. 8
3.4 Bit Error Rate Calculation
The simplest and most widespread modulation technique in
optical wireless communication is intensity-modulation,
direct-detection on-off keying OOK. In this technique, the
receiver is based on the emerging technology of silicon
photomultipliers SiPMs.
15
These photodetector devices
are fabricated in the form of arrays of photodiodes that are
operated in Geiger mode to create a photon-counting de-
vice. If we assume that a large number of photons are re-
ceived, then according to the central limit theorem, the
Poisson distribution can be approximated by a Gaussian
distribution and the BER is given by
11
BER=
1
2
erfc

r
1
T r
0
T

2r
1
T
1/2
+ r
2
T
1/2

. 9
Here r
1
=r
d
+r
bg
+r
s
and r
0
=r
d
+r
bg
, where r
d
and r
bg
repre-
sent the sources of additive noise due to dark counts and
background illumination, respectively, and
erfc =
2

exp
2
d. 10
4 Discussion and Numerical Example
The three types of link models could be used to design
sophisticated networks. It is clear that line of sight using
narrow beam divergence provides the maximum range;
however, in this case the precise locations of the two plat-
forms are required. On the other hand, when it is required
to simultaneously broadcast for example from a subma-
rine to several platforms UUVs or divers, for example,
the best option is to use LOS with a wide beam divergence.
However, if obstructions between the two platforms block
the line of sight, a reective communication link is pre-
ferred. When one party has more resources than the other
one in the link, the modulated retroreector is the best op-
tion.
In this section we simulate the performance of the three
links, using practical values for clean ocean water with an
extinction coefcient of 0.15 m
1
. The values of the simu-
lation parameters are given in Table 1. It is evident from
Fig. 4 that a single LOS underwater link using a pulse-
modulated laser transmitter and a SiPM detector array in
the receiver results in a considerably higher photon count
for a given sensor node separation than a reective or ret-
roreector link. For instance, for a node separation of 30 m,
8,000 photons would be received from a signal in a LOS
link, 2 photons would be received from a retroreector link,
and only 10 would be received in a reective link where the
transmitter depth is 20 m and the receiving nodes are also
at a depth of 20 m. However, if a single point-to-point link
were to fail, the transmitted signal would be lost, while in
the reective underwater network solution a number of
nodes would be expected to receive the signal. Even in the
severe case where several nodes fail, with sufcient node
redundancy there would still be additional nodes that could
relay the signal further.
In Fig. 5 we can see that BER values of 10
4
are ob-
tained for a reective link when the node separation is
40 m, while a BER of 10
4
could be achieved in a LOS link
and a retroreector link when the node separation is 60 m
and 50 m, respectively. From this result it is easy to under-
stand that acceptable BER performance could be achieved
for short ranges on the order of tens of meters for all three
models.
In Fig. 6 we compare the numbers of photons received
for a link operated in turbid harbor water for two cases: a
when only absorption is considered and b when absorp-
tion and scattering are considered. From this gure it is
easy to see that in the absorption case the number of re-
ceived photons reduces from 10
5
to 1 for increases in dis-
tance separation from 1 to 65 m, while in the case of ab-
Fig. 4 Graph showing number of received photons as a function of
transmitter-receiver separation for clean ocean water with extinction
coefcient equal to 0.15 m
1
.
Arnon: Underwater optical wireless communication network
Optical Engineering January 2010/Vol. 491 015001-4
Downloaded from SPIE Digital Library on 28 Jan 2010 to 132.72.138.1. Terms of Use: http://spiedl.org/terms
sorption and scattering the number of received photons
reduces from 10
5
to 1 for increases in distance separation
from 1 to 8 m This result indicates that receiving more
scattered light and performing the required signal process-
ing in the time domain could dramatically improve the per-
formance of an optical wireless system in turbid water.
5 Summary and Conclusions
The results presented indicate that networks based on un-
derwater optical wireless links are feasible at high data
rates for medium distances, up to a hundred meters. Such
networks could serve subsea wireless mobile users. In ad-
dition, by placing multiple relay nodes between the chief
network nodes, messages could traverse very long distances
despite severe medium-induced limitations on the transmis-
sion ranges of individual links. Additional improvements to
the availability of the network could be achieved by a hy-
brid communication system that would include an optical
transceiver and an acoustical transceiver. A hybrid commu-
nication system can provide high-data-rate transmission by
using the optical transceiver. When the water turbidity is
high or the distance between the terminals is large, the sys-
tem can switch to a low data rate using the acoustic trans-
ceiver, thereby increasing the average data rate and avail-
ability. However, the complexity and cost of the system are
increased. In this kind of system, smart buffering and pri-
oritization could help to mitigate short-term data rate reduc-
tion.
Many aspects of the proposed system remain to be in-
vestigated; for example, rigorous modeling of the reective
nature of the ocean-air surface, including ocean surface
roughness, as well as solar radiance penetration. Extensive
studies should be made of the nature of multiple scattering
in different oceanic channels and the limitation of the
Table 1 Parameters used in numerical calculations. We assume

retro
is much greater than the diffraction-limited divergence angle.
Parameter Typical value
Extinction coefcient, clear ocean
m
1

0.1514
Refractive index 1.33643
Critical angle deg 48.44
Transmission wavelength nm 532
Optical efciency of retroreector 0.9
Optical efciency of transmitter 0.9
Optical efciency of receiver 0.9
Average transmitter power W 10
Pulse duration ns 1
Data rate Mbit/s 0.5
Receiver aperture area m
2
0.01
Retroreector aperture area m
2
0.01
Retroreector beam divergence
retro
deg
10
Laser beam divergence angle
0
deg 68
Transmitter inclination angles
min
,
max
deg
0, 68
Dark counting rate MHz 1
Background counting rate MHz 1
Counting efciency % 16
Transmitter depth h m 20
Receiver depth x m 20
Fig. 5 Graph showing BER as a function of transmitter-receiver
separation for clean ocean water with extinction coefcient equal to
0.15 m
1
.
Fig. 6 Graph showing number of received photons for line-of-sight
scenario as a function of transmitter-receiver separation for two
cases: absorption and extinction.
Arnon: Underwater optical wireless communication network
Optical Engineering January 2010/Vol. 491 015001-5
Downloaded from SPIE Digital Library on 28 Jan 2010 to 132.72.138.1. Terms of Use: http://spiedl.org/terms
modulating retroreector range due to light backscattered
into the receiver before reaching the retroreector. Some
seminal theory necessary to describe spatial spreading of an
optical beam in the presence of scattering agents under wa-
ter was presented in Ref. 16.
Future work on these subjects should rene the analysis
and yield more accurate numerical results. Additional open
issues to be addressed at higher layers of the network de-
sign include multiple access, such as wavelength division
multiplexing WDM at blue-green wavelengths, and code
division multiple access CDMA or clustering. However,
the fundamental concept has been shown to be feasible and
practical.
References
1. I. F. Akyildiz, D. Pompili, and T. Melodia, Underwater acoustic
sensor networks: research challenges, Ad Hoc Networks 33, 255
256 2005.
2. J. Heidemann, W. Ye, J. Wills, A. Syed, and Y. Li, Research chal-
lenges and applications for underwater sensor networking, in Proc.
IEEE Wireless Communications and Networking Conf., pp. 228235
2006.
3. T. Dickey, M. Lewis, and G. Chang, Optical oceanography; recent
advances and future directions using global remote sensing and in
situ observations, Rev. Geophys. 441, RG1001 2006.
4. C. Detweiller, I. Vasilescu, and D. Rus, AquaNodes: an underwater
sensor network, in Proc. Second Int. Workshop on Underwater Net-
works, pp. 8588, IEEE 2007.
5. F. Hanson and S. Radic, High bandwidth underwater optical com-
munication, Appl. Opt. 472, 277283 2008.
6. B. Cochenour, L. Mullen, and A. Laux, Spatial and temporal disper-
sion in high bandwidth underwater laser communication links, in
Proc. IEEE Military Communications Conf., pp. 17 2008.
7. J. H. Smart, Underwater optical communication systems part 1: vari-
ability of water optical parameters, in Proc. IEEE Military Commu-
nications Conf., pp. 11401146 2005.
8. J. W. Giles and I. N. Bankman, Underwater optical communications
systems part 2: basic design considerations, in Proc. IEEE Military
Communications Conf., pp. 11401146 2005.
9. D. Kedar and S. Arnon, Subsea ultraviolet solar-blind broadband
free-space optics communication, Opt. Eng. 484, 046001 2009.
10. L. Mullen, B. Cochenour, W. Rabinovich, R. Mahon, and J. Muth,
Backscatter suppression for underwater modulating retroreector
links using polarization discrimination, Appl. Opt. 482, 328337
2009.
11. S. Arnon and D. Kedar, Non-line-of-sight underwater optical wire-
less communication network, J. Opt. Soc. Am. A 263, 530539
2009.
12. D. Kedar and S. Arnon, Optical plankton: an optical oceanic probing
scheme, J. Appl. Remote Sensing 1, 013541 2007.
13. S. Jaruwatanadilok, Underwater wireless optical communication
channel modeling and performance evaluation using vector radiative
transfer theory, IEEE J. Sel. Areas Commun. 269, 16201627
2008.
14. R. P. Bukata, J. H. Jerome, K. Y. Kondratyev, and D. V. Pozdnyakov,
Optical Properties and Remote Sensing of Inland and Coastal Wa-
ters, CRC Press, Boca Raton, FL 1995.
15. P. Eraerds, M. Legre, A. Rochas, H. Zbinden, and N. Gisin, SiPM
for fast photon-counting and multiphoton detection, Opt. Express
1522, 1453914549 2007.
16. B. M. Cochenour, L. J. Mullen, and A. E. Laux, Characterization of
the beam-spread function for underwater wireless optical communi-
cations links, IEEE J. Ocean. Eng. 334, 513521 2008.
Shlomi Arnon is a faculty member in the
Department of Electrical and Computer En-
gineering at Ben-Gurion University, Israel.
There, in 2000, he established the Satellite
and Wireless Communication Laboratory,
which has been under his directorship since
then. During 19981999 Professor Arnon
was a postdoctoral associate Fulbright Fel-
low at LIDS, Massachusetts Institute of
Technology MIT, Cambridge, USA. His re-
search has produced more than fty journal
papers in the area of satellite, optical, and wireless communication.
During part of the summer of 2007, he worked at TU/e and Philips
Lab, Eindhoven, Nederland, on a novel concept of a dual commu-
nication and illumination system. He was a visiting professor during
the summer of 2008 at TU Delft, Nederland. Professor Arnon is a
frequent invited speaker and program committee member at major
IEEE and SPIE conferences in the USA and Europe. He was an
associate editor for the Optical Society of Americas Journal of Op-
tical Networks for a special issue on optical wireless communication
that appeared in 2006, and is now on the editorial board for the
IEEE Journal on Selected Areas in Communications for a special
issue on optical wireless communication. Professor Arnon continu-
ously takes part in many national and international projects in the
areas of satellite communication, remote sensing, and cellular and
mobile wireless communication. He consults regularly with start-up
and well-established companies in optical, wireless, and satellite
communication. In addition to research, Professor Arnon and his
students work on many challenging engineering projects with espe-
cial emphasis on the humanitarian dimension. For instance, a long-
standing project has dealt with developing a system to detect human
survival after earthquakes, with an infant respiration monitoring sys-
tem to prevent cardiac arrest and apnea, and with detection of falls
in the case of epilepsy sufferers and elderly people.
Arnon: Underwater optical wireless communication network
Optical Engineering January 2010/Vol. 491 015001-6
Downloaded from SPIE Digital Library on 28 Jan 2010 to 132.72.138.1. Terms of Use: http://spiedl.org/terms