Underwater Wireless Sensor Networks: A Review

Mohammad Alsulami

, Rafaat Elfouly

and Reda Ammar

University of Connecticut, Storrs, CT, U.S.A.

Rhode Island College, Providence, RI, U.S.A.

Keywords:

Underwater Wireless Sensor Networks, Topologies, Deployment Strategies.

Abstract:

Several disciplines like science, engineering, and biological industry have been inﬂuenced by sensor networks

which have brought sensing and computation into reality. The possibility of having these sensors physically

assigned close to the target whose parameters are to be observed enables remote monitoring of various as-

pects of the physical world. Wireless channeling of information beneath the ocean or generally underwater

has provided the best technological ways of oceanic observations. Ocean bottoms have been monitored tra-

ditionally by deploying oceanographic sensors that obtain information at distinct and ﬁxed ocean zones. The

oceanographic instruments are then recovered when the tasks are completed. This implies that data cannot

be monitored remotely since there is no collaborative communication of obtained data between the collection

point and the monitoring end. The data recorded can also be destroyed in case of a non-successful mission.

Oceanic observations have been made primarily possible by sensor networks carefully laid out under the wa-

ters. Underwater sensor networks can also be achieved wirelessly by establishing communications between

sensors and monitors without major cabling. These are known as Underwater Wireless Sensor Networks

(UWSNs). The UWSNs are comprised of various gadgets like vehicles that can operate autonomously under

the water and sensors. Deployment of these gadgets is done in targeted acoustic zones for the collection of

data and monitoring tasks. Bilateral communication is established between stations based on the ground and

different UWSNs nodes. This enables instantaneous remote monitoring and communication of information

from the speciﬁed oceanic zones to engineering personnel based on the shores. This paper looks at the various

aspects of Underwater Wireless Sensor Networks UWSNs including their importance, applications, network

architecture, requirements, and challenges and in their deployments.

1 INTRODUCTION AND

MOTIVATIONS

Emerging technologies around vehicles that are au-

tonomous and sensor deployment capabilities have

motivated the scheme of networking of sensors un-

der the water. Although this comes with communica-

tion problems, technology on acoustics can be ma-

nipulated to turn the scheme into practicality. For

short-range links, techniques of communication that

are interdependent have been advanced(Amoli, 2016).

These techniques include radio frequency, optical,

and communications that are electrostatic. These are

utilized for ranges between 1m to 10m to exploit the

high bandwidths involved. Large and highly pow-

ered antennas are required for such high bandwidth

signals because of their high attenuation tendencies.

Longer ranges of communication are affected mainly

by sounds of low speed, propagation of diverse paths

that vary in time, and bandwidths that are limited

and depend on distance. Propagation through diverse

paths varying in time coupled with the fact that sounds

have low speeds plus bandwidths that are limited and

dependent on distance brings ramiﬁcations of sub-

standard quality and high throughput(Felemban et al.,

2015; Lloret, 2013).

With advances in microelectromechanical devices,

sensors can be miniaturized with low power consump-

tion, thus increasing the capabilities of processing and

operation in different underwater scenarios. Integra-

tion of internet protocols with UWSNs architectures

can support M2M and IoT structures for up-to-date

monitoring of underwater events. The USWNs nodes

have also been found to be widely compatible with a

variety of sectors. However, the demands and con-

ditions of UWSNs that include accessibility, environ-

mental friendliness, endurance, privacy, and complex-

ity make the acoustic networks of UWSNs distinctive

and difﬁcult to apply practically for developers. Mo-

202

Alsulami, M., Elfouly, R. and Ammar, R.

Underwater Wireless Sensor Networks: A Review.

DOI: 10.5220/0010970700003118

In Proceedings of the 11th International Conference on Sensor Networks (SENSORNETS 2022), pages 202-214

ISBN: 978-989-758-551-7; ISSN: 2184-4380

tivations for UWSNs include (Felemban et al., 2015;

Chandrasekhar et al., 2006):

• The architecture is self-diagnosing since interac-

tive anomalies can be sensed and corrected during

the acquisition of data(Lloret, 2013).

• Allows analyzing the data during data acqui-

sition; therefore, adjustments can be made on

sensors inputs and outputs for desired observa-

tions(Felemban et al., 2015; Amoli, 2016).

• Contrary to cables deployed underwater, UWSNs

offer a distributed structure that allows for ap-

plication to various frameworks. This makes

the UWSNs architectures appreciably ﬂexible for

the provision of support in underwater surround-

ings(Chandrasekhar et al., 2006).

• The cost of deployment is low compared to cabled

sensor networks(Felemban et al., 2015).

UWSNs importance is widely seen in applications

that are swiftly gathering favor in the enablement

of progress around the disciplines of oceanic obser-

vations and monitoring systems. Further, UWSNs

has gained ground in surveillance of deep seas,

earthquake and tsunami monitoring and forecasting,

aquatic animal’s activities and plants tracking, re-

connaissance and surveillance by the military, ex-

traction of gas and oil, monitoring of leaking and

spillage of oils, detection of mines and also in ﬁsh-

eries(Felemban et al., 2015; Amoli, 2016; Chan-

drasekhar et al., 2006). These applications can be

classiﬁed in terms of monitoring aspects, military as-

pects, navigation, and forecasting of disaster occur-

rences and their management. These classes of ap-

plication can then be classiﬁed into respective sub-

classes(Lloret, 2013). Monitoring underwater activi-

ties involves deploying sensors networked to track the

surrounding environment related to its properties and

features of interest. Here, the tangible and substan-

tial environment is monitored. This monitoring ap-

proach of classiﬁcation can also be subdivided into

monitoring the quality of water, aquatic life moni-

toring, and sub-water exploration monitoring. The

quality of water is monitored with regard to the well-

being of aquatic life such as ﬁsh and other living

things above the water. Monitoring of underwater en-

vironment can range from applications to large oceans

to canals. The farming of ﬁsh is considered one of

the best sources of economic growth(Felemban et al.,

2015; Amoli, 2016; Chandrasekhar et al., 2006).

Thus there is a need for uncompromising monitor-

ing of the ﬁsh habitats for a near-perfect living condi-

tion. Deployment of sensors has been accomplished

to monitor the environmental conditions of individ-

ual ﬁsh species to create a conducive environment for

ﬁsh farming. In (Felemban et al., 2015; Lloret, 2013),

UWSNs have been utilized to assess the quantity of

food that has not been eaten and the capital waste on

the farm(Felemban et al., 2015). The design has been

deployed for testing in ﬁsh farming of marine ﬂora

and fauna. A group of wireless network of sensors

was used for precise analytical monitoring of farm

water pollution. Applications have also been devel-

oped for trout farming to keep track of water quality

in the pools. The chemical composition of the water

is monitored for a given period, and an algorithm is

deployed to show the input-output information for a

given number of pools(Felemban et al., 2015; Amoli,

2016; Chandrasekhar et al., 2006).

The algorithm is used to provide alerts in case the

measured conditions surpass the set threshold. These

conditions include conductivity of electricity, oxygen,

and ammonium nitrogen levels. Autonomous vehi-

cles can also be integrated with sensors to collect sam-

ples of water from reservoirs of drinking water. The

processed information is then transmitted to a loca-

tion remotely(Lloret, 2013; Felemban et al., 2015).

UWSNs can also be applied in monitoring river pH

composition by utilizing sensor nodes. These nodes

are made to process the collected data. A compar-

ison is then made between the threshold parameter

values and the collected data. Wireless modules are

then utilized to transmit the processed information to

a remotely located station. Other parameters that can

be monitored in water include temperature, oxygen

dissolved, and pollution(Amoli, 2016). Temperature

sensors are deployed under the water, and the val-

ues are communicated wirelessly and then displayed

in tabular and graphical presentations. The visualiza-

tion of the collected information from under the wa-

ter can be through web applications or cellular plat-

forms. Mostly, ﬂoating nodes with GPRS convey the

underwater information to a remotely located station.

Sensors can be deployed under the water to monitor

pollutants such as wreckages and spillages to monitor

water pollution(Chandrasekhar et al., 2006; Felem-

ban et al., 2015). Monitoring of water habitats in-

cludes applying UWSNs to monitor the coral reefs or

any other plants under the water and monitoring the

activities of marine animals/organisms(Lloret, 2013;

Chandrasekhar et al., 2006).

UWSNs can be applied to monitoring marine ac-

tivities by monitoring the surrounding environment

of the marine organisms through acoustic device net-

works. To transmit the marine environmental con-

ditions, information from one sensor to another can

be accomplished by utilizing TDMA protocols. The

obtained data can then be used to develop ecosys-

tem models capable of predicting changes in the un-

Underwater Wireless Sensor Networks: A Review

203

derwater environment and changes in climate con-

ditions(Almutairi and Mahfoudh, 2017; Felemban

et al., 2015). Currently, deployment of such a sys-

tem has been accomplished in Queensland, Australia,

with sensing capabilities, wireless conveyance of in-

formation to a ground remotely located station, visu-

alization of measured data, and an alert system in case

of anomalies have been deployed. In this scheme, a

rough and ready system that uses a cluster or a topol-

ogy of stars for direct transmission of information to

a buoyant gateway has been used. Data is then trans-

mitted remotely to a control station from the gate-

way. Seashells are also monitored as part of marine

habitat in China, such as Zhejiang province(Almutairi

and Mahfoudh, 2017). UWSNs have an application

in monitoring seismic activities such as oil extrac-

tion from ﬁelds under the water(Aldosari et al., 2020).

Fluctuation of the oil reservoir is studied over time

in a 4-D model for evaluation of the performance of

the oil ﬁelds and application of ad hoc interventions.

Traditionally, onshore ﬁelds are continuously moni-

tored through daily, quarterly, or annual surveillance

through permanent instruments(Aldosari et al., 2020;

Almutairi and Mahfoudh, 2017). Conversely, sub-

merged oil ﬁelds are more demanding because the de-

ployment of sensors is not presently permanent in the

oil ﬁelds underwater. Therefore, monitoring of tec-

tonic ﬁelds is accomplished using ships. The ships are

ﬁxed with an array of sonar as sensors and actuators

made of air cannons(Aldosari et al., 2020; Almutairi

and Mahfoudh, 2017). Evaluating such submerged oil

ﬁelds is highly costly and can only be applied to sub-

merged ﬁelds less frequently. Multiple channels of

tectonic data are collected using individual sensors. A

single sensor is designed to collect 24 bits per channel

at a frequency of 500 Hertz. Data is captured when

an event of seismic nature occurs, consequently giv-

ing 60-kilo bytes of features of data for every sensor

used in a tectonic event. The data rate of 5-kilo bytes

per second is involved for 120s per sensor to transmit

the information for a single hop. For oil ﬁelds that

cover the typical zones of 8km by 8km, sensors are

deployed approximately at a 50m grid(Aldosari et al.,

2020), implying that a highly larger number of sen-

sors are utilized to provide total coverage. A layered

communication channel is also implied because some

nodes of sensors are connected through communica-

tion media that are non-acoustic. Buoys augmented

with high-speed Radiofrequency capabilities are uti-

lized for the wireless transmission of data. Supern-

odes can also be implemented for a ﬁve-by-ﬁve net-

work segment. All sensors are in the vicinity of two

jumps around a supernode, and retrieval of all infor-

mation can be accomplished within an hour(Aldosari

et al., 2020; Almutairi and Mahfoudh, 2017).

Application of UWSNs in exploration involves in-

spection and survey of a substantial amount of sub-

merged minerals(Almutairi and Mahfoudh, 2017; Al-

dosari et al., 2020). The exploration under the wa-

ter can further be classiﬁed into exploring innate re-

sources and monitoring of pipelines and cables for

gas and oil exploration under the water. UWSNs have

been developed in (Felemban et al., 2015) for the lo-

cation and investigation of cast iron crust(Almutairi

and Mahfoudh, 2017). An autonomous underwater

vehicle that is operated remotely has been utilized on-

site to explore the crust minerals attaining a depth of

3000 meters under the ocean surface. The vehicle is

augmented with a system that can do mapping of the

ocean bed. Therefore, it is an emerging technology

to combine an acoustic network of sensors and re-

motely operated underwater vehicles to excavate min-

erals buried deep into the ocean beds(Almutairi and

Mahfoudh, 2017; Aldosari et al., 2020).

Underwater sensor systems can be utilized in the

detection of various facets of applications in military

aspects. Here, AUVs are equipped with sonars for

imaging, detectors for metals, and cameras for mis-

sions in detecting submerged ocean mines(Almutairi

and Mahfoudh, 2017; Felemban et al., 2015). This

setup is also used for port securities(Aldosari et al.,

2020; Lloret, 2013). Submarines are usually aug-

mented with several underwater architectures of sen-

sors to aid in surveillance and monitoring. Detect-

ing submerged mines ensures that military ships ex-

ecute risk-free voyages(Khan et al., 2021; Aldosari

et al., 2020; Chandrasekhar et al., 2006). Since the

mines are composed of unique materials that are fer-

rous, they can be separated from clutter under the

ocean since these clutters are usually nonmetallic.

The application of UWSNs in surveillance encom-

passes detection of foreign intrusion in demarcated

international water spaces such as warships from a

deﬁned enemy. The UWSNs not only assist in detec-

tion and surveillance but can also be utilized for nav-

igation purposes. Since the underwater topology is

exceedingly uneven, arbitrary, untraversed, and expo-

nentially increasing in depth, UWSNs come in handy

in navigation. This provision of navigation capabil-

ity can be extended to civilian ships, boats, and even

ﬁshing vessels. Because of the different transmis-

sion mediums involved, the navigation devices used

on land cannot be used under the sea or ocean, further

cementing the importance of a setup of an underwater

network of sensors(Khan et al., 2021; Almutairi and

Mahfoudh, 2017).

SENSORNETS 2022 - 11th International Conference on Sensor Networks

204

2 UNDERWATER WIRELESS

SENSOR NETWORKS

ARCHITECTURES

Generally, the performance of acoustic networks is

dependent on the design of the topology. Effective

topology increases the reliability of the network of the

sensors and vice versa. Additionally, energy is con-

served in an efﬁcient topology of underwater acoustic

sensors. Therefore, an optimized network model of

UWSNs must be adopted depending on the area of

deployment(Alharbi et al., 2015). Different methods

of classifying UWSNs include classiﬁcation accord-

ing to the mobility of the nodes, i.e., half-static, fully

static, and dynamic models, and classiﬁcation accord-

ing to node depth, i.e., 2D model and 3D model struc-

tures(He et al., 2018; Alsulami et al., 2020b). Since

attenuation is experienced underwater during radio

frequencies, communication between UWSNs must

be accomplished using acoustic mode. The acoustic

linking has been optimized over the years to mitigate

challenges such as delays that are longer, high dispar-

ities, and sensing abilities that are unstable(Alharbi

et al., 2014; Alsulami et al., 2019). Therefore, the

development of efﬁcient topology of sensor networks

has been inﬂuenced by motivating factors such as dif-

ﬁculties presented by compounded underwater sur-

roundings and the need for sensing in marine mon-

itoring applications. In the process of development

for the progress of the 2D model environment to

3D structures evolution, a myriad of UWSNs were

designed with regard to different applications after

proper optimization techniques have been utilized to

improve the performance of the topology conﬁgura-

tion. Hence, formally and generally, it is not wrong to

classify the topologies into 2D, 3D, 4D, and UWSNs

based on vehicles operating autonomously underwa-

ter(He et al., 2018).

2.1 2D Network Models

In an ordinary 2D topology, the nodes of the sen-

sors, which have a smaller size due to unique un-

derwater surroundings, are set upon the ocean bed.

The 2D structure models can be classiﬁed into iso-

morphic and heterogeneous topologies. Initially, ma-

rine investigations are directed towards the coastal ar-

eas. For these observations of marine interests, sen-

sors of similar functionalities are placed on the ocean

bed or seabed to extract data and detection function-

alities. Sinks on the surface can be utilized for the

acquisition of information from nodes under the wa-

ter. This is made possible by an acoustic channel

link. Sinks under the water acquire information from

deep-sea sensors and relay it to ground-based con-

trol stations. Therefore, it is necessary to provide

sinks under the water and transceivers that are acous-

tic. These transceivers should be vertical and hori-

zontal. The horizontal transceivers have the purpose

of communicating with nodes regarding data collec-

tion or command provision as they are being received

from the control station offshore. The data is sent

to the control station on land using the transceivers

that are vertical. The vertical transmitter receivers

must be of sufﬁcient range since the ocean depths

are large. The sinks placed on the surface are sup-

plied with RF transmitters of a wide range and also

acoustic transmitter-receivers for the management of

communications that are parallel by design. Models

that allow for the application of two-dimensional de-

tection at ocean beds with similar detection capabil-

ities in every sensor placed for detection are known

as isomorphic models, as shown in ﬁgure 1 (He et al.,

2018). In the isomorphic model, data is sent to the

Figure 1: A 2D architecture of isomorphic model.

sinks from the sensors. From the sinks, the infor-

mation is then relayed to the control station located

on land. The isomorphic structure is best applied

for detection in shallow areas because of communica-

tion limitations in sensor distances(Alhumyani et al.,

2015; He et al., 2018). To improve model perfor-

mance for marine areas far from the coastal lines, ca-

ble tethering is utilized to connect sinks placed at the

surface and sinks placed under the water. This setup

brings about convenience and improves communica-

tion reliability under the water. Tethering also guaran-

tees the exchange of information in a reliable fashion

between sensors with a simple, convenient deploy-

ment(He et al., 2018). However, the model is suitable

only for comparatively stable networks. This is only

possible using routing structures that do not demand

time to time rationalization for Shallow Ocean water

surrounding.

Additionally, the model is optimized for localized

monitoring of the environment and cannot perform

well for spaces of large expanse. It is therefore in-

effective in areas deemed as deep regions of ocean or

Underwater Wireless Sensor Networks: A Review

205

sea. For the sinks under the water, the collected in-

formation by the sensors is often relayed by utilizing

multi-jump routing. In this model, there are sensors

close to the sinks that are submerged under the wa-

ter(Khan et al., 2021). These sensors are designed to

transmit the collected information and also function

as communication nodes during heavy loads. Due to

the homogeneous capacity of isomorphic model sen-

sors, there is a high-energy consumption by the nodes

and thus a high failure rate. The intersymbol interfer-

ence phenomenon is also likely to occur due to several

sensors sending data at similar times(Felemban et al.,

2015; He et al., 2018). It is inconvenient to deploy

sinks in some regions. Therefore sensors located at

the bottom of the sea are constrained to relay infor-

mation to the sinks at the surface by utilizing audial

communications. This results in high energy demand

by the nodes(He et al., 2018). To solve the above

problems and challenges, a heterogeneous model can

be utilized for such regions. The heterogonous model

incorporates aquatic sensing and a gateway conﬁgu-

ration of center nodes known as hub sensor nodes.

In addition to monitoring the ocean bed, the gate-

way is also utilized to transmit the collected infor-

mation to the sinks at the surface. The gateway has

acoustic transmitter-receiver modules, namely, hori-

zontal and vertical structures. The horizontal conﬁg-

uration enables interaction among the nodes in a clus-

ter. It releases directive information on conﬁguration,

and at the same time, it collects the data sensed(Khan

et al., 2021; Aldosari et al., 2020). The data is sent to

the control station on land using the transceivers that

are vertical. Therefore, this model solves the issue

of rapid consumption of energy in the sensor nodes

for such a conﬁguration. The conﬁguration also re-

lieves the cost of deploying wired cables and there-

fore utilizes the ﬂexible wireless transmission of data.

It is essential to consider the gateway number and

the position of the aquatic clusters when constructing

this model. Additionally, there is a resultant interfer-

ence and decrease in transmission of standard packets

due to the execution of a horizontal collection of in-

formation and vertical forwarding of data at similar

times(He et al., 2018).

2.1.1 Mechanisms of Conﬁguration in 2D

Models

The network conﬁguration is perceived here as the

placement of sensors under the sea or ocean waters to

construct a reliable and stable topology. The deploy-

ment processes in the 2D model conﬁguration are the

deterministic and random approaches(Alsulami et al.,

2020b).

• Deterministic Approach

In the deterministic deployment approach, the state

of the environment being monitored is considered,

and the positions of sensors are organized reason-

ably. Arranging the sensors well involves consis-

tently deploying these sensors for complete cover-

ages and deploying non consistently for occasion-

oriented coverages(Alsulami et al., 2019). For com-

plete coverage of the area being monitored using

a minimum number of sensors, they should be de-

ployed uniformly to minimize superimposed areas

between the sensors. Triangular conﬁguration de-

ployment techniques have been proposed in litera-

ture where the two-dimensional monitored regions

are split into various equal-sided triangles(Alhumyani

et al., 2015). Deployment of sensors is then per-

formed at triangle vertices to realize enormous cover-

age under minimum sensors. The sides of the equal-

sided triangles can be adjusted to achieve maximum

coverage with the assurance that minimal overlapping

regions between sensors will occur(Alsulami et al.,

2020a). Grids of hexagonal or square shapes have

also been proposed as construction structures for two-

dimensional networks. These can be applied to spa-

tially and open deterministic planes of 2Ds. Sensors

can be deployed non-uniformly to tackle the problem

of amorphous areas with several obstacles(He et al.,

2018).

• Random Approach

Random approach deployment is utilized when the

marine surrounding being monitored has a wide vari-

ation of conditions that are not easy to explore com-

pletely. This type of deployment involves sparsely

deploying nodes at distinct locations and intensively

deploying at a large scale(He et al., 2018; Alsulami

et al., 2020a). Sensors are typically deployed ran-

domly in harsh underwater environments by throw-

ing them out of a ship or plane. Algorithms utilized

in deploying randomly consider coverage of network

and connectivity for the realization of complete sen-

sor coverage. However, more sensors are required

for complete coverage with a resultant increase in the

cost of deployment and maintenance(Alsulami et al.,

2020b). Only key regions must be considered for efﬁ-

cient coverage that guarantees less number of utilized

sensors(Alsulami et al., 2020a).

• Clustered Approach Deployment

Here, clustering of sensors is done based on a single-

hop from a gateway made of cluster nodes to a sink

on the ocean surface. Sensors can also be clustered

in a multi-jump communication method. In clustered

approach, algorithms are utilized for the placement of

sensors according to regular grids. This technique of

SENSORNETS 2022 - 11th International Conference on Sensor Networks

206

deployment cannot, therefore, be applied to complex

environments(He et al., 2018; Alsulami et al., 2021).

In this approach, the selection of position is taken into

consideration when clustering unrelated sensor nodes.

Optimization of the position and quantity of nodes

is easy since there is no distinct deﬁnition of topol-

ogy(Alsulami et al., 2020a).

2.1.2 Strategies of Optimization in 2D Models

Mostly, the methods applied for optimization to im-

prove execution in 2D models consider the adjust-

ment of power, modeling of the graphics, and nu-

merous coverages(He et al., 2018). This enhances

the network’s performance by improving connectiv-

ity, coverage and minimizing the required quantity of

nodes. The energy or power consumption pertaining

to the network can also be reduced by optimization

techniques that consider the following: control of fre-

quency, optimization through interpolation, and clas-

siﬁcation of clusters(Alharbi et al., 2015). Delay of

transmission is also reduced, and data quality is im-

proved when execution optimization techniques are

utilized. The aim of optimizing the topology is to

stabilize the network and enhance its performance.

These optimization techniques seek to improve the

network parameters like the interval from one node

to the next, the number of nodes, and how the nodes

are linked(Alsulami et al., 2019).

2.2 3D Network Models

In this model, nodes are deployed in a ﬂoating manner

at varying ocean depths(Alhumyani et al., 2015). The

nodes are anchored at the bottom of the ocean with

wires attached to the anchors to control node depths.

Buoys that are placed horizontally on a plane can also

be utilized in controlling the node depths. The 3D

network structure can be classiﬁed into isomorphic

and heterogeneous. The isomorphic can further be

sub-classiﬁed into dynamic and static design mod-

els(Alhumyani et al., 2015). The static structure de-

signs are placed at various depths to cover large areas

of the region being monitored. Connecting pressure-

sensitive nodes accomplish the varying depth to an

anchor chain or a buoy through a cable whose length

can be adjusted. Anchoring the buoys to the ocean

bottoms mitigates the challenges of exposure to en-

emy targets and navigation problems. Deployment in

the 3D structure of the isomorphic model is as shown

in ﬁgure 2 (He et al., 2018; Ibrahim et al., 2009; Al-

humyani et al., 2015).

The challenges experienced by the 2D models

are avoided in 3D structures, enabling deployment in

complex harsh, and extremely deep regions of the sea

Figure 2: A 3D architecture of UWSNs (Bhaskarwar and

Pete, 2021; He et al., 2018).

or ocean(He et al., 2018; Alsulami et al., 2019). Some

of the drawbacks of 3D environments include: the po-

sitions to anchor the sensors must be initially deter-

mined in the early stages of deployment, the length of

the cable limits the distance of adjustment in the verti-

cal orientation, the energy demand by the sensors de-

termines how frequently the adjustments of the depths

should be done. The anchors that have been used to

hoist the sensors have also been found to be sensitive

to currents from the ocean, basin-scale vortex, plus

various factors of the environment. Quick energy con-

sumption is also experienced when sending data from

sensors at the bottom of the ocean to sinks at the sur-

face because multi-jump routing and nodes of isomor-

phic structures are utilized as relays(Bhaskarwar and

Pete, 2021; He et al., 2018).

3D UWSN model outputs are improved by

augmenting their performance with vehicles

operating autonomously underwater known as

AUVs(Bhaskarwar and Pete, 2021). Dynamic 3D

structure model based on AUVs is as shown in ﬁgure

3 (He et al., 2018).

Figure 3: Dynamic 3D Model based on AUVs (Bhaskarwar

and Pete, 2021).

The AUVs are used for supporting the GPS mod-

ules for tracking of position. AUVs perform analy-

sis of the ﬂow rate with an allowance of controllable

operations since they have a high computation capac-

ity. The AUVs have an adjustable speed of movement

and position. This model is challenging to apply in

large-scale proportions because of the high cost in-

Underwater Wireless Sensor Networks: A Review

207

volved. Consequently, 3D models that are dynamic

and heterogeneous have been proposed and designed

to lower the costs of constructing UWSNs.

2.2.1 Mechanisms of Conﬁguration in 3D

Models

There are different strategies of conﬁguration for var-

ious types of 3D models. The techniques of de-

ployment can be classiﬁed based on node character-

istics. For three-dimensional models that are con-

sidered static, anchoring of every sensor is done on

the ocean bead, and there is a restriction of indepen-

dent movement(Alsulami et al., 2019). The deploy-

ment methods can be classiﬁed as deterministic, self-

adaptive, and force-based virtual deployment (Zhang

et al., 2019).

• Deterministic Approach

This deployment can be classiﬁed as uniform and

non-consistent. It is necessary that UWSNs give a 3D

monitoring of the underwater surrounding, but there

are autonomous movement limitations because of en-

ergy requirements. A minimum number of sensors

should also be utilized to realize maximum place-

ment in a large area(Zhang et al., 2019; Alhumyani

et al., 2015). Therefore, it has been proposed that a

sequence of polyhedral top-up designs with regards

to conjecture by Kelvin be utilized(Alhumyani et al.,

2015).

• Self-adaptive Approach

This deployment technique involves random anchor-

ing of sensors on the bottom of the sea. The depth

of the sensors is also determined randomly(Alsulami

et al., 2020a). The adjustment made on the anchor

length selects the desired depths of the sensors. Ev-

ery individual sensor then relays its ultimate location

to a station onshore. Algorithms have been proposed

under this approach, where the clustering of sensors

is done by controlling the depth of sinking of indi-

vidual sensors(Alsulami et al., 2019). The algorithm

works along with the idea that coverage under 3D

structure ensures continuous interconnection between

the sensors. For the avoidance of superimposition in

horizontal regions, preliminary planning of the posi-

tion of sensors is needed. Since this deployment is

random-based, numerous replicated sensors should be

distributed so that the region being monitored can be

covered completely(Zhang et al., 2019; Alhumyani

et al., 2015).

3 APPLICATIONS OF UWSNs

TECHNOLOGY

UWSN technology ﬁnds its applications in many ar-

eas grouped into three categories: Scientiﬁc, Indus-

trial, and Security applications. The underwater sen-

sor design ranges from simple to complex, while the

prices range from few dollars to thousands of dollars

depending on the monitored parameter. Sensors used

to measure pressure, light penetration, and tempera-

ture are relatively low-priced. On the other hand, sen-

sors used in applications such as estimating the con-

centration of chlorophyll, CO

monitoring, and de-

tecting underwater objects use more complex sensor

technology and hence are more expensive(Heidemann

et al., 2012; Lloret, 2013). Figure 4 gives a summary

of the applications of UWSN under each of the three

categories stated above:

Figure 4: Applications of UWSN Technology (Heidemann

et al., 2012).

Depending on the target data to be collected, de-

ployment of underwater sensor networks is catego-

rized as discussed below:

• Static or Mobile Mode

Static nodes are attached to anchored buoys or at-

tached to the seaﬂoor and remain there collecting and

transmitting information back to the monitoring sta-

tions(Ibrahim et al., 2009; Alhumyani et al., 2015).

On the other hand, mobile nodes are mounted on au-

tonomous underwater vehicles (AUVs)5, gliders, or

drifters(Heidemann et al., 2012). In doing this, one

can monitor data over a large area using the same

hardware. Since the mobile nodes are needed to cover

a more extensive topology, they consume more energy

during the data collection and relay process.

• Short-term and Long-term Monitoring

Deployment of underwater sensors can be for short

periods ranging from a few hours to a day or for

more extended periods ranging from a few months

to several years. Most underwater deployments

happen on a short-term basis due to the harsh aquatic

environment(Heidemann et al., 2012). After a certain

SENSORNETS 2022 - 11th International Conference on Sensor Networks

208

Figure 5: Autonomous Underwater Vehicle – AUV (Blid-

berg, 2001).

period of operation, there arises the need to service

the equipment and allow for recharging, especially

for mobile nodes that run on battery power(Alsulami

et al., 2020a).

The above factors affect the design and opera-

tion of the various equipment deployed for underwa-

ter sensing applications. The choice between static

or mobile deployments and the length of the sensors’

operation period before any service is required varies

depending on the speciﬁc application the UWSN is

applied. This next section discusses the various appli-

cations of UWSNs across multiple sectors.

3.1 Scientiﬁc Applications and Research

Water covers over 70% of the earth’s sur-

face(Felemban et al., 2015). These parts of the

earth underwater remain hugely unexplored and

contain vast resources and data awaiting discovery.

Applications of UWSNs in scientiﬁc research revolve

around environmental monitoring, ocean sampling,

and monitoring of biological activities on the ocean

ﬂoor(Lloret, 2013; Heidemann et al., 2012).

3.1.1 Environmental and Marine Life

Monitoring

Environmental and marine life monitoring involves

monitoring the underwater environment, various ob-

jects of interest, and marine life. Ecological mon-

itoring is split into the tracking of the physical pa-

rameters of the water to determine its quality, mon-

itoring of marine life and their habitats, and perfor-

mance of underwater exploration activities for discov-

ery or research. Tracking the ocean water quality is

mainly geared towards assessing the chemical and bi-

ological waste deposited on the ocean ﬂoor(Collins,

2013). The water’s oxygen levels, temperature, and

pressure are also measured, and the data is sent to on-

shore monitoring stations for further analysis. Moni-

toring the quality of water ensures that aquatic life is

not affected by any human activities(Kiranmayi and

Kathirvel, 2015).

Some of the most common pollution instances in-

clude oil spillages, chemical and nuclear pollution

from facilities close to water bodies that use the wa-

ter in their operations. UWSNs make identifying any

pollution cases easier than manual methods by en-

abling the monitoring stations to get real-time infor-

mation(Kiranmayi and Kathirvel, 2015). The amount

of aquatic life that has been affected can be accu-

rately determined hence enabling timely planning of

countermeasures to salvage the situation. Noticing a

change in some of the nominal operating conditions

of the ocean happens fast, and the possible cause is

identiﬁed before any harm to human or aquatic life

happens(Kiranmayi and Kathirvel, 2015). The under-

water sensors deployed for such applications are pri-

marily static(Bradbeer et al., 2007).

The aim of monitoring aquatic life such as ﬁsh,

mammals, and microorganisms is to observe, study,

and understand their behavior hence further expand

the current scientiﬁc knowledge base(Kiranmayi and

Kathirvel, 2015). The living environment of these un-

derwater creatures is monitored and analyzed. Any

human activities in the vicinity of the study area

are also monitored to determine whether they impact

aquatic life in any way(Felemban et al., 2015; Kiran-

mayi and Kathirvel, 2015). Such systems capture vi-

sual and technical data of the marine organisms and

transmit it wirelessly to a control and monitoring cen-

ter where the data is analyzed and triggers an alert

should an unusual occurrence be detected(Collins,

2013).

3.1.2 Ocean Sampling

Ocean sampling involves the monitoring of under-

water phenomena with the aim of scientiﬁc study.

In this case, the underwater sensors are mounted on

AUVs that move around unexplored regions of the

ocean surface. The area covered by each vehicle is

approximately 500 square meters for every deploy-

ment(Fattah et al., 2020a).

3.1.3 Coral Reefs Study

Coral reefs are underwater ecosystems that are built

by microorganisms that live in the water. They are

studied to check how human activity impacts them. A

lot of data is collected and incorporated into simula-

tions tools. One can project the impact of any change

in the ocean conditions on the thriving of these under-

water ecosystems(Kiranmayi and Kathirvel, 2015).

Both static, as well as mobile nodes are deployed for

Underwater Wireless Sensor Networks: A Review

209

such applications. The static nodes are used to col-

lect data such as temperature and images of the coral

reefs at certain strategic positions of the reef(Fattah

et al., 2020a; Lloret, 2013). AUVs, which are the mo-

bile nodes, are used to collect information from the

stationary nodes and relay it to surface buoy nodes

for transmission to the monitoring and control cen-

ters at the shore. The static nodes are built to with-

stand the harsh underwater environment, while the

mobile nodes are limited by powering issues (batter-

ies) and shorter lifetimes(Kiranmayi and Kathirvel,

2015). Mobile nodes are hence not used for long ex-

cursion periods due to these limitations.

3.2 Industrial

3.2.1 Mineral Exploration

Underwater sensors are used during the exploration

of oil and gas deposits on the ﬂoor of the ocean.

Doing this enables faster and more accurate explo-

ration to establish the presence and viability of such

resources(Maeda et al., 2011). UWSN technology

is also used to monitor and control underwater rigs

to ensure effective operation with minimum pollu-

tion to the surrounding environment. The other min-

eral that has attracted the interest of researchers in re-

cent times is manganese crust(Lloret, 2013; Felemban

et al., 2015; Kiranmayi and Kathirvel, 2015). Under-

water sensor technology has been deployed to iden-

tify, locate and map these underwater deposits. AUVs

mounted with optical and acoustic modules are used

for such applications, facilitating signiﬁcant area cov-

erage during exploration(Maeda et al., 2011). These

AUVs identify, map and monitor these manganese de-

posits to depths of 3000m below the water surface

(Felemban et al., 2015).

3.2.2 Pipeline Monitoring

UWSNs have, in recent times, been incorpo-

rated in the monitoring of underwater oil and gas

pipelines(Jacobi and Karimanzira, 2013). This is

done when oil and gas are extracted from the ocean

ﬂoor and transferred to the shore using pipelines.

Pipelines are also used to interconnect different con-

tinents, making them a critical infrastructure that

needs continuous monitoring(Jassim and Abdelka-

reem, 2020). The health of a vast pipeline can be

determined in a short period of time compared to

when manual methods are used. In doing this, some-

body can take quick remedial actions if a fault is no-

ticed(Jassim and Abdelkareem, 2020; Fattah et al.,

2020a).

3.2.3 Monitoring of Commercial Fisheries

UWSNs are used in monitoring commercial ﬁshing

activities in ﬁsh farms. Fish farming is a most de-

manding activity that calls for continuous monitoring

of the ﬁshes’ habitat to guarantee maximum produc-

tion. In most cases, the ﬁsh live in a closed environ-

ment hence calling for close monitoring of their habi-

tat to ensure that all the living conditions for the ﬁsh

are optimally maintained(Chang et al., 2016; Fattah

et al., 2020a). The parameters monitored are tem-

perature, water pH, NH4 content, amount of uneaten

feed, and fecal waste from the ﬁsh. This is done to de-

termine the quality of the water and determine when

a change is needed(Felemban et al., 2015). UWSN

technology allows for easy management of huge ﬁsh

farms when compared to traditional methods. Large

amounts of data are collected and stored to help make

better decisions in the future to ensure the productiv-

ity of these farms. With the continued ﬂow of real-

time data from UWSNs at their disposal, one is guar-

anteed to make the best decisions to ensure the com-

mercial viability of a ﬁsh farm(Felemban et al., 2015;

Lloret, 2013).

3.3 Security Applications

Applications of UWSNs involve the use of these sen-

sors for monitoring the coastline activities and iden-

tifying any unwanted or unpermitted battleships or

submarines in the vicinity(Jacobi and Karimanzira,

2013; Felemban et al., 2015; Zwanzig, 2018). Any

potential enemies are determined in advance and neu-

tralized before launching an attack, and the secu-

rity teams can plan for offenses to neutralize the en-

emy before they strike. SeaWeb was one of the

earliest underwater communication technology de-

ployed by the military for detecting and communicat-

ing with submarines(Heidemann et al., 2012). The

mode of communication underwater is usually lim-

ited to short distances compared to terrestrial com-

munication. Current technological advances are look-

ing into how this communication can be expanded

since most parts of the underwater world remain un-

explored and unexploited(Felemban et al., 2015). Op-

tical and acoustic sensors are used in mine detection

under the water. Similar to terrestrial mines, under-

water mines pose a threat to aquatic activities carried

out by security agencies(Dong et al., 2015). Hence,

underwater sensor technologies help avert loss of life

when used for reconnaissance activities to identify

and neutralize any hidden threats positioned on the

seaﬂoor(Zwanzig, 2018).

SENSORNETS 2022 - 11th International Conference on Sensor Networks

210

3.3.1 Disaster Identiﬁcation and Management

Tsunamis have been a natural disaster that has posed

considerable harm to human life and activities car-

ried out close to seas and oceans. UWSNs are de-

ployed to monitor the conditions of the ocean ﬂoor,

and if some seismic activities are noted, the sensors

send a warning to the shore stations. This informa-

tion can then be relayed to those near the seas, advis-

ing them to move to higher grounds(Alsulami et al.,

2019; Zwanzig, 2018). In doing this, damage to prop-

erty and loss of life is averted. Underwater volcanos

and earthquakes are also monitored to gather informa-

tion on any upcoming or expected seismic activities.

At times, the ruthless behavior of oceans limits the

traditional methods of gathering information and ren-

ders some areas of the sea inaccessible. This is where

the underwater sensor technology comes in handy to

guarantee continued monitoring of the oceanic con-

ditions despite the rough nature of the ocean wa-

ters(Dong et al., 2015). In doing this, disasters are

averted through early detection and relaying of infor-

mation(Lloret, 2013).

One of the adverse effects of global warming has

been increased ﬂooding around the world. This has

led to the loss of life and destruction of property when

the ﬂoods strike unexpectedly. UWSN technology

has found an application in monitoring such events

to ensure that timely alerts are sent out, saving lives

and reducing the destruction of property(Jassim and

Abdelkareem, 2020; Amoli, 2016). Aquatic vitals

are monitored by some stationary nodes then trans-

mitted to a remote shore station for analysis of any

impending ﬂoods. Some of the parameters monitored

include water levels, thrust, and intensity of ﬂow.

Other information that is monitored includes temper-

ature, humidity, and amount of precipitation(Arima

et al., 2014). After gathering this information, pre-

diction tools are used to determine whether a ﬂood

is imminent and when one might occur. This infor-

mation guides emergency services on any evacuation

that needs to be done and when it should be done.

Other forms of disasters that may occur can be artiﬁ-

cial, such as oil spillages(Arima et al., 2014; Felem-

ban et al., 2015). These are bound to cause ecologi-

cal instabilities in life on and under the water if not

adequately mitigated(Amoli, 2016; Alsulami et al.,

2020a; Dong et al., 2015).

3.3.2 Assisted Navigation

In assisted navigation, modern-day ships can get in-

formation about impending threats such as rocks,

shoals, and submerged vehicles(Gallagher et al.,

2016). This helps the crew plan for their voyages

and avoid accidents such as what begot the titanic

from happening. Traditional systems would depend

on the manual observation of threats then communi-

cating them to the ship’s navigation team for action.

Any miscalculation of danger would tend to be fatal,

as was the case with the titanic(Gallagher et al., 2016;

Lee et al., 2004; Felemban et al., 2015).

4 CHALLENGES AND

LIMITATIONS OF UWSNs

Although UWSNs have grown in the application, they

present several limitations in their deployment. These

challenges require to be addressed by research to con-

tinuously improve the reliability, efﬁciency, and ap-

plicability of these UWSNs(Akyildiz et al., 2005).

The challenges and limitations of UWSNs include

Limited Bandwidth, Propagation delay, Delay Vari-

ance, Link reliability, complex acoustic environment,

common standard and interface, sensor heterogene-

ity, hardware-related limitations, communication, and

visualization challenges(Fattah et al., 2020b). To

clearly understand the challenges of UWSNs, it is im-

portant to study the differences that these networks

have when compared to terrestrial systems(Ryecroft

et al., 2018). Firstly, underwater sensor networks

are expensive when compared to terrestrial networks.

The high cost is due to the complex design of un-

derwater transceivers and the protection hardware re-

quired for protection against the harsh environmen-

tal conditions that underwater systems are exposed

to(Akyildiz et al., 2005). Secondly, underwater wire-

less sensor nodes are sparsely deployed when com-

pared to terrestrial nodes (Akyildiz et al., 2005). This

sparse deployment is occasioned by the high cost of

the underwater sensor nodes and the harsh deploy-

ment environment(Ryecroft et al., 2018). In terms

of power requirements, UWSNs require more power

than their terrestrial counterparts. USWNs require

complex signal processing at the transceivers to com-

pensate for noise and signal loss along the underwater

transmission channel(Alsulami et al., 2019; Bhanu-

mathi et al., 2019).

Due to intermittency of data transition in under-

water systems, USWNs require additional memory

for caching to mitigate the intermittency(Iqbal and

Lee, 2015; Fattah et al., 2020b). On the other hand,

terrestrial systems do not require huge memory for

caching because data transmission is stable. Addi-

tionally, data from underwater sensor networks are

more spatially correlated than that of terrestrial sen-

sor networks(Lloret, 2013; Akyildiz et al., 2005). The

reason for this is the large distances between sensor

Underwater Wireless Sensor Networks: A Review

211

nodes in underwater sensor networks (Akyildiz et al.,

2005). These differences between underwater sensor

networks and terrestrial systems make the deployment

of underwater sensor networks and research in these

networks challenging(Akyildiz et al., 2005). The fol-

lowing sections look at the challenges of underwater

sensor networks in more detail and how researchers

are addressing these challenges(Akyildiz et al., 2005).

4.1 Limited Bandwidth

Underwater wireless sensor networks require acous-

tic channels to communicate. Acoustic channels have

a limited bandwidth of below 30kHz due to absorp-

tion(Kilfoyle and Baggeroer, 2000). The underwa-

ter environmental factors make underwater commu-

nication channels highly variable(Awan et al., 2019).

These factors make the bandwidth dependent on the

frequency of transmission and the distance between

two communication nodes. The acoustic channel de-

ployed in shallow waters will be affected by a high-

temperature gradient, surface noise, and multipath ef-

fect compared to a channel deployed in deep waters

(Awan et al., 2019). For acoustic channels, the band-

width increases with an increase in depth and tem-

perature. The bandwidth also decreases with an in-

crease in distance(Bhanumathi et al., 2019; Kilfoyle

and Baggeroer, 2000). Due to bandwidth dependence

with distance, it requires that the Underwater sensor

network nodes be deployed as close to each other as

possible. This would mean that the cost of deploy-

ing these nodes would be high, and also, the net-

work power requirements would be high. To address

the challenge of limited bandwidth, researchers have

looked at the possibility of UWSNs accessing shared

channels(Akyildiz et al., 2005; Ryecroft et al., 2018).

To access shared channels, the media access con-

trol protocol (MAC) is used to coordinate the UWSN

nodes and ensure the validity of data sent through the

channel (Akyildiz et al., 2005).

4.2 Propagation Delay

High propagation delays are inherent in Underwater

Sensor Networks. The protocols designed for terres-

trial radio networks usually ignore propagation de-

lays because they are small. These protocols can-

not work well in underwater systems because of the

high delay experienced in acoustic networks (Chen

et al., 2010). Propagation delays in UWSNs are ﬁve

times higher for radio networks when compared to

terrestrial networks (Pompili et al., 2006). When the

UWSNs are used for real-time monitoring, for exam-

ple, in surveillance applications, this delay is unde-

sirable(Ryecroft et al., 2018; Kilfoyle and Baggeroer,

2000). Protocols that take into consideration this de-

lay need to be developed speciﬁcally for real-time

monitoring in underwater applications. These pro-

tocols should restore connectivity quickly when lost

and decongest congested links by dynamic rerout-

ing to minimize propagation delay. An example of

such protocol is the store and forward mechanism de-

veloped by the Delay-Tolerant Networking Research

Group. This protocol uses middleware between the

application and lower layers to resolve intermittent

connectivity and long delays (Akyildiz et al., 2005;

Kilfoyle and Baggeroer, 2000). Delays in UWSNs

for some applications, such as those used for seis-

mic monitoring, are occasioned by producing a huge

amount of data when these sensors are suddenly ac-

tivated. Such sensor networks require careful design

to minimize the propagation delay by tuning existing

networks to the characteristics of the underwater en-

vironment(Akyildiz et al., 2005; Awan et al., 2019).

4.3 Path Loss

Path loss is the decline in the power density of

an electromagnetic signal as it propagates through

a medium(Alhumyani et al., 2015; Ibrahim et al.,

2009). In UWSNs, path loss leads to attenuation

of the signal. Attenuation in UWSNs occurs when

acoustic energy is converted to heat(Kilfoyle and

Baggeroer, 2000; Awan et al., 2019). Signal attenua-

tion in underwater application increases with distance

and frequency, as shown in ﬁgure 6 (Akyildiz et al.,

2005; Fattah et al., 2020b).

Attenuation is also caused by reverberation and

scattering, and dispersion. Another critical source of

path loss is geometric spreading(Kilfoyle and Bag-

geroer, 2000). UWSNs rely on acoustic waves for

signal transmission. Acoustic waves are highly sus-

ceptible to geometric spreading. Geometric spreading

refers to the propagation of sound energy due to the

expansion of the wavefront (Min et al., 2012). Geo-

metric spreading increases with distance but is inde-

pendent of frequency(Min et al., 2012).

4.4 Unpredictable and Unreliable

Underwater Environment

The unpredictability of the underwater environment

makes it very hard to design and deploy UWSNs.

Water activities are unpredictable, the water pres-

sure is high, and the uneven seawater depth is unpre-

dictable(Felemban et al., 2015; Amoli, 2016).

And also, nodes in underwater sensor networks

are continuously moving due to water currents. Lo-

SENSORNETS 2022 - 11th International Conference on Sensor Networks

212

Figure 6: Attenuation vs. frequency and distance (Akyildiz

et al., 2005).

calization of these nodes is therefore crucial for reli-

able transmission of information(Almutairi and Mah-

foudh, 2017; Kilfoyle and Baggeroer, 2000; Min

et al., 2012). Terrestrial localization of nodes cannot

be applied to underwater networks because the under-

water conditions will disorganize the nodes and net-

work topology, making the nodes unreliable(Akyildiz

et al., 2005; Chandrasekhar et al., 2006).

5 CONCLUSION

In conclusion, we wanted, in this paper, to provide re-

searchers and readers who are interested in UWSNs

valuable overview about this promising technology.

We started our paper by showing the importance of

this technology nowadays and presenting main moti-

vations to have such technology. Then, We described

the different architectures that have been developed

by researchers so far. After that, we summarized some

of the deployment approaches that are used to deploy

nodes in the UWSNs. Moreover, we surveyed some

of the applications of UWSNs. We also discussed

core requirements of UWSNs. We discussed in this

paper some of the signiﬁcant challenges and limita-

tion of UWSNs. We concluded this paper by present-

ing and comparing three different meduims that are

used in UWSNs.

REFERENCES

Akyildiz, I. F., Pompili, D., and Melodia, T. (2005). Under-

water acoustic sensor networks: research challenges.

Ad hoc networks, 3(3):257–279.

Aldosari, H., Elfouly, R., Ammar, R., and Alsulami, M.

(2020). Performance of new monitoring architectures

for underwater oil/gas pipeline using hyper-sensors.

In 2020 IEEE Symposium on Computers and Commu-

nications (ISCC), pages 1–6.

Alharbi, A., Alhumyani, H., Tolba, S., and and, R. A.

(2014). Underwater sensing and processing networks

(uspn). In 2014 IEEE Symposium on Computers and

Communications (ISCC), pages 1–7.

Alharbi, A., Ibrahim, S., Ammar, R., and Alhumyani, H.

(2015). Performance analysis of efﬁcient pipeline ar-

chitectures for underwater big data analytics. In 2015

IEEE Symposium on Computers and Communication

(ISCC), pages 1014–1019.

Alhumyani, H., Ammar, R., Alharbi, A., and Tolba, S.

(2015). Efﬁcient surface-level gateway deployment

using underwater sensing and processing networks. In

OCEANS 2015 - MTS/IEEE Washington, pages 1–6.

Almutairi, A. and Mahfoudh, S. (2017). Deployment pro-

tocol for underwater wireless sensors network based

on virtual force. International Journal of Advanced

Computer Science & Applications, 8(11):241–249.

Alsulami, M., Elfouly, R., and Ammar, R. (2021). A re-

liable underwater computing system. In 2021 4th

IEEE International Conference on Industrial Cyber-

Physical Systems (ICPS), pages 467–472.

Alsulami, M., Elfouly, R., Ammar, R., and Aldosari, H.

(2020a). Deployment of multiple computing systems

in underwater wireless sensor networks. In 2020 IEEE

International Symposium on Signal Processing and

Information Technology (ISSPIT), pages 1–6.

Alsulami, M., Elfouly, R., Ammar, R., Aldosari, H., and Al-

enizi, A. (2019). Ideal number of computers for real-

time underwater computing systems. In 2019 IEEE

International Symposium on Signal Processing and

Information Technology (ISSPIT), pages 1–5.

Alsulami, M., Elfouly, R., Ammar, R., and Alenizi, A.

(2020b). A modiﬁed k-medoids algorithm for de-

ploying a required number of computing systems in

a three dimensional space in underwater wireless sen-

sor networks. In 2020 IEEE International Symposium

on Signal Processing and Information Technology (IS-

SPIT), pages 1–5.

Amoli, P. V. (2016). An overview on current researches on

underwater sensor networks: Applications, challenges

and future trends. International Journal of Electrical

and Computer Engineering, 6(3):955.

Arima, M., Yoshida, K., and Tonai, H. (2014). Development

of a four-channel underwater passive acoustic moni-

toring system for the use of underwater vehicles. In

2014 Oceans - St. John’s, pages 1–6.

Awan, K. M., Shah, P. A., Iqbal, K., Gillani, S., Ahmad,

W., and Nam, Y. (2019). Underwater wireless sen-

sor networks: A review of recent issues and chal-

lenges. Wireless Communications and Mobile Com-

puting, 2019.

Bhanumathi, S., Rajkishore, S., and Kowshik, S. (2019).

Localized aided energy efﬁcient optimal relay selec-

tion protocol for underwater wireless sensor networks.

In 2019 3rd International Conference on Trends in

Electronics and Informatics (ICOEI), pages 880–885.

Bhaskarwar, R. V. and Pete, D. J. (2021). Cross-layer design

approaches in underwater wireless sensor networks: A

survey. SN Computer Science, 2(5):1–26.

Blidberg, D. R. (2001). The development of autonomous

underwater vehicles (auv); a brief summary. In Ieee

Icra, volume 4, page 1. Citeseer.

Underwater Wireless Sensor Networks: A Review

213

Bradbeer, R. S., Hodgson, P., Lam, K., and Ku, K.

K. K. (2007). Establishment of a three dimen-

sional, real-time marine environment monitoring sys-

tem, databuoy™ i, in the hoi ha wan marine park,

hong kong, with connection to the internet. In 2007

14th International Conference on Mechatronics and

Machine Vision in Practice, pages 174–178.

Chandrasekhar, V., Seah, W. K., Choo, Y. S., and Ee,

H. V. (2006). Localization in underwater sensor net-

works: survey and challenges. In Proceedings of the

1st ACM international workshop on Underwater net-

works, pages 33–40.

Chang, J.-E., Huang, S.-W., and Guo, J.-H. (2016). Hunt-

ing ghost ﬁshing gear for ﬁshery sustainability using

autonomous underwater vehicles. In 2016 IEEE/OES

Autonomous Underwater Vehicles (AUV), pages 49–

53.

Chen, Y.-S., Juang, T.-Y., Lin, Y.-W., and Tsai, I.-C. (2010).

A low propagation delay multi-path routing protocol

for underwater sensor networks. Internet Technology

Journal, 11(2):153–165.

Collins, A. (2013). Application of subsea wireless technol-

ogy to environmental monitoring. In 2013 OCEANS -

San Diego, pages 1–5.

Dong, Y., Zhang, G., He, X., and Tang, J. (2015). Informa-

tion fusion in networked underwater target detection.

In OCEANS 2015 - Genova, pages 1–4.

Fattah, S., Gani, A., Ahmedy, I., Idris, M. Y. I., and Tar-

gio Hashem, I. A. (2020a). A survey on underwater

wireless sensor networks: Requirements, taxonomy,

recent advances, and open research challenges. Sen-

sors, 20(18):5393.

Fattah, S., Gani, A., Ahmedy, I., Idris, M. Y. I., and Tar-

gio Hashem, I. A. (2020b). A survey on underwater

wireless sensor networks: Requirements, taxonomy,

recent advances, and open research challenges. Sen-

sors, 20(18):5393.

Felemban, E., Shaikh, F. K., Qureshi, U. M., Sheikh,

A. A., and Qaisar, S. B. (2015). Underwater sensor

network applications: A comprehensive survey. In-

ternational Journal of Distributed Sensor Networks,

11(11):896832.

Gallagher, D. G., Manley, R. J., Hughes, W. W., and Pilcher,

A. M. (2016). Development of an enhanced underwa-

ter navigation capability for military combat divers. In

OCEANS 2016 MTS/IEEE Monterey, pages 1–4.

He, M., Chen, Q., Dai, F., and Zheng, X. (2018). Topo-

logical conﬁguration and optimization in underwa-

ter acoustic sensor networks: A survey. Inter-

national Journal of Distributed Sensor Networks,

14(8):1550147718795359.

Heidemann, J., Stojanovic, M., and Zorzi, M. (2012). Un-

derwater sensor networks: applications, advances and

challenges. Philosophical Transactions of the Royal

Society A: Mathematical, Physical and Engineering

Sciences, 370(1958):158–175.

Ibrahim, S., Ammar, R., and Cui, J. (2009). Geometry-

assisted gateway deployment for underwater sensor

networks. In 2009 IEEE Symposium on Computers

and Communications, pages 932–937.

Iqbal, Z. and Lee, H.-N. (2015). Deployment strategy anal-

ysis for underwater cooperative wireless sensor net-

works. In 2015 International Conference on Infor-

mation and Communication Technology Convergence

(ICTC), pages 699–703.

Jacobi, M. and Karimanzira, D. (2013). Underwater

pipeline and cable inspection using autonomous un-

derwater vehicles. In 2013 MTS/IEEE OCEANS -

Bergen, pages 1–6.

Jassim, W. M. and Abdelkareem, A. E. (2020). Performance

enhancement of oil pipeline monitoring for underwa-

ter wireless sensor network. In 2020 International

Conference on Computer Science and Software Engi-

neering (CSASE), pages 38–43.

Khan, M. F., Bibi, M., Aadil, F., and Lee, J.-W. (2021).

Adaptive node clustering for underwater sensor net-

works. Sensors, 21(13):4514.

Kilfoyle, D. B. and Baggeroer, A. B. (2000). The state of

the art in underwater acoustic telemetry. IEEE Journal

of oceanic engineering, 25(1):4–27.

Kiranmayi, M. and Kathirvel, A. (2015). Underwater wire-

less sensor networks: applications, challenges and de-

sign issues of the network layer-a review”. Interna-

tional Journal of Emerging Trends in Engineering Re-

search, 3(1):05–11.

Lee, P., Jeon, B., Kim, S., Choi, H., Lee, C., Aoki, T.,

and Hyakudome, T. (2004). An integrated naviga-

tion system for autonomous underwater vehicles with

two range sonars, inertial sensors and doppler veloc-

ity log. In Oceans ’04 MTS/IEEE Techno-Ocean ’04

(IEEE Cat. No.04CH37600), volume 3, pages 1586–

1593 Vol.3.

Lloret, J. (2013). Underwater sensor nodes and networks.

Maeda, F., Takagawa, S., Asada, A., Maki, T., Sasaki,

Y., and Yamamoto, F. (2011). Next r amp;d

of measurement method of offshore minerals using

the low-frequency acoustic mills-cross transmission

and receiving system multibeam sub-bottom proﬁl-

ing method and ”natsushima” cruise experiment. In

2011 IEEE Symposium on Underwater Technology

and Workshop on Scientiﬁc Use of Submarine Cables

and Related Technologies, pages 1–5.

Min, H., Cho, Y., and Heo, J. (2012). Enhancing the reli-

ability of head nodes in underwater sensor networks.

Sensors, 12(2):1194–1210.

Pompili, D., Melodia, T., and Akyildiz, I. F. (2006). Routing

algorithms for delay-insensitive and delay-sensitive

applications in underwater sensor networks. In Pro-

ceedings of the 12th annual international conference

on Mobile computing and networking, pages 298–309.

Ryecroft, S. P., Shaw, A., Fergus, P., Kot, P., Muradov,

M., Moody, A., and Conroy, L. (2018). Requirements

of an underwater sensor-networking platform for en-

vironmental monitoring. In 2018 11th International

Conference on Developments in ESystems Engineer-

ing (DeSE), pages 95–99. IEEE.

Zhang, W., Wang, J., Han, G., Zhang, X., and Feng, Y.

(2019). A cluster sleep-wake scheduling algorithm

based on 3d topology control in underwater sensor

networks. Sensors, 19(1):156.

Zwanzig, C. (2018). Ship-based exploration of sub-

marine gas hydrates and environmental monitoring

with multibeam echo sounders. In 2018 OCEANS -

MTS/IEEE Kobe Techno-Oceans (OTO), pages 1–8.

SENSORNETS 2022 - 11th International Conference on Sensor Networks

214