RetractedArticlePDF Available

Abstract and Figures

The Underwater Internet of Things (UIoTs), as a specific genre of Internet of Things (IoTs), represent networked devices that exploit sensors and objects (deployed underwater) to collect and process oceanic data. The oceanic data represents a multitude of valuable information and exploratory artefacts, including but not limited to underwater temperatures, acidity, types and classification of marine life, quality of sea water, and minerals. When applied to context-sensitive information collected from UIoTs, data analytics can provide valuable insights into the stakeholders (e.g. environmentalists, marine explorers, oceanographers) for decision making and intelligence about underwater data (i.e. smart oceanography). We propose to unify the concept of UIoTs with data analytics to (i) enable the sensor-driven collection of context-sensitive oceanic data, (ii) apply data analytics to derive useful information, and (iii) provide reporting and data-driven intelligence to the stakeholders. To support this, we have developed a layered architecture and evaluated it using a real-world case study on analysing oceanic data. Results of evaluation streamline the usability and efficiency of the architecture in terms of (a) sensors throughput (i.e. stability), (b) query response time (i.e. performance), and (c) ease of use for stakeholders (i.e. usability). The proposed solution complements existing research and development on IoTs for smart cities and specifically contributes to UIoTs in thesmart oceans. Results and their evaluation demonstrate the architecture’s applicability as a proposed solution and provide guidelines to engineer and develop emerging and next generation of UIoTs for ocean data analytics.
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=tijr20
IETE Journal of Research
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/tijr20
Underwater Internet of Things to Analyse Oceanic
Data
Yasser Alharbi & Aakash Ahmad
To cite this article: Yasser Alharbi & Aakash Ahmad (2022): Underwater Internet of Things to
Analyse Oceanic Data, IETE Journal of Research, DOI: 10.1080/03772063.2022.2027286
To link to this article: https://doi.org/10.1080/03772063.2022.2027286
Published online: 14 Feb 2022.
Submit your article to this journal
Article views: 20
View related articles
View Crossmark data
IETE JOURNAL OF RESEARCH
https://doi.org/10.1080/03772063.2022.2027286
Underwater Internet of Things to Analyse Oceanic Data
Yasser Alharbi and Aakash Ahmad
College of Computer Science and Engineering, University of Ha’il, Ha’il 81481, Saudi Arabia
ABSTRACT
The Underwater Internet of Things (UIoTs), as a specific genre of Internet of Things (IoTs), represent
networked devices that exploit sensors and objects (deployed underwater) to collect and process
oceanic data. The oceanic data represents a multitude of valuable information and exploratory
artefacts, including but not limited to underwater temperatures, acidity, types and classification
of marine life, quality of sea water, and minerals. When applied to context-sensitive information
collected from UIoTs, data analytics can provide valuable insights into the stakeholders (e.g. environ-
mentalists, marine explorers, oceanographers) for decision making and intelligence about underwa-
ter data (i.e. smart oceanography). We propose to unify the concept of UIoTs with data analytics to
(i) enable the sensor-driven collection of context-sensitive oceanic data, (ii) apply data analytics to
derive useful information, and (iii) provide reporting and data-driven intelligence to the stakehold-
ers. To support this, we have developed a layered architecture and evaluated it using a real-world
case study on analysing oceanic data. Results of evaluation streamline the usability and efficiency
of the architecture in terms of (a) sensors throughput (i.e. stability), (b) query response time (i.e. per-
formance), and (c) ease of use for stakeholders (i.e. usability). The proposed solution complements
existing research and development on IoTs for smart cities and specifically contributes to UIoTs in
thesmart oceans. Results and their evaluation demonstrate the architecture’s applicability as a pro-
posed solution and provide guidelines to engineer and develop emerging and next generation of
UIoTs for ocean data analytics.
KEYWORDS
Internet of things; smart
cities; software architecture;
software engineering
1. INTRODUCTION
Internet of Things (IoTs), as the pervasive technology,
are driven by a collection of networked devices and sen-
sors that ingest context-sensitive data from surrounding
environment(s) such as monitoring trac congestions,
analyzing air pollution, or analytics oceanic variables
[1]. IoTs play a fundamental role in enabling smart sys-
tems and infrastructures across a multitude of domains,
including but not limited to home and industrial automa-
tion, vehicular networks, service robots and urban man-
agement [2]. International Data Corporation (IDC) a
premier company on market research and intelligence
predicts that by 2025 41 billion devices (4 IoT devices
perperson)willbeconnectedtoIoTplatformswith
an estimated revenue generation of 670 billion dollars
[3]. Rapidly increasing adoption of IoT technologies and
infrastructure can be primarily attributed to context-
sensitive devices that can unify hardware sensors, soft-
ware applications that control those sensors, and network
protocols that interconnect the sensors to operationalise
IoT systems [1,4]. Real-world examples of such context-
sensitive devices can be smartwatches, location trackers,
and surveillance cameras, etc., to support smart health,
transportation, and security management in smart city
systems [5]. The Underwater Internet of Things (UIoTs)
is a specic genre of IoTs that can exploit several under-
water sensors to sense context-sensitive from relating to
oceandata[6].Typicalexamplesofsuchcontext-sensitive
information can be oceanic data in terms of temperature,
acidity, the concentration of marine life, and underwater
minerals that can be of strategic importance for stake-
holders, such as oceanographers and environmentalists
[7]. While there is a lot of research and development on
IoT systems [1,2,4]andtheirroleinsmartcities[5], much
lesshasworkedhasbeendonethatfocusesondeveloping
and operationalising UIoTs for data-driven intelligence
from oceanic information. In recent years, there has been
a growing need for research and development to leverage
existing IoT applications and services beyond traditional
smart city systems to explore smart oceans [7].
Data analytics represents a structured process to collect,
process, transform, and analyse the data that can help
individuals and organisations derive key insights and dis-
cover trends that empower stakeholders with data-driven
intelligence and decision support [8]. IoT-driven data
analytics is a rapidly emerging discipline of data science
that enables sensors to ingest context-sensitive, transmit
© 2022 IETE
2 Y. ALHARBI AND A. AHMAD: UNDERWATER INTERNET OF THINGS TO ANALYSE OCEANIC DATA
it to the backend server(s) for data processing, and pro-
vide analytics for smart systems [9]. IoT systems pro-
vide foundations to operationalise such applications that
are fundamental to designing, developing, and deploying
data-driven smart systems that are being adopted across
continents from Europe [11]toAsia,Africa[12], and
the United States [10]. IoT-driven data analytics appli-
cations represent a plethora of technical and nancial
opportunities, with International Data Corporation fore-
casting that by the end of 2022, nancial revenues gen-
erated by big data analytics from IoT applications will
touch approximately 274.3 billion dollars. [13]. In recent
years, research and development eorts have focused on
applying principles, methods, algorithms, tools, and data
science technologies to IoT systems [14]. However, a
lack of research and development leverages data analyt-
ics techniques for UIoT systems [6,7]. The Smart Ocean
Technologies initiative pinpoints that UIoTs represent an
innovative class of IoT systems with strategic benets and
expected revenues in the context of smart ocean analyt-
ics [15]. However, UIoTs that orchestrate many sensors
deployed underwater also entail some critical challenges
[16]. As an example of such challenges, IoT devices,
sensors, services, and applications that sense context-
sensitive oceanic data i.e. ingesting underwater data
using sensors rely on hardware equipment, networking
protocols, and software systems to compute the correla-
tion between several oceanic variables such as underwa-
ter temperature (°C), the acidity of the water (pH) and
their impacts on marine life [6,7,15]. The roadmap and
directions about engineering IoT systems [13] leverage
empirical evidence from various IoT projects to stream-
line the needs for engineering frameworks, patterns, tools
and empirical foundations, which can aid IoT application
engineers and developers to design, develop, deploy and
maintain a class of IoT systems eectively and eciently.
Solution Overview: The proposed research aims to unify
the principle and practices of engineering IoT systems
with methods and tools of data analytics in the context
of UIoTs, as illustrated in Figure 1. Specically, in this
research we focus on developing and validating a UIoT
system that deploys many sensors underwater to ingest
contextual data (e.g. temperature, acidity, sunlight under-
water, and its impact on marine life). Once the data have
been collected from the sensors, data analytics performed
ataremoteservercanprovideusefulinsightsandintel-
ligence to the stakeholders. An overview of the proposed
solution is presented in Figure 1that illustrates a layered
system architecture of the solution comprising (i) a Sen-
sor Layer (ingesting underwater data), (ii) an Analytics
Layer (performing analytics for ingested data), and (iii)
anUser Interface Layer (reporting and visualisations for
Figure 1: Generic View of the Proposed Solution
end-users). Each of these layers performs specic tasks.
Forexample,thesensorlayeraccumulatessenseddata
from deployed sensors and transmits the accumulated
data to the analytics layer for its processing and deriv-
ing data intelligence. Each of the above-mentioned layers
is detailed later in a dedicated section to demonstrate
the applicability of the proposed solution, as shown in
Figure 1.
Existing work and proposed contributions: In recent years,
initiatives like Ocean-of-Things [27]promotecollabora-
tive and open-source practices to develop scalable solu-
tions for monitoring underwater data. In a similar con-
text,theresearchersin[25,26]haveappliedmachine
learning approaches with IoTs, referred to as UIoT
(underwater IoT), to collect, process, and analyse acous-
ticsounds.Theproposedsolutioninthisresearchaims
to complement the existing work in [2527]andspeci-
cally contributes an architecture that guides algorithmic
implementations (i.e. solution implementation) and tool
support (i.e. solution automation) for mine and analyse
oceanic data. We hypothesise : “An appropriate modu-
larisation (layering of the solution) can help to develop
ocean data mining solution that support ingesting data,
analysing it, and presenting it to the end-users”.Further
comparisons of the proposed research with the existing
work are performed in a subsequent section of this paper.
Theproposedsolutionisvalidatedwithacasestudy
on ocean data analytics by using the ISO/IEC-9126 that
evaluates the quality of software-intensive systems [17].
Evaluation of the proposed architectural solution primar-
ily focuses on validating the usability and eciency of the
solution in terms of sensors (1) sensors’ throughput (i.e.
stability of data ingestion), (2) query response time (i.e.
the performance of data storage and retrieval), and (3)
usability of the system (i.e. ease of use for stakeholders).
We outline the primary contributions of this research as
Y. ALHARBI AND A. AHMAD: UNDERWATER INTERNET OF THINGS TO ANALYSE OCEANIC DATA 3
Leveraging the state-of-the-art on data analytics in
the context of IoT systems to systematically archi-
tect,develop,andvalidateaspecialisedgenreofIoT
systems, referred to as UIoTs.
Developing a Layered architecture enabling reusab-
ility and modularisation that pioneers architecture-
centric engineering of UIoTs.
Validatingtheproposedsolutionwithacasestudy
on ocean data analytics for scenario-based demon-
stration and criteria-based evaluation of the pro-
posed solution.
Thenoveltyoftheproposedsolutionisbasedonengi-
neering and architecting UIoTs as an emerging class
of IoT systems that enable ocean data analytics in the
context of smart systems and infrastructures [2,4]. In
recent years, many research and development eorts have
streamlined the need for architecting IoT-driven systems
for smart ocean technologies [15,16]. However, no solu-
tion demonstrates (a) how toarchitectUIoTsinthecon-
text of data analytics?, and (b) what validation methods
can be employed for system evaluation. Therefore, we
outline the primary objectives of this research as
Developing an architecture (as a system blue-print),
implementing it (providing concrete specication) and
validating it (evaluating functionality and quality) of the
proposed solution that can exploit underwater sensors
to ingest data that can be analysed to gather key insights
about oceanic data.
The implications for proposed research can be benecial
for
Advancing state-of-the-art on UIoTs, particularly
unifying the concept of UIoTs with methods and
techniques of data analysis for smart ocean technolo-
gies.
Developing a layered architecture as a reference
model and providing guidelines and validation crite-
ria to engineer emerging and next-generation solu-
tions for IoTs.
Paper Organisation:ResearchContextandMethodol-
ogy are presented in Section 2. Related Work is detailed
in Section 3. Research Methodology is presented in
Section 4. Software architecture for UIoTs and its imple-
mentation are detailed in Section 4. The case study and
details of evaluations are in Section 5. Conclusions and
Future Work is discussed in Section 6.
2. RESEARCH CONTEXT AND PRELIMINARIES
We now present the research background and some pre-
liminaries to contextualise (i) IoUTs for ocean data in
Figure 2: Building Blocks for Underwater Internet of Things
Section 2.1 and (ii) data analytics for IoT-driven systems
in Section 2.2 that help us to pinpoint the challenges and
potential of the proposed research. Figure 2is used to
illustrate and conceptualise the core elements and ter-
minologies introduced in this section and will be used
throughout the paper.
2.1 UIoTs for Ocean Data Analytics
UIoTs as a concept and its underlying technologies are
being exploited in smart city systems to address various
challenges for ocean analytics [5,7]. Specically, UIoT
solutions are being promoted as part of smart ocean
technologies [15] that rely on many sensors that sense
context-sensitive underwater data and communicate it to
the backend servers, as illustrated in Figure 2.
Sensor Layer: The data ingested from sensors are stored,
retrieved, and processed at the server to provide use-
ful insights and key intelligence for the stakeholders
and assist them in decision-making. From a technical
perspective, as shown in Figure 2in UIoTs, underwa-
ter things refer to a collection of sensors (S1,S
2,... ,
SN)thatingestcontextualdataaboutmarinelifeand
underwater conditions and communicate such data to
a sensor’ bridge (B). For example, an individual sensor
SNingests context-sensitive information from underwa-
ter (e.g. underwater temperature °C) and adds auxil-
iary information such as date/time, sensor identity, geo-
proximity of the sensor, and communicates the data to
B. After accumulating the data from all the sensors, the
bridge transmits the data to a remote server. At the sen-
sor layer, the data communication between sensors S
4 Y. ALHARBI AND A. AHMAD: UNDERWATER INTERNET OF THINGS TO ANALYSE OCEANIC DATA
and bridge B is enabled via Bluetooth. Another possi-
bility can be Near Field Communication (NFC), but it
is constrained by the distance between communicating
devices. The intra-layer communication, i.e. data trans-
mission from the bridge to the analytics server, is enabled
via Wi-Fi, as Bluetooth communication is not supported.
Analytics Layer: Theserver(S)managesthestorageof
sensor data via backup and retrieval with the primary
purpose of performing analytics on the data. The anal-
ysis can reveal key insights from data that include but are
not limited to discovering recurring patterns and trends
(temperature value at specic time and location), a corre-
lation between oceanic variables (eects on temperature
°C on the acidity of water pH) of underwater data. Such
information can help stakeholders, such as geologists,
sea explorers and environmentalists, gain insights into
oceanic data, for instance, to know if the acidity of water
is aected by temperature and what impacts (temperature
°C and acidity pH) can exert on marine life.
User Interface Layer : Moreover, the user interface layer
inUIoTsactsasaninterfacebetweendataanalytics
and end-users with a presentation, customisation, and
reporting of data. Despite the potential and strategic ben-
ets of UIoTs, there are several challenges that must
be addressed for increased adoption of such systems in
smart city technologies [11,16]. Some of these challenges
relatetothesenorlayerwheretheresourcepoverty(i.e.
limited power, storage and processing capabilities) can
hinder the performance of the sensors that are portable,
context-sensitive, but resource-poor from a computa-
tional perspective. To ensure correct functionality and
desired quality of the system, sensors’ performance needs
to be evaluated in terms of sensors’ throughput and relia-
bility of data transmission to the server. Failing to address
such or alike such challenges impacts the functionality
and quality of the UIoT systems [9,17].
2.1.1 Data Analytics for IoT-driven Systems
IoT systems and their enabling technologies are funda-
mental to operationalising smart systems and infrastruc-
tures that vary from pervasive healthcare, autonomous
transportation, home automation, and digital oceanogra-
phy [2,8]. In recent years, academic research and devel-
opment and some key industrial players (e.g.,Google,
Microsoft, Amazon), have been continuously striving
to develop solutions for IoT-driven data analytics [9].
IoT devices and sensors produce and consume mission-
critical data such as route planning, environmental pol-
lution levels, and health symptoms. Such data can reveal
key insights about patterns and trends of a specic
mission (i.e. healthcare or transportation) [12]. In the
context of UIoTs, sensor data incorporate many oceanic
variables and their correlation, such as underwater tem-
perature, acidity, and concentration of marine life. Data
analytics can be applied to such oceanic data to extract
useful information about oceanography [16]. Data ana-
lytics for UIoTs can given decision support in smart
ocean initiatives that vary from ocean mining to classi-
fying marine life and predicting the quality of water [6].
Despite these benets, several challenges also exist while
developing and operationalising UIoTs. These challenges
relate to the scalability of analytics systems and the accu-
racy of predictions. To address such challenges there
is a need for systematic development of solutions that
can leverage existing data analytics techniques in UIoT
systems [8,9].
We conclude that IoT technologies and data analytics
methods provide strategic benets to design, develop,
and deploy UIoT systems. However, many challenges,
such as resource poverty and heterogeneity of sensors
and scalability and accuracy of data analytics, must be
ensured. Based on the background details and high-
lighted challenges we will discuss the proposed solution
anditsevaluationsindedicatedsectionsofthepaper.
3. RELATED WORK
We now present the most relevant existing research from
two dierent perspectives that include (i) solutions that
enable IoT-driven analytics of oceanic data in Section 3.1
and (ii) potentials and challenges analytics and analysing
of oceanic data in Section 3.2. At the end of the section,
a conclusive summary objectively compares the existing
research with a proposed solution to justify the scope and
contributionsofthisresearch.
3.1 Solutions for IoT-driven Analytics of Oceanic
Data
From a smart city perspective, recently many research
and development initiatives have been put forward
that exploit IoT solutions to enable intelligent and
autonomous computing such as smart healthcare, trans-
portation, and service robotics for home automa-
tion [10,12,18]. Such solutions from other application
domains, such as healthcare and vehicular networking,
can be customised to be applied to IoT-based and sensor-
driven analysis of ocean data [19,20]. The volume and
velocity of underwater data ingested by IoT devices
and sensors provide valuable insights to geologists and
oceanographers that were not possible with traditional
ways of data analytics [21]. In the past, oceanographic
studies relied on handheld devices that could be taken
Y. ALHARBI AND A. AHMAD: UNDERWATER INTERNET OF THINGS TO ANALYSE OCEANIC DATA 5
underwater to measure and analyse information such as
the surface of the sea, underwater temperatures, level
of chlorophyll, and underwater pollution [19]. Replac-
ing such ad-hoc methods of data collection, the research
in [22] proposes a technique where IoT-sensors collect
oceanic data and provide a web-based interface to display
the data to end-users. The solution addresses the issues
of manual data collection and its distribution; however,
no analytics are performed or critical insights extracted
about the underwater data. In the context of oceano-
graphic analytics data, the study [23]streamlinessome
critical challenges such as eciency and reliability of data
collection and processing that are missing in the existing
solutions.
UIoT,asaspecicgenreofIoTs,isclassiedasa
distributed interconnection (i.e. networking) of het-
erogeneous sensors (i.e. hardware) and objects (i.e.
software) to monitor ocean data [15,24]. Extending the
concept of UIoTs, the researchers in [25,26]haveapplied
machine learning approaches with IoTs, referred to as
UIoT (underwater IoT), to collect, process, and analyse
acoustic sounds. The UIoT solution streamlines the role
of interconnected sensors in data analytics and highlight
challenges about the reliability of data collection in such
circumstances [20]. Some recent initiatives, such as the
Ocean-of-Things initiative, [27]aimtopromotecollabo-
rative and open-source practices to develop scalable solu-
tions for monitoring underwater data. The study in [28]
exploits the concepts and principles of Ocean of Things to
track mammal marine life and their population. Research
eorts are underway to develop o-the-shelf components
(sensors and their control software) that can be deployed
in the context of pervasive healthcare, crowd-sensing,
and ocean mining [29].
3.2 Potential and Challenges for Analysing
Oceanic Data
Adopting and/or implementing data analytics solutions
can be complex, time-consuming and dicult to scale
due to the dynamic and context-sensitive nature of IoT
systems. To optimise the performance, decrease com-
plexity, and increase the reliability of UIoTs, machine
learning solutions are being proposed for underwater
IoTs [30]. An overwhelming majority of data analytics
techniques use dynamic models of data interaction that
are received via IoT devices [31]. Some recently devel-
oped solution for IoT-driven data analytics [32,33]aims
to address issues such as security [34], infrastructure
[32], and storage [33] aspects for UIoTs. From underwa-
ter UIoTs, many research studies have been conducted
for analytics data [35], extracting patterns from marine
data [36,37], and discovering knowledge about ocean
variables from multiple sources.
Conclusive Summary:Theframeworkin[9]reviews12
academic and industrial processes that enable IoT-driven
analytics of sensors data. The research provides empir-
ical insights, recommendations, and needs for futuristic
solutionsthatexploitIoTsensorstocollectdataandapply
dataanalyticsapproachestogatherkeyinsightsfromsen-
sor collected data. The proposed solution extends the
concepts from IoT-DA and specically applies them to
underwater data in the contest of UIoTs. The review of
the most existing research highlights a lack of research
and development that unies the concept of IoTs and
data analytics to operationalsie smart systems speci-
cally focused on ocean mining. There is a dire need
to develop frameworks and solutions beyond existing
research [19,22] to enable solutions that ensure the func-
tionality and desired quality of UIoTs [27,29]. Based on
the discussion of the related work, technical contribu-
tions and scope of the proposed solution are discussed
in the remainder of this paper.
4. RESEARCH METHOD
We now present details of the research method that high-
light some methodological steps to plan, conduct, and
evaluate the proposed research, as illustrated in Figure 3.
Step I Investigating the Existing Solutions on Data Ana-
lytics for IoTs
Before conceptualising the proposed solution, we need to
analyse state-of-the-art (recurring challenges and exist-
ingmethods)dataanalyticsforUIoTs,asinFigure3.
We followed the guidelines and recommendations of
conducting the systematic literature review of the exist-
ingresearch[38]. The systematic review helped us fol-
low many pre-dened steps to objectively analyse the
challenges, proposed solutions, and other aspects of
IoT-driven data analytics. Specically, as per the steps
of the SLR, we identied, qualitatively selected, and
Figure 3: Overview of Research Methodology
6 Y. ALHARBI AND A. AHMAD: UNDERWATER INTERNET OF THINGS TO ANALYSE OCEANIC DATA
synthesised existing solutions (peer-reviewed published
research studies) as part of the review. The results of the
review are already presented in Section 3. Investigating
the existing solutions has helped us identify the limita-
tions, potentials, and need for designing and architecting
futuristic solutions for UIoTs.
Step II 2 Designing the Architecture for Data Analytics of
UIoTs
After investigating the existing solutions, we develop
an architecture that conceptualises the solution for IoT-
driven data analytics, as in Figure 3. From an engineering
perspective, an architectural representation refers to a
collection of layers (i.e. system blueprint), their under-
lying objects, tools and technologies that solve problems
[1,2]. The architecture is primarily based on a layered
design that allows separation of concerns (i.e. data sens-
ing, data analytics and data presentation, see Figure 2)for
the proposed solution. Details of architectural design are
presented in Section 5.
Step III Validating the Proposed Framework
Finally, once the architecture has been designed and
implemented, we need to validate the solution to assess
its functionality and quality. We have used the ISO-IEC-
9126 model [17] to evaluate the quality of the proposed
architecture. We primarily focus on evaluating the sensor
throughput, availability, and reliability to measure system
functionality and quality. Results of the architectural val-
idation are presented in Section 6. The limitations of the
proposed solution as potential threats to the validity of
research are detailed in Section 6.4.
As per the details and steps of methodology from
Figure 3,theremainderofthispaperfocusesonarchi-
tectural development, its implementation, and validation
to streamline the core contributions of this research.
5. ARCHITECTURAL VIEW AND
IMPLEMENTATION OF UIOTS
We now present details of the proposed software archi-
tecture for data analytics of UIoTs. The architecture of
software-intensive systems as an IEEE standard 1471
standard [39]actsasablueprinttoprovideaglobal,
i.e. abstract view of the system by hiding complex and
implementation-specic details, as in Figure 4.Speci-
cally, as per IEEE standard 1471, software architecture
comprises architectural components and connectors that
enable a system to perform computations, data storage,
Figure 4: Architectural View of the Proposed UIoT System for
Analytics Oceanic Data. ([a] Domain View in terms of Real World
Deployment, [b] Architectural View for System Implementation)
and interactions [2]. To generalise the solution presenta-
tion, the architectural view is described (in Section 5.1)
as an implementation neutral and independent of any
specic tools and technologies of implementation. Tools,
technologies, and implementation details are provided
as supplementary material to this proposed research (in
Section 5.2).
5.1 Software Architecture for UIoTs
Architectural components act as computational entities
or data stores, whereas architectural connectors enable
interconnection between components and connectors.
We discuss the software architecture of the proposed
solution in the context of Figure 4.Thearchitecturalview
of the proposed solution, as in Figure 4, pinpoints the
benets, including
Architectural layering helps to decompose the over-
all solutions (by creating the separation of concerns)
in the form of (i) A Data Sensing Layer, (ii) A
Data Analytics Layer, and (iii) A User Interface (i.e.
Data Presentation) Layer, each of which is illustrated
in Figure 4and detailed below. By decomposing
the solution into architectural layers, architects and
system developers can focus on a specic concern
during system development. For example, IoT devel-
opers can focus on the data sensing layer, while
front-end developers can focus on the user interface
layer.
Y. ALHARBI AND A. AHMAD: UNDERWATER INTERNET OF THINGS TO ANALYSE OCEANIC DATA 7
Modularising system development by treating each
layerasanindividualmodule.Modulardevelop-
ment enables system development as a set of func-
tional routines and procedures that provides cus-
tomised functionality based on parameterised cus-
tomisation of the inputs [38,39].
Patterns or architectural styles can be reused as best
practices and reusable knowledge to solve recur-
ring problems of architectural design. In Figure 4,
by exploiting architectural abstractions we can apply
the layered architecture pattern to design the sys-
tem [1,4]. In the context of architectures for soft-
intensive systems [38], patterns promote reuse and
enhance the quality of system development.
In Figure 4,weprovidetwoviewsofthesolutionreferred
to as (i) domain (a.k.a. the real world) view and (ii) archi-
tectural (a.k.a. system design) view. First, as in Figure 4.
(a), domain or real world view highlights a real-world
presentation of the systems, operational environment
and the corresponding challenges. Second, as in Figure 4.
(b), architectural or system design view shows the build-
ing blocks of the system in terms of architectural com-
ponents and connectors of the system. The system view
helps designers and architects to sketch the system in
terms of architectural components and connectors of
the systems represented as either Unied Modeling Lan-
guage (UML), graph-based model, or state chart dia-
grams. For example, in the domain view, Figure 4illus-
tratestheunderwaterviewwithmarineobjects,suchas
marine life, underwater temperature, and pollution that
need to be measured and monitored. In contrast, the
system view shows the corresponding components like
Sense_Temperature to sense the underwater tempera-
ture using UIoTs and transmits it to other components
named Sense_SIIM. Further details of each layer and
both views are presented below.
5.1.1 Layer I: Data Sensing Layer Ingesting
Underwater Data
From the system implementation and operations point of
view,thedatasensinglayerdealswithcollectionofdata
from all the sensors deployed underwater. In Figure 4,
data sensing can be done with deployed sensors where
each deployed sensor has a unique identity (Sensor ID).
For example, the sensor with an identity Sensor-ID_A
monitors and captures the data about underwater tem-
perature (°C). The Scientic Instrument Interface Mod-
ule (SIIM) is deployed to act as a mediator between the
deployed sensors (data sensing layer) and server (data
analytics layer). Specically, the SIIM acts as a media-
tor to accumulate the data from all the sensors, adds
additional information (TimeDate, Geo-location, ID of
sensor), and transfers the sensor data to the server. As
in Figure 4, the underwater data ingested by SIIM from
Sensor-ID_A,i.e. underwater temperature (27 °C, Date-
Time: 190821_05:13:22 and Geo-location: 22.308824,
38.925052)foritstransmissiontotheserver.Dataare
continuously ingested from the sensors, but data trans-
mission from sensors to the server takes place period-
ically, usually N minutes interval (where N is a user-
preferred time interval in minutes). Figure 4illustrates
that the component named collectedData at the data
sensing layer transmits the data via a connector data-
Transmit to the component named dataStorage at the
server.
5.1.2 Layer II: Data Analytics Layer Analysing the
Underwater Data
The data analytics layer, as a bridge between data sensing
and data presentation, is primarily responsible for man-
aging and analysing the data collected from the sensors
for its presentation and useful insights for the stake-
holders. As the rst step, once the packaged data are
receivedfromthesensorlayerviapackagedData compo-
nent and are stored in the repository as in Figure 4.(b).
The data relating to the sensor (i.e. oceanic data), includ-
ing SIIM, Sensor_ID, Sensor_Value are stored indepen-
dent of other data such as the geo-location, date time,
andvalueoftemperature.Inadditiontothedataman-
agement, this layer supports data analytics, for example,
patterns of underwater temperature values (°C), their
impacts on acidity (pH value), and marine life. This layer
supports (i) analysing the historical data based on two-
time intervals (historical value =endtime–starttime)
and (ii) analysing the current value of the temperature.
For example, the data from Sensor: Sensor-ID_A trans-
mit data to the server: packaged data with current values:
underwater temperature is 24.8 °C on date 22092020 at
time 13:08:17 at location 22.308824, 38.925052.
5.1.3 Layer III: Data Presentation Layer Analysing
the Underwater Data
Finally, the data presentation layer provides a user inter-
face to the stakeholders to visualise the analysed data.
More specically, the stakeholders as the end-users of the
system rely on the visual presentation of the underwa-
terdatafortheirdecision-making.Forexample,theuser
can view the graphical bars of temperature values over a
while to get insights about temperature values over time.
Moreover, he/she can view the impact of temperature
on underwater acidity. The user interface layers provide
human interfacing with the backend data where human
users can query, view, and manipulate data via available
user interfaces.
8 Y. ALHARBI AND A. AHMAD: UNDERWATER INTERNET OF THINGS TO ANALYSE OCEANIC DATA
5.2 Implementation of the Solution
We now discuss the solution implementation that realises
the architecture from Figure 4. Specically, we outline
the tools and technologies for the implementation in
Section 5.2.1. Details about the solution source code and
algorithmic steps are presented in Section 5.2.2.
5.2.1 Tools and Technologies for Solution
Implementation
We illustrate the details of implementation based on
Figure 4that streamlines the tools and technologies for
the implementation, as illustrated in Figure 5.Speci-
cally, architectural layering in Figure 4(Section 5.1) acts
as a blueprint to exploit the tools and technologies and
their ow for developing and operationalising the solu-
tion in Figure 5,detailedbelow.
Implementing Layer 1 Sensing Layer: Ter b abl u IoT
sensors have been used to collect underwater data.
These sensors are used to collect data transmitted
to the SIIM. SIIM that accumulates data and acts as
a bridge between sensors and their data storage is
implemented as Raspberry Pi (a single board con-
troller) to manage data packaged into a le for its
transmission to the server. The packaged data from
thesensorsaremanagedasaCSV (Comma Sepa-
rated Values) le that can be manipulated with C#
(C-Sharp) source code.
Implementing Layer 2 Analytics Layer: This layer
manages and analyses the data provided from the
sensing layer. Therefore as a rst step, sensors’ data
fromCSVlesarestoredintheMicrosoft SQL
Server. The data server is deployed and managed
on Microsoft Windows Platform (Windows 7). The
data server can be queried using the SQL (Struc-
tured Query Language) to retrieve the data as and
when required. Based on user querying, the analysed
data can be presented as reports using the CSV le.
Python language is used to handle the data libraries
(e.g. Numpy, and Pandas).
Implementing the Layer 3 User Interface Layer:
Finally, to present the data to the end-users Analytics
Reports are generated at the server-side and pre-
sented at the client-side. Specically, to manage and
process the generated report, the server-side imple-
ments C# source code to process the data retrieved
using SQL.Theserver-sidecodeisimplemented
using Microsoft Visual Studio. To present the data
to the user, the client-side uses JavaScript. The gen-
erated reports can be viewed by the user using an
HTML (Hypertext Markup Language) compatible
web-client (browser).
Figure 5: Integration of Tools, Technologies, and their Flow to
Implement the Solution, ([a] Sensors for IoTs, [b] Data manage-
ment for analytics, [c] Presentation for end-users)
Figure 6: Case Study on Ocean Data Analytics, ([a] Location 1, [b]
Location 2, [c] Sensor Selection, [d] Location Selection, [e] Sensor
Value)
5.2.2 Case Study on Ocean Data Analytics (System
Prototype)
We now present the case study on analytics oceanic data,
as illustrated in Figure 6. Specically, Figure 6highlights
that we were able to collect the data from (a) Red Sea,
Figure 6(a,b) Arabian Gulf, Figure 6(b).
Thesetwolocationsprovidethenecessaryoceanicdata
to validate the system’s functionality.
Y. ALHARBI AND A. AHMAD: UNDERWATER INTERNET OF THINGS TO ANALYSE OCEANIC DATA 9
Figure 7: Overview of Evaluating the Sensors’ Throughput, ([a] X-Axis: Sensor Data Transmission Number of Days, [b] Y-Axis: Sensor Data
Transmission in Kilobytes)
As presented in Figure 6(c), there were ve deployed
sensors (S1: Temp e r atu r e , S2: pH value, S3: Dissolved
Oxygen, S4: Chlorophyll, S5: Sea Level)andone
of the sensors to collect underwater temperature is
selected. Figure 6(d) shows the selection of a specic
location (Arabian Gulf with coordinates 26.7001897,
50.2480422)thatcanenabletheusertoselecttheloca-
tionforsensordatabasedonthelistofavailablelocations.
Finally, as in Figure 6(e), the interface allows the selec-
tion of sensors, correlation, and the location for the value
of a specic sensor. For example, Figure 6(e) presents the
value of temperature on the value of pH (acidity) at both
locations.
It is vital to mention that the developed system is a
prototypeonlythatprovidestoolsupportasaproof-of-
the-concept. We are in the process of further developing
the system with the incremental addition of new sen-
sors and extended functionality. In this regard, we have
deployed the source code1(executable specications and
algorithmic details) to make it available for the public
and move towards an open-source initiative for collab-
orative and community-driven development of systems
that can operationalise ocean analytics and complement
initiatives like Ocean-of-Things [27]
6. VALIDATION OF THE SOLUTION
To validate the solution, we use the ISO/IEC-9126 model
for evaluating software and system quality [17]. Specif-
ically, based on the guidelines and criteria in ISO/IEC-
9126, we aim to evaluate (i) sensors’ throughput (i.e.
performance) in Section 6.1, (ii) query response time
Figure 8: Overview of Query Response Time for Data Retrieval,
([a] X-Axis: Number of Sensors, [b] Y-Axis: Time Taken in
Milliseconds)
(i.e. eciency) in Section 6.2, and (iii) ease of use (i.e.
usability) in Section 6.3, each detailed below. The visu-
alised summary of the evaluation results is presented in
Figures 79to guide the discussion of evaluation.
Evaluation Congurations: The congurations for evalu-
ation refer to hardware and software components, tools,
and other artefacts that need to be congured to eval-
uate the proposed solution, as shown in Figure 4.The
hardware components primarily included the sensors
(Ter a B lu) with SIIM implemented as Raspberry Pi with
data management and analytics executed on a Windows
7Machine(corei7and32GBruntimememory).The
software set-up and congurations include the Microsoft
dot net framework (C#, SQL Server). The scripts to eval-
uate sensors’ throughput and query response automate
10 Y. ALHARBI AND A. AHMAD: UNDERWATER INTERNET OF THINGS TO ANALYSE OCEANIC DATA
Figure 9: Likert Scale to Summarise End-Users’ Responses for
System Usability, ([a] Number of Participants (U1 U7), [b] 5-Point
Criteria for Evaluating Usability)
the system-level testing, written in Python language, exe-
cuted on Jupyter Notebook.
6.1 Evaluating Sensors’ Throughput
(Transmission Stability)
Sensors’ throughput is a means to analyse the data pro-
duced from the underwater sensors (see Figure 4)and
canhelpusassessthestabilityofdatacommunicated
from each sensor. Analysing the sensors’ throughput can
help us to identify (a) any disruptions, (b) variations,
(c) inconsistencies of data received from the sensors over
a specic period. As in Figure 7,sensors’throughput
data in Kilobytes (KBs) are presented on the vertical axis,
and the horizontal axis represents the time duration to
gather that data (number of days) for experimental pur-
poses. Sensors, throughout from all of the ve sensors,
are illustrated as a specic and dedicated bar graph that
moves along the the horizontal and vertical axes. Specif-
ically, the sensors’ data transmission values represent the
amount of data transmitted in KBs over a specic number
of days (i.e. data over time). For experimental purposes
andtopresenttheresultshere,wehavemeasuredthesen-
sors’ throughput for 30 consecutive days (1-month trials).
The SIIM collects data from the server and logs it every
15 min to the server. The sensors’ throughput data in
Figure 6represent a stable and consistent throughput by
all the sensors. For example, the temperature sensor (S1
as Blue coloured bar) consistently transmitted data (>
2000 KB) during each of the 30-day trials.
We conclude that the uctuation in sensor values for
dissolved oxygen (S4 as Green coloured bar) that var-
ied between 1200 and 1700 KB data over 30 days. This
variation suggests that for this specic sensor, data read-
ing is not as frequent as the temperature sensor, or there
is a further need to investigate the throughput by this
sensor.
6.2 Evaluating the Query Response Time (Data
Retrieval Time)
Query response time refers to the time it takes to query
the dataset and retrieve the results. Evaluating the query
response time is vital to analyse system performance to
retrieve the desired datasets (structured and unstruc-
tured) in a specic period. Figure 8illustrates the query
response time for data retrieval from the analytics layer
(Layer 2) to the presentation layer (Layer 3). Figure 8
presents accumulated values of 00 trials, where the aver-
ageofthevalueof10trialsisrepresentedasoneinstance
of the bar graph. For example, as in Figure 8,theblue
colouredbargraphonthersttrialindicatesthatbased
on 10 trials, it took an average of 13 milliseconds to
retrieve data related to a single sensor. We have three
cases of query response time as
Response time from one sensor:Thequeryresponse
for data from one sensor, such as values of tempera-
ture (°C).
Response time from two sensors:Thequeryresponse
time for data from two sensors such as the impacts
of temperature on the acidity of the water (impacts
of °C on pH)
Response time from multiple sensors:Thequery
response time of more than two sensors, such as the
impacts of temperature on the acidity of the water
and concentration of marine life (impacts of °C on
pH on dissolved oxygen: DO).
Figure 8represents a bar graph for the query response
time that illustrates many trials (horizontal axis) vs. the
time taken to retrieve the query data (time in millisec-
onds). The query response values seem consistent and
reasonable. Based on 100 trias, the response time for sin-
gle sensors is (11–15 milliseconds), for two sensors, it
is (15–18 milliseconds), and for multiple sensors (19–27
milliseconds).
6.3 Evaluating the Usability of System (Ease of
System Use)
As part of the evaluation, we need to evaluate system
usability by engaging potential end-users to use sys-
tem interfaces and seek their feedback. Specically, we
engaged 7 end-users (2oceanographers,3dataanalysts,
and 2 geologists, identied as U1 to U7 in responses) and
provided them a tutorial about the use of the system (see
Figure 6as an illustration of some of the user interfaces).
Y. ALHARBI AND A. AHMAD: UNDERWATER INTERNET OF THINGS TO ANALYSE OCEANIC DATA 11
After getting the brieng, the end-users were asked to
use the system and answer a few questions as part of the
exit survey on system usability. We used 5-point crite-
ria to evaluate system usability, as in Figure 9.Figure9
provides a graphical summary (Likert scale: Strongly
Agree, Agree, Neutral, Disagree, Strongly Disagree)of
theend-usersfeedbackaboutsystemusability,asdetailed
below. In addition to the 5-point criteria (Customisation,
Information Presentation, Data Query ing, Navigation,
Interface Usability), we also asked for open-ended feed-
back from the end-users, if they like to suggest or recom-
mend any future improvements in system usability.
The results of end-users- based evaluations suggest that
the overall majority of the end-users Strongly Agree or
Agree with the usability criteria of the system. Some of
the end-users had disagreed on data customisation, inter-
face navigation, and usability. While investigating the
reason for such disarmament, via open-ended question,
we identied that end-users U4 indicated that “some-
times the provided information was quite overwhelm-
ing and that creates a confusion during navigation and
data customisation”. Moreover, as part of end-user’s sug-
gestion“thereshouldbeastep-wiseguideonhowto
customise the data on correlating sensors”. Similarly, the
end-user U7 highlighted that . .. interfaces can be made
more usable with audio-visual support for critical actions
like selecting a particular sensor to view its data”.
Based on the overall feedback and end-users responses,
we concluded that system usability is appropriate; how-
ever, there is a need for better information management
and audio-visual support to empower end-users while
navigating through the system or customising the data as
part of future work.
6.4 Threats to Validity of Research
Threats to the validity of research refer to some con-
ditions or pre-dened constraints that could potentially
invalidate the results. These threats represent potential
limitations or specic constraints as a consequence of
the design, implementation, and evaluation of the solu-
tion. These threats need to be highlighted, so they can be
addressed as per future work to improve the rigour of the
solution.
Threats to internal validity relates to the design of
research solutions that may have some inherent bias in
methodology or solution construct. Specically, we have
a limited number of sensors and controlled experiments
that provide data. Increased trails, more sensors, and
diversity of data collection can help to minimise this
threat as part of future work.
–ThreatstoExternalvalidityrefers to potential limita-
tions and constraints that can invalidate or impact the
results of external validation. In the scope of the current
solution,weonlyrelyonasinglecasestudyforevalua-
tion (Section 6.1–Section 6.3). However, future research
requires more case studies and use cases to accumulate
diversied underwater data sets to improve the rigour
of system validation as part of external validity of the
solution.
7. CONCLUSIONS AND NEEDS FOR FUTURE
RESEARCH
7.1 Conclusive Summary
Internet of Things (IoTs)-driven systems are the back-
bone for sensor-driven technologies to enable smart
systems and infrastructures. The Underwater Internet
of Things (UIoTs) represent a specic genre of IoTs
that orchestrate many context-sensitive sensors and
dynamically composable software services to ingest data
from under the sea. The oceanic data contains valu-
able information and exploratory artifacts, including but
not limited to, types and classication of marine life,
quality of seawater, and underwater minerals. In this
research, we have unied the concept of UIoTs with data
analytics to (i) enable the sensor-driven collection of
context-sensitive oceanic data, (ii) apply data analytics
to derive useful information, and (iii) provide reporting
and data-driven intelligence to the stakeholders. We have
developed a layered architecture and evaluated it using a
real-world case study on analysing oceanic data. Results
of evaluation streamline the usability and eciency of the
architecture in terms of (a) sensors throughput (i.e. sta-
bility), (b) query response time (i.e. performance), and
(c) system usability (i.e. ease of use). The proposed archi-
tecture complements the existing research and develop-
ment on IoTs for smart cities and specically contributes
to UIoTs in smart oceans. Results and their evaluation
demonstrate the architecture’s applicability as a proposed
solution and provide guidelines to engineer and develop
emerging and next generation of UIoTs for ocean data
analytics. The primary contributions of this research can
be outlined as
Exploiting the principle and practices of software
architecture that unies IoTs and data analytics to
systematically architect, develop, and validate a spe-
cic genre of IoT systems, referred to as UIoTs.
12 Y. ALHARBI AND A. AHMAD: UNDERWATER INTERNET OF THINGS TO ANALYSE OCEANIC DATA
Validatingtheproposedsolutionwithacasestudy
on ocean data analytics for scenario-based demon-
stration and criteria-based evaluation of the pro-
posed solution.
Thenoveltyoftheproposedsolutionisbasedonarchi-
tectingUIoTs,anemergingclassofIoTsystemsthat
enable ocean data analytics in the context of smart sys-
tems and infrastructures. The implications for proposed
research can be benecial for
Advancing state-of-the-art on IoTs, particularly uni-
fying the concept of UIoTs with methods and tech-
niques of data analytics for smart ocean technolo-
gies.
Developing a layered architecture as a reference
model and providing guidelines and validation crite-
ria to engineer emerging and next-generation solu-
tions for IoTs.
7.2 Dimensions of Future Research
Our vision for future research mainly focuses on expand-
ing the scope of solution validation with more case stud-
ies and practical scenarios that further validate the solu-
tion. We plan to incorporate more case studies to eval-
uate the solution’s eectiveness with dierent data and
deployment congurations. IoT systems in general and
UIoTs contain critical data that needs to be secure to
ensure its integrity and trustworthiness. Future research
also requires the implementation of secure management
andtransmissionofdatatoincreasethereliabilityand
trustworthiness of the UIoTs.
Future work can also extend the current architecture
(Figure 4) to develop a framework that assists with min-
ing patterns of oceanic data. Patterns can help to investi-
gate recurring trends of data in an oceanic context.
NOTE
1. GitHub: IoTOcean: https://tinyurl.com/IoUTSource
DISCLOSURE STATEMENT
No potential conict of interest was reported by the author(s).
FUNDING
“This research has been funded by Scientic Research Dean-
ship at University of Ha’il Saudi Arabia through project
number BA-2015”.
REFERENCES
1. M.Fahmideh,A.A.Abbasi,A.Behnaz,J.Grundy,andW.
Susilo, “Software Engineering for Internet of things,” IEEE
Trans. Sof t ware Eng. , 1–18, 2021.DOI:10.1109/TSE.2021.
3070692.
2. A. Gaur, B. Scotney, G. Parr, and S. McClean, “Smart
city architecture and its applications based on IoT,”
Procedia. Comput. Sci., Vol. 52, pp. 1089–1094, 2015.
DOI:10.1016/j.procs.2015.05.122
3. Available: https://www.idc.com/getdoc.jsp?containerId =
prAP46737220.
4. X. Larrucea, A. Combelles, J. Favaro, and K. Taneja, Soft-
ware engineering for the internet of things,” IEEE Softw.,
Vol. 34, no. 1, pp. 24–28, 2017.DOI:10.1109/MS.2017.28
5. A. Zanella, N. Bui, A. Castellani, L. Vangelista, and
M. Zorzi, “Internet of things for smart cities,” IEEE
Internet Things J., Vol. 1, no. 1, pp. 22–32, 2014.
DOI:10.1109/JIOT.2014.2306328
6. S.Battula,M.N.V.S.S.Kumar,S.K.Panda,U.M.Rao,G.
Laveti, and B. Mouli. “Online ocean monitoring using edge
IoT.” In Global Oceans 2020: Singapore–US Gulf Coast,
2020, pp. 1–7. IEEE.
7. S. Jiang, “Marine Internet for Internetworking in oceans:
Atutorial,Future Internet, Vol. 11, no. 7, pp. 146, 2019.
DOI:10.3390/11070146
8. V.Moustaka,A.Vakali,andL.G.Anthopoulos,“Asys-
tematic review for smart city data analytics,” ACM Com-
puting Surveys (CSUR, Vol. 51, no. 5, pp. 1–41, 2018.
DOI:10.1145/3239566
9. A. Ahmad, et al., “Software Engineering for IoT-driven
data analytics applications,” IEEE. Access., Vol. 9, pp. 48197
–48217, 2021.DOI:10.1109/ACCESS.2021.3065528
10. S. America. Internet of things and system control, 2021.
11. C. Manville, et al.Mappingsmartcitiesintheeu,2014.
12. J.-H. Lee, and M. G. Hancock. Toward a framework for
smart cities: A comparison of Seoul, San Francisco and Ams-
terdam, research paper. California: Yonsei University and
Stanford University, 2012.
13. D.Vesset,C.Gopal,C.W.Olofson,D.Schubmehl,S.Bond,
andM.Fleming.“Worldwidebigdataandanalyticssoft-
ware forecast, 2018–2022,” IDC, Framingham, MA, USA,
Tech. Rep. US44803719, 2018.
14. A. Yassine, S. Singh, M. S. Hossain, and G. Muhammad,
“Iot big data analytics for smart homes with fog and
cloud computing,” Future Gener. Comput. Syst., Vol. 91,
pp. 563–573, 2019.DOI:10.1016/j.future.2018.08.040
15. T.Qiu,Z.Zhao,T.Zhang,C.Chen,andC.L.PhilipChen,
Underwater Internet of Things in smart ocean: system
Y. ALHARBI AND A. AHMAD: UNDERWATER INTERNET OF THINGS TO ANALYSE OCEANIC DATA 13
architecture and open issues,” IEEE Trans. Ind. Inf., Vol. 16,
no. 7, pp. 4297–4307, 2019.DOI:10.1109/TII.2019.2946618
16. D.M.Toma,T.OReilly,J.delRio,K.Headley,A.Manuel,
A. Bröring, and D. Edgington, “Smart sensors for inter-
operable smart ocean environment,” in OCEANS 2011
IEEE-Spain, 2011, pp. 1–4. IEEE.
17. J. Estdale, and E. Georgiadou, Applying the iso/iec
25010 quality models to software product,” in European
Conference on Software Process Improvement, Springer,
pp. 492–503.
18. E. Ahmed, I. Yaqoob, A. Gani, M. Imran, and M.
Guizani, “Internet-of-Things-based smart environments:
state of the Art, taxonomy, and open research chal-
lenges,” IEEE Wirel. Commun., Vol. 23, pp. 10–16, 2016.
DOI:10.1109/MWC.2016.7721736
19. C.-C. Kao, Y.-S. Lin, G.-D. Wu, and C.-J. Huang, “A com-
prehensive study on the underwater Internet of Things:
applications, challenges, and channel models,” Sensors,
Vol. 17, pp. 1477, 2017.DOI:10.3390/s17071477
20. T.Qiu,Z.Zhao,T.Zhang,C.Chen,andC.P.Chen,“Under-
waterinternetofthingsinsmartocean:systemarchi-
tecture and open issues,” IEEE Trans. Ind. Inf., Vol. 16,
pp. 4297–4307, 2019.DOI:10.1109/TII.2019.2946618
21. D. Halpern, Satellites, oceanography and society.Elsevier,
2000.
22. O.L.Osen,H.Wang,K.B.Hjelmervik,H.Sch,et al., “Orga-
nizing data from industrial Internet of Things for maritime
operations,” in OCEANS 2017-Aberdeen, IEEE, pp. 1–5.
23.C.Hu,Y.Pu,F.Yang,R.Zhao,A.Alrawais,andT.
Xiang, “Secure and ecient data collection and storage
of IoT in smart ocean,” IEEE Internet Things J.,Vol.7,
pp. 9980–9994, 2020.DOI:10.1109/JIOT.2020.2988733
24. Fraunhofer. Smart Ocean Technologies Solutions for
Responsible Ocean Use, 2021.
25. E. H. Belghith, F. Rioult, and M. Bouzidi, Acous-
tic diversity classi_er for automated marine big data
analysis,” in 2018 IEEE 30th International Conference
on Tools with Articial Intelligence (ICTAI), IEEE,
pp. 130–136.
26. J. Song, H. Xie, and Y. Feng, “Correlation analysis method
for ocean monitoring big data in a cloud environment,”
J. Coast. Res., 24–28, 2018.
27. A. A. Saucan, and M. Z. Win, “Information-seeking sen-
sorselectionforocean-of-things,IEEE Internet Things
J., Vol. 7, pp. 10072–10088, 2020.DOI:10.1109/JIOT.2020.
2992509
28. J.Waterston,J.Rhea,S.Peterson,L.Bolick,J.Ayers,and
J. Ellen, “Ocean of things: aordable maritime sensors
with scalable analysis,” in OCEANS 2019-Marseille, IEEE,
pp. 1–6.
29. N. Wright, and H. Chan, “Low-cost internet of things
ocean observation,” in OCEANS 2016 MTS/IEEE Mon-
terey, IEEE, pp. 1–5.
30. M. Popescu, P. J. Dugan, M. Pourhomayoun, D. Risch,
H. W. Lewis III, and C. W. Clark. Bioacoustical peri-
odic pulse train signal detection and classication using
spectrogram intensity binarization and energy projection,
arXiv preprint arXiv:1305.3250; 2013.
31. J. Schmidhuber, “Deep learning in neural networks: An
overview,” Neural Netw., Vol. 61, pp. 85, 2015. 117.
DOI:10.1016/j.neunet.2014.09.003
32. Y. Demchenko, C. De Laat, and P. Membrey, “De_ning
architecture components of the big data ecosystem,” in
2014 International Conference on Collaboration Tech-
nologies and Systems (CTS), IEEE, pp. 104–112.
33. Y. Du, Z. Wang, D. Huang, and J. Yu, “Study of migra-
tion model based on the massive marine data hybrid
cloud storage,” in 2012 First International Conference
on Agro-Geoinformatics (Agro-Geoinformatics), IEEE,
pp. 1–4.
34. K.Yang,X.Jia,K.Ren,R.Xie,andL.Huang,“Enablinge-
cient access control with dynamic policy updating for big
data in the cloud,” in IEEE INFOCOM 2014-IEEE Confer-
ence on Computer Communications, IEEE, pp. 2013-2021.
35. E. Y. Chang, H. Bai, and K. Zhu, “Parallel algorithms for
analytics large-scale rich-media data,” in Proceedings of
the 17th ACM International Conference on Multimedia,
pp. 917–918.
36. C.K.-S.Leung,andY.Hayduk,“Analyticsfrequentpatterns
from uncertain data with MapReduce for big data analyt-
ics,” in International Conference on Database Systems for
Advanced Applications, Springer, pp. 440–455.
37. C.K.-S.Leung,R.K.MacKinnon,andF.Jiang,“Reduc-
ing the search space for big data analytics for interesting
patterns from uncertain data,” in 2014 IEEE International
Congress on Big Data, IEEE, pp. 315–322.
38. D. Budgen, and P. Brereton, “Performing systematic liter-
ature reviews in software engineering,” in Proceedings of
the 28th International Conference on Software Engineer-
ing, 2006, pp. 1051–1052.
39. R. Kumar, and G. Dhiman, “A comparative study of Fuzzy
optimization through Fuzzy number,” International Jour-
nal of Modern Research, Vol. 1, pp. 1–14, 2021.
14 Y. ALHARBI AND A. AHMAD: UNDERWATER INTERNET OF THINGS TO ANALYSE OCEANIC DATA
AUTHORS
Ya s s e r A l h a r b i is currently an Assistant
Professor in the College of Computer Sci-
ence & Engineering at the University of
Hail in Saudi Arabia. He obtained his
Ph.D. in Computer Sciences from the Uni-
versity of Essex, the United Kingdom in
2018. He did his MS in Information Tech-
nology, in 2009 at the Queensland Uni-
versity of Technology, Australia. His areas of interest are Big
Data in the Network, IoT, Fog computing and Grid and Cloud
Computing Systems & Network which include Resource and
applicationmanagementinclouds(VMsAllocationandmigra-
tion), Performance analysis, modelling and optimization, and
Energy eciency (Green Cloud).
Email: y.alharbi@uoh.edu.sa
Aakash Ahmad is currently working as an
Assistant Professor at the College of Com-
puter Science and Engineering, University
of Ha’il, Saudi Arabia. Aakash completed
his PhD in Software Engineering from the
School of Computing, Dublin City Uni-
versity, Ireland. Aakash also worked as a
Postdoctoral Researcher in Software and
Systems Sections, IT University of Copenhagen, Denmark.
Aakash’s research interest are in the areas of Software Engineer-
ing for Internet of Things and Mobile Computing systems.
Corresponding author. Email: a.abbasi@uoh.edu.sa
... Underwater wireless optical communication (UWOC) [1], [2] is a high bandwidth, low-cost deploy-ability system. There is a lot of interest in UWOC because of its applications in the internet of underwater things (IoUT) [3], [4], underwater wireless sensor networks (UWSN) [5], oil, gas and mineral exploration, navigation, military etc. The major impairment in a UWOC link is underwater turbulence [6], [7] which causes irradiance fluctuations. ...
Article
Full-text available
The major problems in an underwater wireless optical communication (UWOC) link are turbulence induced fading and pointing errors. In this paper, we have investigated the bit error rate (BER) performance of spatially coupled low-density parity-check (SC-LDPC) coded horizontal UWOC link over a strong turbulent channel model with pointing errors. The performance of this link for different channel and code parameters has been studied using simulations. It has been observed that a rate 1/2 ARJA protograph based SC-LDPC code with graph lifting factor of 256 gives a coding gain of 47 dB at a BER of 10 -4 for strong turbulence channel model with pointing errors. An analytical BER expression for UWOC link under strong turbulence with pointing errors for On-Off Keying modulation technique has been derived. A multidimensional protograph based extrinsic information transfer algorithm has been developed to obtain the decoding thresholds for different channel parameters and code rates. We have also studied the SC-LDPC coded vertical UWOC link performance for some specific strong turbulence channel parameters with pointing errors and observed that as the link length increases from 20 m to 40 m , the performance gap between the hypothetical and cascaded channel models increases from 1.1 dB to 5.5 dB .
... It encompasses both stationary and mobile sensors/actuators and underwater robots. [1]. Subsurface monitoring and monitoring in seas and oceans is critical due to their diverse environmental and industrial applications [2]. ...
... Underwater sensors and objects can thus collect real-time data, support decision-making, planning and protect the biodiversity of marine life. The advantages are the speed, simplicity and efficiency of data acquisition analysis for various stakeholders [1,17,6]. ...
Conference Paper
Full-text available
Water Smart Cities reflect the trend of climate change, but it is necessary to implement new approaches to water resources management, which will also reflect other trends, such as water stress levels, urbanization or technology development.. The aim of the article is to identify the latest approaches to water resources management in the concept of Water Smart Cities, summarize the findings of the analysis and point out the benefits of their use for today's cities. By analyzing the relevant articles, 12 innovative approaches and technologies for water management in the urban environment were identified. Their benefits are mainly lower consumption, efficient management, prediction, monitoring or protection of biodiversity, quality and quantity of water resources. The main findings of the article can serve as an analytical basis for strategic management of cities, academia, water institutions or communities of interested citizens.
Preprint
Full-text available
Oceans cover about 70% of the Earth's surface, and about 95% remain unexplored for various reasons. Underwater wireless communication has been revolutionised with the help of Internet of Things networks. Recently, applications such as collecting marine data, marine monitoring, disaster prevention, historical exploration, oil and gas exploration, shipwreck exploration, maritime security, and monitoring of aquatic species and water pollution, and several applications have become possible. However, the problem of routing, information transfer, and resource preservation in the fluid underwater environment remains unsolved. In this research, we were looking for a solution to make routing more efficient, considering the various and practical criteria for the quality of service of the underwater IoT network. One of the outstanding features of this research is the possibility of dynamically weighing the parameters involved in routing and taking optimal and distributed decisions between network components. The proposed method has achieved acceptable results in terms of quality of service compared to recent methods.
Article
Full-text available
Internet of Things based systems (IoT systems for short) are becoming increasingly popular across different industrial domains and their development is rapidly increasing to provide value-added services to end-users and citizens. Little research to date uncovers the core development process lifecycle needed for IoT systems, and thus software engineers find themselves unprepared and unfamiliar with this new genre of system development. To ameliorate this gap, we conducted a mixed quantitative and qualitative research study where we derived a conceptual process framework from the extant literature on IoT, through which 27 key tasks for incorporation into the development processes of IoT systems were identified. The framework was then validated by means of a survey of 127 IoT practitioners from 35 countries across 6 continents with 15 different industry backgrounds. Our research provides an understanding of the most important development process tasks and informs both software engineering practitioners and researchers of the challenges and recommendations related to the development of next-generation of IoT systems.
Article
Full-text available
Internet of Things Driven Data Analytics (IoT-DA) has the potential to excel data-driven operationalisation of smart environments. However, limited research exists on how IoT-DA applications are designed, implemented, operationalised, and evolved in the context of software and system engineering life-cycle. This article empirically derives a framework that could be used to systematically investigate the role of software engineering (SE) processes and their underlying practices to engineer IoT-DA applications. First, using existing frameworks and taxonomies, we develop an evaluation framework to evaluate software processes, methods, and other artefacts of SE for IoT-DA. Secondly, we perform a systematic mapping study to qualitatively select 16 processes (from academic research and industrial solutions) of SE for IoT-DA. Thirdly, we apply our developed evaluation framework based on 17 distinct criterion (a.k.a. process activities) for fine-grained investigation of each of the 16 SE processes. Fourthly, we apply our proposed framework on a case study to demonstrate development of an IoT-DA healthcare application. Finally, we highlight key challenges, recommended practices, and the lessons learnt based on framework’s support for process-centric software engineering of IoT-DA. The results of this research can facilitate researchers and practitioners to engineer emerging and next-generation of IoT-DA software applications.
Article
Full-text available
Ever-increasing human activities in oceans require handy, reliable and cost-effective high-speed network access similar to terrestrial Internet services to be available on and under water. However, terrestrial Internet cannot be extended to oceans seamlessly due to huge differences between terrestrial and oceanic environments, while satellite services are still very expensive especially for ordinary users with communication quality susceptible to weather conditions, and cannot cover underwater networks either. Therefore, marine Internet, proposed about six years ago, is the earliest scheme that tries to systematically address the internetworking issue in oceans and still in its infancy stage. This tutorial aims to introduce the principle, architecture and applications of marine Internet, along with discussion on oceanic environments for communication, currently available and under developing communication systems in oceans, as well as challenging issues necessary for further studies to foster the development of marine Internet.
Article
Full-text available
Smart cities (SCs) are becoming highly sophisticated ecosystems at which innovative solutions and smart services are being deployed. These ecosystems consider SCs as data production and sharing engines, setting new challenges for building effective SC architectures and novel services. The aim of this article is to “connect the pieces” among Data Science and SC domains, with a systematic literature review which identifies the core topics, services, and methods applied in SC data monitoring. The survey focuses on data harvesting and data mining processes over repeated SC data cycles. A survey protocol is followed to reach both quantitative and semantically important entities. The review results generate useful taxonomies for data scientists in the SC context, which offers clear guidelines for corresponding future works. In particular, a taxonomy is proposed for each of the main SC data entities, namely, the “D Taxonomy” for the data production, the “M Taxonomy” for data analytics methods, and the “S Taxonomy” for smart services. Each of these taxonomies clearly places entities in a classification which is beneficial for multiple stakeholders and for multiple domains in urban smartness targeting. Such indicative scenarios are outlined and conclusions are quite promising for systemizing.
Article
We propose a general sensor selection (SS) methodology for ocean-of-things (OoT) where a sensing network performs multi-object tracking (MOT) under resource constraints. SS methods address the combinatorial problem of determining the best subset of sensors that maximizes a suitable reward function for a fixed cardinality. The novelty of this work is twofold. First, we propose a tractable information-theoretic reward function for MOT-OoT with an unknown and time-varying number of objects such as ocean vessels. A tractable reward function is essential in order to rapidly evaluate a sensor subset, which is crucial in the high-dimensional problems encountered in OoT. Second, we propose a general cross-entropy SS (CE-SS) methodology that efficiently estimates the probabilities of sensor activations and determines the optimal sensor subset according to the proposed reward function and under the imposed cardinality constraint. The CE-SS algorithm avoids exhaustive searching over the space of all sensor subsets, which is intractable for most OoT applications. The CE-SS methodology, coupled with the proposed reward function, is capable of selecting sensors that lead to more accurate estimators for both the number of vessels and their trajectories than random selection. We demonstrate the effectiveness of our method via numerical simulation and, for multi-vessel tracking in OoT, with a network of acoustic sensors deployed off the coast of Italy.
Article
Due to the abundant marine resources, smart ocean has attracted much attention of the government, industry and academy. The Internet of Things (IoT) architectures for smart ocean have been proposed to collect a various of data from ocean, thereby assisting environmental protection, military reconnaissance and so on. However, few researchers have paid attention to the security and privacy issues of data collection and transmission. In this paper, for the unreliable underwater environment, we present a secure, efficient and complete data collection, transmission and storage scheme for IoT in smart ocean. Specially, to prolong the lifetime of underwater node, two novel data compression algorithms (lossy data compression algorithm (LCA) and lossless data compression algorithm (NLCA)) are also proposed. Moreover, due to the vulnerability of underwater nodes, we also propose a corresponding IoT framework and data collection pattern to resist single point failure attack. Besides, to guarantee the confidentiality, reliability, and integrity of transmitting data, Elliptic Curve-ElGamal (EC-ElGamal) and Elliptic Curve Digital Signature Algorithm (ECDSA) are employed. Consensus algorithm and blacklisting mechanism are also employed to detect and address failure or malicious nodes. Finally, The security analysis demonstrates that our scheme is able to resist many typical attacks for underwater nodes, such as manipulation attacks, Distributed Denial of Service (DDoS) attacks, malicious node injection attacks and so on. Additionally, relevant experimental results show that the scheme are feasibility and efficiency.