Easy and Secure Handling of Sensors and Actuators as Cloud-Based Service

Abstract

The internet of things is rapidly becoming a fact of life. There remain, however, various challenges with respect to the concept of Industry 4.0 before this new reality is fully integrated into the industrial sector. One such challenge is ensuring security, an essential issue in terms of both management and operational technology involving the configuration and administration of physical equipment. Another is integrating equipment from different manufacturers which must be configured and/or administered with incompatible protocols. Likewise, remote access to systems must be carried out from a browser in the web frame. This paper presents a set of tools, available as a cloud-based service, which facilitates access to the physical equipment located in a business. The access enables the end user to see and/or modify the configuration and management of data in a secure, controlled, organized and collaborative manner. In order to unify different protocols, the service makes use of the Modbus TCP/IP (Transmission Control Protocol/Internet Protocol) protocol, one of the most commonly used by industry in local networks. The local Modbus communications are encapsulated in the Websocket protocol to make them accessible from the cloud. The tools proposed in this paper are based on open hardware platforms and free software, and they permit local and remote sensors and actuators to be fully and easily accessed. Hence, their chief advantage is that they allow access from any internet connection, via a browser, to the physical equipment located in an industry by means of user profiles in accordance with the security level set by each individual company.

Full text

Received December 16, 2019, accepted December 31, 2019, date of publication January 10, 2020, date of current version January 17, 2020.
Digital Object Identifier 10.1109/ACCESS.2020.2965639
Easy and Secure Handling of Sensors and
Actuators as Cloud-Based Service
REYES SÁNCHEZ-HERRERA , MARCO A. MÁRQUEZ ,
AND JOSÉ M. ANDÚJAR , (Senior Member, IEEE)
Huelva University, Escuela Técnica Superior de Ingeniería, 21007 Huelva, Spain
Corresponding author: Reyes Sánchez-Herrera (reyes.sanchez@die.uhu.es)
ABSTRACT The internet of things is rapidly becoming a fact of life. There remain, however, various
challenges with respect to the concept of Industry 4.0 before this new reality is fully integrated into the
industrial sector. One such challenge is ensuring security, an essential issue in terms of both management
and operational technology involving the configuration and administration of physical equipment. Another
is integrating equipment from different manufacturers which must be configured and/or administered with
incompatible protocols. Likewise, remote access to systems must be carried out from a browser in the
web frame. This paper presents a set of tools, available as a cloud-based service, which facilitates access
to the physical equipment located in a business. The access enables the end user to see and/or modify
the configuration and management of data in a secure, controlled, organized and collaborative manner.
In order to unify different protocols, the service makes use of the Modbus TCP/IP (Transmission Control
Protocol/Internet Protocol) protocol, one of the most commonly used by industry in local networks. The
local Modbus communications are encapsulated in the Websocket protocol to make them accessible from
the cloud. The tools proposed in this paper are based on open hardware platforms and free software, and they
permit local and remote sensors and actuators to be fully and easily accessed. Hence, their chief advantage
is that they allow access from any internet connection, via a browser, to the physical equipment located in
an industry by means of user profiles in accordance with the security level set by each individual company.
INDEX TERMS Modbus, remote access, open source hardware/software, EJS/EJSS, data acquisition,
Internet-of-Things, Industrial Internet of Things, cloud, Industry 4.0.
I. INTRODUCTION
Companies are now developing digital strategies which
enable them to incorporate all the advantages and possibil-
ities of the digital world into their production lines. The
information generated by these advances will be digitally
accessible from computers and other networked devices,
the analysis and evaluation of which opens up a wide range
of possibilities. This information can be considered a trans-
formable raw material that can be given added value. The
possibilities of digitalization are greatest in the industrial
field, in which advanced digital techniques drive increases in
productivity [1], such as the introduction of augmented reality
in the assembly and maintenance of facilities, and the use of
autonomous robots in production lines. If, in addition, all a
company’s systems are interconnected and accessible, then
The associate editor coordinating the review of this manuscript and
approving it for publication was Theofanis P. Raptis .
the company has achieved full digitalization and connectivity
and can be described as an Industry 4.0 company.
The cornerstones of Industry 4.0 are the advances in dig-
ital and communications technologies. In the new industrial
paradigm everything is interconnected. In addition, advances
facilitated by the internet are tending to replace limited, local
visions of industry with more global and distributed ones.
This new vision enables total connectivity without the need
for resources to be physically adjacent.
The degree of connectivity sought by the Internet of Things
(IoT) or the Internet of Everything (IoE) [2, p.], [3], [4]
already exists in the personal and social spheres. We live
in a society permanently connected to the network, with an
ever growing number of connected devices. The use of smart
phones, for example, is increasingly widespread, and almost
half the population have profiles on social networks. All these
devices and applications generate data which we are largely
unaware of. Nor do we control the applications installed
on our devices for managing these data, as they exact our
VOLUME 8, 2020 This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see http://creativecommons.org/licenses/by/4.0/ 10433

R. Sánchez-Herrera et al.: Easy and Secure Handling of Sensors and Actuators as Cloud-Based Service
mandatory consent which cannot be subsequently retracted.
In addition, our online activity, like the devices, is constantly
registered on an unknown server. In short, we have little
control over our information and our devices. Moreover, they
are vulnerable to those who know to obtain it and have the
will to do so, without mentioning the consequences of discon-
necting the server. In the industrial sphere, where information
management and plant operation coexist, security and privacy
are absolute essentials. In such a context, loss of informa-
tion and control of devices is unthinkable, and consequently
the adoption of IoT within this sector, referred to as the
Industrial IoT (IIoT) [5], [6] represents a major challenge to
industry.
Nevertheless, total connectivity has advantages for the
industry that cannot be ignored. Among these is a much
more open and immediate communications model [7] for
all types of information exchange, whether between devices
(M2M, Machine to Machine) [8], or between devices and
people (M2P, Machine to People). The integration of manage-
ment and production levels within industry means facilitating
the exchange of information between the two. According to
the current tendency to integrate the management information
into the content management systems (CMS), cloud technolo-
gies have a very important role to play [9]. These technologies
can be managed by the company itself or provided by third
partly entities (IaaS, Infrastructure as a Service) that guaran-
tee information security and privacy.
The adoption of Industry 4.0 by an industrial company
involves the digitalisation and integration of all its commu-
nication processes. Moreover, to use CMS, that integration
must be carried out into the cloud. Digitalisation makes it
necessary to use very different communications technologies
to those currently used in the industry. The limitations of local
operation through local area network (LAN) can be overcome
with SCADA (Supervisory Control and Data Acquisition)
systems, which use different protocols, such as Fieldbus,
Profinet [10] to carry out the communication between equip-
ment. One of the most commonly used in heterogeneous
scenarios (involving devices from different manufacturers) is
Modbus TCP/IP [11], which has the following advantages in
comparison with similar protocols:
•It is an open protocol, used locally (without internet
access) across a wide range of industrial scenarios.
•It is interoperable with a large number of indepen-
dent devices supporting the protocol, for example, fre-
quency inverters, PLCs (programable logical control),
I/O devices, etc.
•It can act as a link between devices and industrial net-
works running different protocols specific to different
manufacturers (proprietary software).
•It is widely used on open hardware platforms, such as
SBCs (Single Board Computers) and MCUs (MicroCon-
troller Units).
These design features make Modbus ideal for situations
involving a multiplicity of devices and software, linked
together in LANs over the TCP/IP protocol.
It is possible for companies to access their data and devices
remotely through the use of cloud-based services via a stan-
dard internet connection and a browser. However, several
technical challenges need to be resolved. This paper presents
solutions to the connectivity issues associated with the LAN
and – a more complex challenge – the internet. In section 3.1,
it also deals with connectivity via browser, the use of which
introduces an additional layer of complications.
Against this background, this paper presents a set of tools
for accessing all the devices on a company’s industrial net-
work by means of Modbus. The devices are accessed via the
cloud and allow data queries to be carried out and devices
to be configured and/or administrated in a secure, controlled,
organized and collaborative environment. The specific nature
of these design features is as follows: security is ensured by
data encryption and the use of user profiles for access; control
refers to the ability to apply different filters and user criteria,
such as the time zone; the tools are organized in the sense
that they allow access to be sequenced through, for example,
a reservation system; and finally, they are collaborative in that
they allow concurrent access to the same resource. In short,
the tools permit access to an industrial network from any
internet connection by means of a browser with a user profile
enabled and the required level of security specified by each
company, and as such they fulfil all system requirements fac-
ing companies for integration into Industry 4.0 and into CMS.
The outline of the remainder of the paper is as fol-
lows. Section 2 provides a review of the technical literature.
Section 3 gives a detailed description of the set of tools
proposed; it is divided into two subsections, the first dealing
with the restrictions imposed by the browsers, as mentioned
above, and the second, the technical aspects. Section 4 cov-
ers the aspects relating to integration in the cloud, while
Section 5 presents a case study. It is divided in two subsec-
tions, the first to describe the case and the second to present
the obtained results. Finally, in Section 6, some conclusions
are drawn.
II. STATE OF THE ART
The communications scenario described in the introduction
must integrate the traditional communications model cur-
rently used in industrial frameworks, denominated the com-
puter integrated manufacturing (CIM) pyramid, with the new
technologies emerging from cloud computing. The highest
levels in CIM architecture correspond to information tech-
nology (IT), and are traditionally related to the commer-
cial management, whilst the lowest levels correspond to
the operational technology (OT) and are directly related to
the physical devices involved in the manufacturing process.
In the majority of companies, CIM remains embedded in local
communications.
In addition, in many cases, unlike the IT applications,
the OT devices are not interconnected via a LAN, and
instead use proprietary software oriented towards closed
systems. Integration of the IT and OT layers would allow
the communication between them, making production more
10434 VOLUME 8, 2020

R. Sánchez-Herrera et al.: Easy and Secure Handling of Sensors and Actuators as Cloud-Based Service
efficient and improving profits [12]. IT/OT convergence
enables direct control of the complete monitoring system
in a company through automation and integration of com-
munications systems and industrial networks. Without this
convergence, a certain degree of operative inefficiency is
inevitable and the work flow cannot be optimized.
However, this convergence is not easy to achieve. Commer-
cial platforms exist, but they are rarely compatible with other
applications within the same company because each manu-
facturer provides its own products and solutions, [13]–[17].
One proposal that is becoming increasingly popular is that
of open solutions [18]. In this regard, Botta et al. [19] pro-
mote research efforts to define standard protocols, libraries,
languages, and methodologies to develop the full potential
of IoT. An alternative to commercial packages can be found
in open hardware and free software platforms, which not
only offer more control over the system and data, but also
allow developers to work at lower levels, and, more impor-
tantly, do not require configuration data to be sent to external
databases, [20].
As a result, low-cost platforms, such as Arduino [21],
Phidget [22] or Raspberry Pi [23], are becoming increas-
ingly popular. They display characteristics analogous to more
expensive proprietary devices, and are highly suitable for
monitoring and controlling systems via the internet, such as
real-time monitoring of fuel cells using LabVIEW [24], and
Arduino [25], [26], accessing applications in solar energy
facilities [27], [28], developing the IoT [29], [30], remotely
operating robotic applications [31], [32] and transmitting
data [33], [34, p.], among others.
There are two procedural options for achieving the kind of
convergence illustrated by the examples above of devices and
industrial networks across different platforms:
1. The substitution of the protocols used up to now by
others compatible with access from the cloud, such as
HTTP (HyperText Transfer Protocol), Websocket and
SSE (Server Sent Event). A major disadvantage of this
option is that it discards the protocols currently used by
the majority of devices in operation in manufacturing
and other fields.
2. The design of an appropriate runways to make it possi-
ble to access the protocols currently used in the industry
from the cloud.
This paper proposes a solution of the second type, using
Modbus TCP/IP as the ideal protocol to unify heterogeneous
industrial networks. However, the availability of the unified
platform is not the only requirement for the development
of remote monitoring and control systems. The integration
of the relevant hardware/software into a communications
network is also required. Remote systems can be classified
as open or closed according to their degree of heterogeneity
and accessibility, [35]. Hence, if the devices involved in the
remote system are from a single manufacturer or compati-
ble software, and are all connected to the same LAN [36],
the remote system will be closed or proprietary. It becomes a
little more open if – assuming the devices remain mutually
compatible - it is accessible via the Internet [37], [38].
Finally, if the system involves mutually incompatible devices
or devices from different manufacturers, and is also accessi-
ble via the internet, it can be considered fully open.
Some fully open systems can be found in the literature,
but not many. Furthermore, those that are discussed tend to
be local ad hoc solutions [39]–[41], with the result that the
technology is non-transferable from one system to another.
For example, in [42] a new flexible framework based on social
instant messaging (IM) application architecture is proposed
for integrating an offline remote experiment into a commu-
nication node via a unique identifier. However, the system is
only valid for plants controlled with LabView.
In the approach outlined in this paper, the problem of
heterogeneity is solved by the use of the unification protocol
TCP/IP, while accessibility is achieved through integrating
the systems involved with Modbus, as addressed in a paper
by the same authors [43] in which a set of tools were pre-
sented for accessing a set of devices unified by Modbus
over the internet. This solution is easy to implement over
open platforms. In [43], easy java simulations (EJS)[44] is
proposed as a suitable medium for implementing the applica-
tion which accesses the physical devices (henceforth the user
interface, UI). EJS is an open-source authoring tool created
by Francisco Esquembre for building discrete simulations.
An important feature of EJS is that it enables the develop-
ment of the UI. However, it does not incorporate the tools
necessary for connecting hardware. To overcome this prob-
lem, the authors [43] present a set of software tools, called
elements, which are integrated within EJS, and provide a
simple way of connecting physical devices across the network
(through the use of controllers). Finally, they materialize the
unification of communications through the corresponding
EJS elements (software elements), which hide the details of
the communications technology so as to make it easier to
develop applications when the designer has little knowledge
of network-based applications.
In short, the procedure proposed in [43] makes it possible
to access accessible devices unified by Modbus TCP/IP over
the internet. Establishing a connection between any internet-
enabled PC (or SBC) and the physical system in ques-
tion requires an application programming interface (API).
However, APIs are not supported by the new generation of
browsers, so in order to make the proposed software resident
in the cloud, further software is needed. This new software
has been developed by the authors and is the subject of this
paper.
III. DESIGN OF THE USER INTERFACE IN THE
FRAMEWORK OF THE NEW GENERATION
OF BROWSERS
Cloud technology is becoming the most used to integrated
IT in the internet. One of the causes is the fact that a cloud-
based service does not require any specific application to be
installed on the end user PC in order to carry out its function.
However, access to the service is necessarily carried out from
VOLUME 8, 2020 10435

R. Sánchez-Herrera et al.: Easy and Secure Handling of Sensors and Actuators as Cloud-Based Service
a browser. This means that, for the integration of IT and
OT layers of the CIM, they both must operate within the
execution framework of the browser, a consequence which
imposes limitations on the interaction with the local operating
system and the possibilities of remote communications. Such
limitations include, for example, the capacity of the local
operating system to store data, execute an application or ini-
tiate a socket accessible from the operating system.
With this in mind, subsection 3.1 details limitations
imposed by the new generation of browsers, while
subsection 3.2 presents the set of tools which is the main
focus of the paper: a cloud-based, browser-enabled system for
accessing heterogeneous industrial networks unified under
Modbus TCP/IP.
A. CLOUD ACCESS BY TH E USER
For security reasons, current browsers do not allow the execu-
tion of local applications, or connections between themselves
and the operating system of the machine on which they are
running. This places a limitation on remote access, so it is
necessary to use HTTP or supported protocols over HTTP
(such as Websocket or SSE) to access networked devices
(both internet and local network). However, communication
between the different devices, systems and plants in a com-
pany have to date generally been carried without taking into
account the question of remote accessibility regarding OT
(as this has only recently become viable). Consequently,
the applications used in the design of such elements do not
support integration with web technology and are often not
easily adapted.
Until recently, the most common means of integrating web
technology with functions such as the execution of appli-
cations on the operating system, access to files and remote
access was through Java in the form of an API. Never-
theless, as mentioned above, this kind of integration is not
longer allowed by the new browsers, as they do not allow
the running of Java applets to reproduce the client-server
architecture.
One means available to application programmers for
retaining the capacity to connect the browser with the indus-
trial network is a JavaScript engine which allows them to
create connections based on HTTP. This is the procedure
adopted in this paper for the integration of the application
with the browser, in cases where the controller has been
developed within open platforms.
The Modbus master and slave elements presented in [42]
unify heterogeneous industrial networks and make them
accessible from the internet by means of applets generated
in EJS. However, those elements do not enable the system to
be stored in, and accessed from, the cloud. The following sub-
section presents various easy Java and JavaScript simulations
(EJSS) [44] elements designed to overcome this limitation,
encapsulating the Modbus protocol on top of HTTP-based
protocols.
B. PROPOSED EJJS ELEMENT S
EJSS is a new version of EJS to implement JavaScript
interfaces. Four elements within the EJSS framework have
been created to encapsulate the Modbus protocol on top
of those based on HTTP. They are jointly denominated
OverWebsocket (OW), and complement the elements pre-
sented in [43] and denominated Modbus (MB). Their oper-
ation is as follows: there are two sets of OW elements,
the MasterOverWebsocket (MOW) and the SlaveOverWeb-
socket (SOW). MOW is composed of ClientMasterOverWeb-
socket (CMOW) and ServerMasterOverWebsocket (SMOW).
They open communication links in HTTP protocols
between the industrial network control application (INCA),
developed in EJS, and the UI, developed in EJSS and
executed on the end user PC. There are two SOW ele-
ments, the ClientSlaveOverWebsocket (CSOW) and Server-
SlaveOverWebsocket (SSOW), whose function is the same as
that of the MOW’s. The use of the MOW or SOW elements
is determined by the configuration of the MB elements in the
industrial network.
Hence, on the one hand, if the element included in the
INCA is the MasterModbus (MMB), the MOW elements
will be used to encapsulate the Modbus communications on
top of HTTP. Conversely, if the element in the INCA is the
SlaveModbus (SMB), SOW will be used. On the other hand,
the corresponding server element (SMOW or SSOW) must
be located in the INCA, which has a valid and fixed IP. And
the corresponding client element (CMOW or CSOW) must
be located in the UI, which runs on the end user PC with
any IP.
Among the HTTP protocols available, the proposed
OW elements use Websocket to establish the communications
links. In addition, the information is encapsulated in JSON
(JavaScript Object Notation) as this is the most suitable for
use in JavaScript. Like the MB elements, the OW also gen-
erate an API. However, the new one does allow write / read
calls to be made from the browser to the Modbus registers that
configure and manage the physical devices of the industrial
network.
In the same way as the MB elements, OW only requires
inclusion in the UI and the INCA. This can be done by select-
ing them with the mouse from the window ‘‘Elements for the
model’’ and dropping them into the‘‘List of elements’’ shown
in figures 1 and 2. Figure 1 presents that operation in EJSS.
As indicated above, CMOW and CSOW are available (both
of which can be included in the UI). In the case shown in fig-
ure 1 it is CMOW which has been included, which also cor-
responds to the case presented in figure 2. Figure 2 presents
the EJS framework and the INCA. Here, the MB element
is necessarily MMB, and consequently SMOW must also
be included. SMOW requires the following methods for its
configuration:
SMOW.setModbus(MMB)
SMOW.port(8000)
10436 VOLUME 8, 2020

R. Sánchez-Herrera et al.: Easy and Secure Handling of Sensors and Actuators as Cloud-Based Service
FIGURE 1. The EJSS application with available client elements. These
appear on the right. On the left, CMOW has been selected and is included
in the UI.
FIGURE 2. The EJS application with MB and OW elements. In this case,
MMB has been required in the INCA and hence also SMOW. This case is
the corresponds to the UI illustrated in figure 1.
The first of these connects the MB and OW elements. The
second indicates the communications port between SMOW
and CMOW in the UI, which in the case is the 8000. Hence the
following method must be included in figure 1: masterOver-
Websocket.port(8000).
In all the methods described above, the word occurring just
before the period is precisely the name of the corresponding
element, as it appears on the EJS / EJSS element screen,
differentiating between upper and lower case letters. Thus,
in the figure 2 methods, SMOW occurs just before the period,
as does masterOverWebsocket in figure 1. Any name of the
element can be used and is chosen by the programmer when
are dropped in the ‘‘List of elements’’ panel.
If the required element in the INCA is SMB, then the
OW for inclusion would be SSOW in the INCA and CSOW
FIGURE 3. FIGURE 3. EJSS application with the OW elements available on
the right. On the left, the SARLAB element has already been included in
the UI.
in the UI. The methods for their configuration would be the
same.
IV. INTEGRATION AS A CLOUD SERVICE
The high degree of data availability and expandability in
the cloud make this service an excellent way to increase
the efficiency of IoT systems. Data storage and analysis can
be done in the cloud instead of on IoT devices, and hence
this is where all the services described in this paper are
located, making data and resources permanently available for
all devices connected to the LAN.
To this end, the final step of the procedure proposed in
this paper is the publication of the access to the controllers
in the cloud in an organized, controlled and transparent way,
irrespective of the communication protocol used.
In order to achieve this, all the physical devices in the
industrial network are controlled by means of the Mod-
bus protocol, whether the platform is open or commercial.
In addition, as indicated above, the HTTP protocol used is
Websocket.
The element designed to implement the communication
links is SARLAB (an acronym derived from its Spanish
name, Sistema de Acceso Remoto a LABoratorios, or remote
laboratory access system in English), as illustrated in figure 3,
and described in [43]. Regarding communication architec-
ture, SARLAB uses a middleware layer to make the commu-
nication process between the user computer and the controller
connected to the industrial network [45]–[47] transparent.
Middleware works as a software distributed abstraction layer,
and is located between the application and the lower layers
(operating system and network layer). The software devel-
oped in the middleware layer provides an API which hides
the complexity of the general communication problem.
In this way, SARLAB creates a service in the cloud
allowing direct access from the UI, which it registers in
the cloud IaaS. The communication layers in the industrial
network are presented in figure 4. The upper is the cloud
VOLUME 8, 2020 10437

R. Sánchez-Herrera et al.: Easy and Secure Handling of Sensors and Actuators as Cloud-Based Service
FIGURE 4. Communications arquitecture.
FIGURE 5. Wheastone brigde.
layer, and going down, network layer with the routers, fog
layer where controllers are connected and finally device layer
with physical devices. The links between the end user com-
puter and the fog layer are identified by the corresponding
URL (uniform resource locator).
SARLAB controls access from outside the cloud according
to the corresponding user profiles. However, SARLAB does
not limit direct access to the internet from the controllers in
the fog layer. Consequently, events occurring in the industrial
network of a company can make use of the services in the
cloud without being controlled by SARLAB. Figure 4 also
represents the communication links created by SARLAB,
based on [48]. As can be seen, SARLAB controls the commu-
nications and access to the fog layer from the cloud so that the
FIGURE 6. A general view of the use case plant.
end user can remotely access the industrial network by means
of the cloud service.
V. USE CASE
A. DESCRIPTION
In this section, a use case is presented, consisting of a Wheat-
stone bridge, the corresponding circuit for which can be
seen in figure 5. R2 in figure 5 represents a set of resistors,
each of which is of unknown value and needs to be ascer-
tained. The system also provides the possibility of connecting
two or three resistors in series or parallel. Figure 6 represents
a general view of the plant. The set of resistors indicated
above is labelled 1.a. The element labelled 1.b is a set of
wires, each of which is of unknown value and also needs to
be found out. Figure 7 represents the UI, where the user can
choose a set of resistors, after choosing between resistors and
wires.
10438 VOLUME 8, 2020

R. Sánchez-Herrera et al.: Easy and Secure Handling of Sensors and Actuators as Cloud-Based Service
FIGURE 7. The use case user interface.
Elements R1 and R2 in figure 5 correspond to the rule at
the bottom of figure 6 (labelled 2). A wire with a resistor
of known resistance is divided into two parts by the isolated
movable device (labelled 3 and henceforth denominated car-
riage). Finally, Rx in figure 5 is a resistor of known resistance
with a value of 100 if the user choses the resistors, and
1if the wires. This element is labelled 4 in figure 6.
Although this physical system is relatively simple, its char-
acteristics allow different technical details regarding imple-
mentation to be illustrated. The intention is to show the
potential in the functioning of the procedure rather than the
complexity of the physical system itself. In fact, the procedure
presented in this paper allows the inclusion of as many sen-
sors and actuators as necessary, regardless of the complexity
of the physical system.
In addition to the image shown in figure 6, the UI shows an
amplified view of the isolated movable element on the right
in figure 7, allowing the exact position on the rule to be clearly
seen.
Regarding the architecture of the use case, the core of
the system is a Raspberry Pi board [23] running the EJS
where the INCA has been developed. On the one hand, INCA
makes the communications with the UI, and on the other
hand, it controls the set of relays and the carriage movement
according to the orders received from the UI.
Figure 8 diagrams the structure of the complete system.
The connections between the end user PC and the cloud can
be seen, and likewise those between the cloud and the con-
troller in the fog layer, which in this case is only the Raspberry
Pi board. Likewise, the Raspberry Pi board is networked to
the Arduino Mega with Shield Ethernet, which controls the
carriage movement. These communications are carried out by
Modbus TCP/IP, with the MB and OW elements in the INCA,
as shown in figure 9.
In fact, the INCA involves an MMB element which is in
communication with the Arduino board via the LAN. A Mod-
bus TCP/IP slave, available in the Arduino libraries, runs on
the Arduino board, which controls the carriage. The com-
munications between those two elements are encapsulated
in the Websocket protocol by means of a SMOW element,
which corresponds to the CMOW element included in the
UI (figure 10). In addition, the INCA controls the relays by
GPIO (general purpose input/output) according to the end
user selection in the UI. In order to do this, an SMB element
has been included in the INCA, which receives the UI orders
by means of the SSOW element. Both can be seen in figure 9.
In figure 10, the CSOW element can be seen, which is in
communication with the earlier SSOW. Finally, the Rasberry
Pi also switches the physical devices on when an authorized
FIGURE 8. General scheme of the use case with the communications between the user and the physical system.
VOLUME 8, 2020 10439

R. Sánchez-Herrera et al.: Easy and Secure Handling of Sensors and Actuators as Cloud-Based Service
FIGURE 9. Elements in the use case INCA.
FIGURE 10. Elements in the use case UI.
user accesses them and switches them off when he/she leaves.
SARLAB verifies the permissions of users at point of access
to the UI.
Therefore, this use case represents a physical system con-
nected to an industrial network, whose INCA has been imple-
mented in EJS using the MB and OW elements, and the UI
in EJSS using OW. The physical system consists of different
devices, including relays controlled by digital signals and
DC-DC motors controlled by PWM to establish the position
of a mobile physical element. The nature of the physical
system could have been any other whose control could be
implemented digitally and analogically.
B. ACCESS AND RESULTS
The use case described in the previous subsection is in oper-
ation and use at the University of Huelva. Access can be
made from anywhere with an Internet connection worldwide
through a CMS, which in this case is moodle (one of the
most widespread learning management systems in the world).
The CMS, whose URL is http://sarlab3.uhu.es/sarlab/,
includes a reservation system to regulate the access; although
it can be also concurrent between remote users or remote and
local users.
In addition, any authorized user (identified in the CMS) can
access and manage the plant through any electronic device
(PC, tablet or smartphone).
The result of applying the proposed procedure and set
of tools to the described physical system in the previous
subsection makes its access secure, controlled, organized
and collaborative. Secure because it is done in an encrypted
way and with the corresponding user profiles. Controlled
because it may be subject to filters with different operating
criteria, such as time zone. Organized because access can be
sequenced by, for example, a reservation manager. Finally,
collaborative because concurrent access to the physical plant
is allowed.
In addition, the physical plant is available through the
internet 24 hours a day, 7 days a week and the UI can combine
real content (usually observed through some electronic device
such as cameras and HMD displays) and virtual computer-
generated content (augmented reality), adequately superim-
posed to improve the usability.
All these characteristics obtained in the use case can be
extrapolated to other kind of systems as for example an indus-
trial plant or smartdevices installed in a building. Although
the use case is a plant in the university field; actually,
the scalability of the developed solution is total, there is no
technical barrier to incorporate more and more sensors and
actuators. In fact, the main problem in industrial fields can be
the security of communications and this is solved with this
proposal.
VI. CONCLUSION
This paper presents a set of tools to facilitate the adoption
of Industry 4.0 across all sectors of the business world.
Although many industries have already migrated their IT
processes to the cloud, the OT component remains acces-
sible only at local level. The tools presented in this paper,
based on open hardware and free software platforms, and
implemented in EJS/EJSS, enable IT and OT to be inte-
grated and accessed remotely. The integration of systems and
devices running different protocols is achieved through Mod-
bus TCP/IP within a single of level of the LAN. At the same
time, access to devices is made available as a cloud-based
service (via a browser) by means of communications encap-
sulated in Websocket. The result is a secure, controlled, orga-
nized and collaborative access to physical systems, which
are in this way available 24 hours a day, 7 days a week
through the internet by means of a UI that can combine real
content and virtual computer-generated content (augmented
reality), adequately superimposed. The solution and results
addressed in this research are extrapolated to other sectors
as the industry, service or household. Really, the application
scope is not a problem, because the solution is completely
general.
10440 VOLUME 8, 2020

R. Sánchez-Herrera et al.: Easy and Secure Handling of Sensors and Actuators as Cloud-Based Service
REFERENCES
[1] M. Savastano, C. Amendola, F. Bellini, and F. D’Ascenzo, ‘‘Contextual
impacts on industrial processes brought by the digital transformation of
manufacturing: A systematic review,’’ Sustainability, vol. 11, no. 3, p. 891,
Feb. 2019.
[2] L. Jiang, L. Da Xu, H. Cai, Z. Jiang, F. Bu, and B. Xu, ‘‘An IoT-oriented
data storage framework in cloud computing platform,’’ IEEE Trans. Ind.
Informat., vol. 10, no. 2, pp. 1443–1451, May 2014.
[3] G. Roussos, S. S. Duri, and C. W. Thompson, ‘‘RFID meets the Internet,’’
IEEE Internet Comput., vol. 13, no. 1, pp. 11–13, Jan. 2009.
[4] C. Floerkemeier, C. Roduner, and M. Lampe, ‘‘RFID application develop-
ment with the accada middleware platform,’’ IEEE Syst. J., vol. 1, no. 2,
pp. 82–94, Dec. 2007.
[5] S. Kwon, J. Jeong, and T. Shon, ‘‘Toward security enhanced provisioning
in industrial IoT systems,’’ Sensors, vol. 18, no. 12, p. 4372, Dec. 2018.
[6] J. Park, ‘‘Advances in future Internet and the industrial Internet of Things,’’
Symmetry, vol. 11, no. 2, p. 244, Feb. 2019.
[7] L. Yang, S. Yang, and L. Plotnick, ‘‘How the Internet of Things technology
enhances emergency response operations,’’ Technol. Forecasting Social
Change, vol. 80, no. 9, pp. 1854–1867, Nov. 2013.
[8] B. W. Khoueiry and M. R. Soleymani, ‘‘A novel machine-to-machine
communication strategy using rateless coding for the Internet of Things,’’
IEEE Internet Things J., vol. 3, no. 6, pp. 937–950, Dec. 2016.
[9] T. Usländer and T. Batz, ‘‘Agile service engineering in the industrial
Internet of Things,’’ Future Internet, vol. 10, no. 10, p. 100, Oct. 2018.
[10] S. Jaloudi, ‘‘Communication protocols of an industrial Internet of Things
environment: A comparative study,’’ Future Internet, vol. 11, no. 3, p. 66,
Mar. 2019.
[11] The Modbus Organization. Accessed: Apr. 17, 2019. [Online]. Available:
http://www.modbus.org/
[12] K. L. S. Sharma, ‘‘20—Information technology-operation technol-
ogy convergence,’’ in Overview of Industrial Process Automation,
K. L. S. Sharma, Ed., 2nd ed. Amsterdam, The Netherlands: Elsevier,
2017, pp. 359–375.
[13] L. De La Torre, M. Guinaldo, R. Heradio, and S. Dormido, ‘‘The ball
and beam system: A case study of virtual and remote Lab enhancement
with Moodle,’’ IEEE Trans. Ind. Informat., vol. 11, no. 4, pp. 934–945,
Aug. 2015.
[14] J. Chacón, G. Farias, H. Vargas, A. Visioli, and S. Dormido, ‘‘Remote
interoperability protocol: A bridge between interactive interfaces and engi-
neering systems,’’ IFAC-PapersOnLine, vol. 48, no. 29, pp. 247–252, 2015.
[15] M. A. Prada, J. J. Fuertes, S. Alonso, S. García, and M. Domínguez,
‘‘Challenges and solutions in remote laboratories. Application to a remote
laboratory of an electro-pneumatic classification cell,’’ Comput. Educ.,
vol. 85, pp. 180–190, Jul. 2015.
[16] U. Hernandez-Jayo and J. Garcia-Zubia, ‘‘Remote measurement and
instrumentation laboratory for training in real analog electronic experi-
ments,’’ Measurement, vol. 82, pp. 123–134, Mar. 2016.
[17] I. González, A. Calderón, A. Mejías, and J. Andújar, ‘‘Novel networked
remote laboratory architecture for open connectivity based on PLC-OPC-
LabVIEW-EJS integration. Application in remote fuzzy control and sen-
sors data acquisition,’’ Sensors, vol. 16, no. 11, p. 1822, Oct. 2016.
[18] S. Weyer, M. Schmitt, M. Ohmer, and D. Gorecky, ‘‘Towards indus-
try 4.0—Standardization as the crucial challenge for highly modular,
multi-vendor production systems,’’ IFAC-PapersOnLine, vol. 48, no. 3,
pp. 579–584, 2015.
[19] A. Botta, W. De Donato, V. Persico, and A. Pescapé, ‘‘Integration of cloud
computing and Internet of Things: A survey,’’ Future Gener. Comput. Syst.,
vol. 56, pp. 684–700, Mar. 2016.
[20] B. P. Wong and B. Kerkez, ‘‘Real-time environmental sensor data:
An application to water quality using Web services,’’ Environ. Model.
Softw., vol. 84, pp. 505–517, Oct. 2016.
[21] Arduino—Home. Accessed: Feb. 4, 2019. [Online]. Available: https://
www.arduino.cc/
[22] Phidgets Inc. Products for USB Sensing and Control.
Accessed: Apr. 17, 2019. [Online]. Available: https://www.phidgets.com/
[23] Raspberry Pi Foundation. Raspberry Pi—Teach, Learn, and Make
with Raspberry Pi. Accessed: Feb. 4, 2019. [Online]. Available:
https://www.raspberrypi.org
[24] LabVIEW 2019—National Instruments. Accessed: May 20, 2019.
[Online]. Available: http://www.ni.com/es-es/shop/labview/labview-
details.html
[25] A. Calderón, I. González, M. Calderón, F. Segura, and J. Andújar, ‘‘A new,
scalable and low cost multi-channel monitoring system for polymer elec-
trolyte fuel cells,’’ Sensors, vol. 16, no. 3, p. 349, Mar. 2016.
[26] T. Ewing, P. T. Ha, J. T. Babauta, N. T. Tang, D. Heo, and H. Beyenal,
‘‘Scale-up of sediment microbial fuel cells,’’ J. Power Sources, vol. 272,
pp. 311–319, Dec. 2014.
[27] H. Gad and H. E. Gad, ‘‘Development of a new temperature data acqui-
sition system for solar energy applications,’’ Renew. Energy, vol. 74,
pp. 337–343, Feb. 2015.
[28] F. Salamone, L. Belussi, L. Danza, M. Ghellere, and I. Meroni, ‘‘An open
source low-cost wireless control system for a forced circulation solar
plant,’’ Sensors, vol. 15, no. 11, pp. 27990–28004, Nov. 2015.
[29] G. Barbon, M. Margolis, F. Palumbo, F. Raimondi, and N. Weldin, ‘‘Tak-
ing Arduino to the Internet of Things: The ASIP programming model,’’
Comput. Commun., vols. 89–90, pp. 128–140, Sep. 2016.
[30] A. Solano, R. Dormido, N. Duro, and J. Sánchez, ‘‘A self-provisioning
mechanism in openstack for IoT devices,’’ Sensors, vol. 16, no. 8, p. 1306,
Aug. 2016.
[31] A. Cela, J. Yebes, R. Arroyo, L. Bergasa, R. Barea, and E. López,
‘‘Complete low-cost implementation of a teleoperated control system for
a humanoid robot,’’ Sensors, vol. 13, no. 2, pp. 1385–1401, Jan. 2013.
[32] C.-T. Chao, M.-H. Chung, J.-S. Chiou, and C.-J. Wang,‘‘A simple interface
for 3D position estimation of a mobile robot with single camera,’’ Sensors,
vol. 16, no. 4, p. 435, Mar. 2016.
[33] D. Piromalis and K. Arvanitis, ‘‘SensoTube: A scalable hardware design
architecture for wireless sensors and actuators networks nodes in the
agricultural domain,’’ Sensors, vol. 16, no. 8, p. 1227, Aug. 2016.
[34] M. R. Senouci, A. Mellouk, N. Aitsaadi, and L. Oukhellou, ‘‘Fusion-based
surveillance WSN deployment using Dempster–Shafer theory,’’ J. Netw.
Comput. Appl., vol. 64, pp. 154–166, Apr. 2016.
[35] R. Sanchez-Herrera, A. Mejias, M. A. Marquez, and J. M. Andujar,
‘‘A fully integrated open solution for the remote operation of pilot plants,’’
IEEE Trans. Ind. Informat., vol. 15, no. 7, pp. 3943–3951, Jul. 2019.
[36] A. M. Borrero and J. M. A. Márquez, ‘‘A pilot study of the effectiveness
of augmented reality to enhance the use of remote labs in electrical engi-
neering education,’’ J. Sci. Educ. Technol., vol. 21, no. 5, pp. 540–557,
Oct. 2012.
[37] V. L. Lasky, D. K. Liu, S. J. Murray, and Y. K. L. Choy, ‘‘A remote PLC
system for e-Learning,’’ in Proc. 4th ASEE/AaeE Global Colloq. Eng.
Educ., 2005, p. 1655.
[38] M. Domínguez, J. J. Fuertes, M. A. Prada, S. Alonso, and A. Morán,
‘‘Remote laboratory of a quadruple tank process for learning in control
engineering using different industrial controllers,’’ Comput. Appl. Eng.
Educ., vol. 22, no. 3, pp. 375–386, Sep. 2014.
[39] M. Casini, D. Prattichizzo, and A. Vicino, ‘‘The automatic control tele-
lab: A user-friendly interface for distance learning,’’ IEEE Trans. Educ.,
vol. 46, no. 2, pp. 252–257, May 2003.
[40] N. Faltin, A. Böhne, J. Tuttas, and B. Wagner, ‘‘Distributed team learning
in an Internet-assisted laboratory,’’ in Proc. Int. Conf. Eng. Educ., 2002,
pp. 18–22.
[41] T. A. Fjeldly, J. O. Strandman, and R. Berntzen, ‘‘LAB-on-WEB—A com-
prehensive electronic device laboratory on a chip accessible via Internet,’’
in Proc. Int. Conf. Eng. Educ., 2002, pp. 1–5.
[42] N. Wang, G. Song, and X. Chen, ‘‘Framework for rapid integration of
offline experiments into remote laboratory,’’ Int. J. Online Biomed. Eng.,
vol. 13, no. 12, p. 192, Dec. 2017.
[43] A. Mejías, R. Herrera, M. Márquez, A. Calderón, I. González, and
J. Andújar, ‘‘Easy handling of sensors and actuators over TCP/IP net-
works by open source hardware/software,’’ Sensors, vol. 17, no. 1, p. 94,
Jan. 2017.
[44] Welcome to Easy Java Simulations Home Page. Accessed: Apr. 17, 2019
[Online]. Available: http://fem.um.es/Ejs/
[45] M. A. Razzaque, M. Milojevic-Jevric, A. Palade, and S. Clarke, ‘‘Middle-
ware for Internet of Things: A survey,’’ IEEE Internet Things J., vol. 3,
no. 1, pp. 70–95, Feb. 2016.
[46] D. F. Sadok, L. L. Gomes, M. Eisenhauer, and J. Kelner, ‘‘A middleware
for industry,’’ Comput. Ind., vol. 71, pp. 58–76, Aug. 2015.
[47] N. Cai, M. Gholami, L. Yang, and R. W. Brennan, ‘‘Application-
oriented intelligent middleware for distributed sensing and control,’’ IEEE
Trans. Syst., Man, Cybern. C, Appl. Rev., vol. 42, no. 6, pp. 947–956,
Nov. 2012.
[48] M. A. Marquez, R. S. Herrera, A. Mejias, F. Esquembre, and J. M. Andujar,
‘‘Controlled and secure access to promote the industrial Internet of
Things,’’ IEEE Access, vol. 6, pp. 48289–48299, 2018.
VOLUME 8, 2020 10441

R. Sánchez-Herrera et al.: Easy and Secure Handling of Sensors and Actuators as Cloud-Based Service
REYES SÁNCHEZ-HERRERA was born in
Huelva, Southwest of Spain. She received the
Industrial Engineering degree and the Ph.D. degree
(Hons.) in electrical engineering, in 1995 and
2007, respectively. She is currently a Professor
with the Department of Electrical Engineering,
University of Huelva. She worked in electrical
engineering within the first research group to
which she belonged. In electronics and communi-
cations in the second. She is the main research of
the applied multidisciplinary research group with the University of Huelva.
Her main interests include electrical power quality, renewable energy sys-
tems, and engineering education.
MARCO A. MÁRQUEZ received the Industrial
Engineering degree from the University of Seville,
Seville, Spain, in 1979, and the master’s degree
in industrial engineering from the University of
Huelva, Huelva, Spain, in 2009. He was an
Associate Professor with the Department of Elec-
tronic Engineering, Computer Systems and Auto-
matic Control, University of Huelva, from 1994 to
2002. He is currently a Researcher with Huelva
University. He is also a regional Instructor of Cisco
Systems with the CIT Business University Foundation, University of Cádiz,
Cádiz, Spain. His research interests include new e-learning technologies and
the communications aspects in remote labs. His developments are the basis
UNILABS communications systems, a network of Spanish universities that
share its laboratories through the Internet.
JOSÉ M. ANDÚJAR (Senior Member, IEEE) was
born in Huelva, Spain. He received the Ph.D.
degree, in 2000. He is currently a Full Professor
of systems engineering and automatic control with
the University of Huelva, Spain. Throughout his
professional life he has received 23 awards and
academic honors. He has conducted ten Doctoral
Theses with eight prizes. He has 12 international
patents. He has more than 300 publications, among
them more than 100 articles published in indexed
journals in the ISI Journal Citation Reports. Specifically, he has 43 quartile
Q1 publications in 19 different journals; most of these journals are among
the top ten in their categories, and several are number one. He has an H-
index (SCOPUS) = 22. He has led or co-led 50 research projects funded by
public institutions and companies. His main research interests are control
engineering, renewable energy systems, and engineering education.
10442 VOLUME 8, 2020