Sensors 2020, 20, 5603 4 of 19
properties [
24
], but it does not contain information related to the integration of the fieldbuses and
devices from the industrial environment.
Another important IIoT reference architecture is the Reference Architectural Model Industry
4.0 (RAMI 4.0) [
25
,
26
]. This is based on a 3-D model where the axes are Live Cycle, Value Stream,
and Hierarchy Levels, and it is service-oriented architecture. The hierarchy level of the architecture
is organized in the following levels: product, field devices, control devices, station, work centers,
enterprise, and connected world. The value stream level of the architecture is related to the different
component functionalities, it includes a communication layer and it is organized the following
layers with a high level of abstraction: asset, integration, communication, information, functional,
and business.
Other IoT reference architectures are organized on layers, such that defined by the International
Telecommunication Union (ITU) that consists of four layers: devices, network, service support and
application support, and application [
27
]. The device layer includes the device capabilities for interaction
for the communication network and gateway capabilities for supporting multiple communication
networks. The network layer includes networking and transport capabilities, service support, and the
application support layer includes the generic and specific support for the application layer.
In [
28
], the authors proposed a case study on the growth of big data in IIoT systems, a classification
of key concepts, and a presentation of key frameworks and continued with the presentation of future
technologies, opportunities, and challenges. These researches concluded that augmented reality,
IoT devices, cyber-physical systems and Industry 4.0 Big Data and Analytics (BDA) platforms are
at an early stage in IIoT systems and that solutions should be found in developing new standards
that allow interoperability between various platforms but also processing capability of the end-to-end
applications for concentric computing systems. A solution to improve the blockchain scalability of
IIoT systems by guaranteeing system security, latency but also decentralization, that was proposed in
the article [
29
] have led to performance optimization with a deep reinforcement learning technique
(DRL). The authors obtained results that demonstrated they can obtain a better efficiency compared to
the basic parameters of the system.
In another research paper [
19
], the authors concluded based on conducted researches that the
success of IIoT can be hindered for different reasons such as challenging collaboration between various
heterogeneous IIoT systems, efficient data management, large, solid, and flexible data technologies,
IIoT protocols, operating systems, reliable IIoT systems but also the coexistence of wireless technologies.
In addition to a survey of existing definitions of IIoT in [
30
], the authors also present a proposal
for a new definition of what the IIoT means. The authors also present an analysis of a framework
for IIoT devices based on security-related issues surrounding IIoT but as well as an analysis of the
relationships between cyber-physical systems and Industry 4.0.
A secure fog-based IIoT architecture by suitably plugging a number of security features that
reduce the trust and burden on the cloud and resource-constrained devices, but also that reduce latency
in decision making that improves the performance, is presented in another research paper from the
specialized literature [
31
]. Also, the authors demonstrated that by offloading several computationally
intensive tasks to the fog nodes, the battery life of the resource-constrained of end devices is greatly
saved. The validation of the architecture was demonstrated through theoretical analyses, practical
experiments, but also through simulation and testbed implementation.
In [
32
], an IIoT application for a sewage treatment plant is proposed. This IIoT solution uses control
station systems based on the STMicroelectronics STM32 microcontrollers to update the automation
system and to activate the IIoT concept. Basically, the STM32-based control stations act as gateways
between field devices connected to fieldbuses and their connection within the local network and
further to the Internet for remote monitoring and cloud connections. The monitoring operations can
be performed remotely from PCs or mobile devices such as smartphones and tablets. The authors
concluded that this solution provide the real-time performances and reliability required for monitoring
and controlling the sewage treatment plant.