1.1.2 Agent/Firmware Layer: User Interface Layer
The agent layer interconnects the gateway for device communication with all the other components. It is based on the agent's modular architecture suitable for robotic and production machines controlled by sensors. The IoT agent is linked to the IoT controller, and checks the framework environment during the time, by acquiring data by sensors properly programmed by firmware. Firmware is developed through software and hardware interfaces. Integrated development environment (IDE) platforms are usually adopted for firmware development as user interfaces. Agent‐based computing (ABC) tools are able to implement agent functions in modern distributed applications; multi‐agent systems (MASs) are ensembles of agents interacting in the same framework [31]. The MASs support edge and cloud computing, interconnecting innovative applications such as machine learning and blockchain.
1.1.3 Gateway and Enterprise Service Bus Layer
The gateway is an important node executing data routing and switching functions of the information network, using different protocol types. The gateway behaves as a provider, and it is interfaced between the IoT middleware and the agent layer. This layer is responsible for publishing and subscribing the services, message routing, and allows the communication between platforms. The enterprise service bus (ESB) enables service‐oriented architecture (SOA) applications. Typically, the ESB is adopted to collect the digital data sources of different technologies adopted in the same information network, such as big data systems and AI algorithm datasets, by allowing data transfer between different DB systems [32]. ESB solutions are suitable to integrate old technologies with new ones improving the information system and Industry 4.0 implementations. The gateway system covers the role of collector of the packets coming from different sensors applied in the specific production line field. Communication takes place through IEEE 802.11 (WI‐FI/5 GHz) or MODBUS standards, following the most common mechanisms for asynchronous communication including Message Queuing Telemetry Transport (MQTT) protocol [30]. The architecture of the gateway follows Industry 4.0 mechanisms, where the gateway represents the main local component made by:
A sensor manager, able to locally manage the sensor nodes.
A local DB manager with regulated accesses.
An engine optimizer capable of processing advances.
The application programming interfaces (APIs) integrating the IoT sensor network with the cloud environment by the use of mobile devices.
The IoT gateway is designed to implement the following features:
IoT network management (local knowledge of connected devices and resources, local management of sensors or network nodes, local data availability).
System intelligence (ability to perform process optimization locally, integrity of the information received).
Distributed logic (the data from the nodes or sensors are stored in the local DB and managed for the central system).
The ESB is a software infrastructure carrying out services in complex SOAs and supporting horizontal, vertical and end to end integrations. The main advantage of this infrastructure is to interconnect and to interface heterogeneous technologies including big data systems, ensuring data synchronization, data security, messaging, intelligent routing managed by AI algorithms, and transformations services. The ESB assists the developer in integrating applications, therefore providing the infrastructure necessary to implement routing, translation, and other integration of features.
1.1.4 IoT Middleware
The middleware function [21] is to connect different typologies of programs. This interface is part of the architecture enabling sensor connectivity and application layers. The middleware manages important functionalities, such as collecting and selecting the received data from the IoT devices, by providing access monitoring for applications. The security is mainly performed in the middleware system by [33]:
User identification.
Identity management.
A secure data communication system.
Secure storage.
A secure software execution environment.
Secure contents.
Security resistance.
The middleware improvement is related to the following specifications:
Interoperability between devices of different technologies by managing heterogeneous interfaces.
Managing devices performing load balancing.
Use of API calls.
Scalability by supporting the communication between a large number of devices.
Big data interconnection and big data analytics tools.
AI algorithm interconnection activating sensor data processing.
Authentication and implementation of access control improving security and privacy.
Running algorithms on different cloud services.
Data extraction and data migration improving innovative DB and big data systems.
Middleware is classified as event based when all the components interact with each other through events, or service oriented when service providers are used for resource management in SOAs, or DB oriented, or application oriented.
1.1.5 Processing Layer
The processing layer is very important for the orientation of the industry network on facilities to Industry 5.0, including big data systems and AI algorithms, able to activate advanced data processing and the setting of all parameters of the architecture of Figure 1.3. The data processing is suitable for predictive maintenance, for machine failure predictions, for the self‐adaptive production, and in general for assisted production.
1.1.6 Application Layer
This layer is responsible for delivering various application services. These services are provided through the middleware layer. The application services are suitable for, logistics, BI, rapid prototyping, reverse engineering (RE) and other advanced industry applications.
1.1.7 File Transfer Protocols
File transfer of the company information systems is executed through file transfer protocols. The main protocols adopted by the information network systems are listed in Table 1.5.
Table 1.5 Other protocols usable in industry.
| File transfer protocols | Details | References |
|---|---|---|
| File Transfer Protocol | Client/serverTCP connectionProtocol interpreter;Data transfer processData and command separate connections | [34] |
| Simple Mail Transfer Protocol |
Connection orientedText basedClient/server
|