ARAUJO, A. S.; http://lattes.cnpq.br/7586196643780905; ARAUJO, Aramis Sales.
Résumé:
5G mobile networks play a fundamental role in connected smart factories, providing signif icant improvements in security and Quality of Service (QoS) factors compared to previous
generations. The ability to transmit large amounts of data with low latency enables remote
control and monitoring in critical scenarios, involving the control of machines, industrial
devices, and the monitoring of smart sensors. Considering the scenario in which 5G mo bile networks are applied to industry, the opportunity to prioritize data transmission flows
and network services using network orchestration tools becomes evident. For this purpose,
Intent-Based Network Systems (IBNs) are tools that combine artificial intelligence tech niques and network orchestration capabilities to comply with high-level defined business
intents, improving communication effectiveness and promoting the adequate functioning of
the systems involved. In this context, this work aims to evaluate the integration between the
data and statistics from Internet of Things (IoT) devices, observed within a connected indus trial environment, and the metrics provided by the 5G network utilized in smart factories, in
the use of neural network models for the task of classifying network problem scenarios like
the increase of communication latency. Therefore, the quantitative evaluation of these mod els’ performance highlights the potential of IoT devices’ data in the task of network problem
scenario classification. Bringing out the potential to automate and reconfigure network sys tems in a fast, secure, and autonomous way, promoting compliance with the rules defined in
the business model of the specific industrial environment.
