Context-aware monitoring for IoT: an approach based on Agents, and Federated Learning

This paper proposes a software architecture model applied for context-aware monitoring of Smart Buildings(SB), and it works in two layers: Fog and Edge computing and uses the Multi-Agent System(MAS) and Federated Learning(FL). The objective of this architecture will allow context-aware monitoring and the capture of anomalies and offer specific functionalities that depend on the user roles. As a result, we modeled the complete software architecture considering Reactive, edge-fog computing, and MAS components, following the best practices from the literature and the 2413-IEEE standard.