Federated Learning-Driven Edge Intelligence Framework for Maritime Monitoring

In this paper, we propose an edge intelligence framework for maritime monitoring that combines edge computing and federated learning to optimize communication delay and model convergence. The framework reduces data transmission and accelerates the model convergence process by employing compression techniques, upload timing strategies, and weighted aggregation methods. Simulation results show that these optimizations significantly improve system efficiency and provide a scalable solution for real-time decision making in resource-constrained environments.