Edge Intelligent Computing in Aqua Environments

Nowadays, an ambitious target of the next generation networks is to develop edge intelligence ecosystem able to efficiently operate in heterogeneous domains. Toward this end, the underwater environment requires a special attention, since they are recognized as the most challenging domain. Within this context, this chapter aims at illustrating a self-intelligent ground–aqua integrated system where the emerging semantic communication paradigm is envisaged to counteract the hostile behavior of underwater channels. In particular, we discuss in this chapter the use of deep convolution neural networks-based encoder–decoder architecture for this purpose. Numerical results are included here to show the better behavior of the proposed system in comparison with the conventional alternative that does not provide the use of the semantic communications approach. Finally, a specific performance evaluation analysis is devoted to the analysis of the convergence behavior of the proposed Federated Learning (FL) procedure in reference to the cross ground–aqua system considered to highlight its advantages with respect to a classical implementation.