FedResilience: A Federated Learning Application to Improve Resilience of Resource-Constrained Critical Infrastructures

Critical Infrastructures Overview: Past, Present and Future

Industrialized societies depend on the proper functioning of a whole range of technological infrastructures, such as electricity, road and railway networks and telecommunications which, due to their importance, are generically referred to as critical infrastructures (CIs). Technical failures, natural disasters and malicious events, if not terrorist, could have devastating effects on these infrastructures. The events of the last few years have accelerated efforts to identify and designate CIs at national and European levels and have reinforced concerns about increasing their protection in sensitive sectors for the safety of the individual and the community. The aim of this research is to provide the basic elements to understand the issue along with the reasons for its importance both at national, European and international level. In particular, after analyzing the origin of the problem, a systematic literature review is carried out to study the current research around future perspectives relating to the management of Cis, with particular focus on three research questions: RQ1 “What types of risk assessment methods are used to manage CIs?”, RQ2 “What are the environmental risk mitigation strategies for CIs?” and RQ3 “What is the role of the human factor in the prevention of risks for CIs?”. The results aim to be guidelines for decision makers and researchers interested in this topic.

Hierarchical Federated Learning for Edge-Aided Unmanned Aerial Vehicle Networks

Federated learning (FL) allows UAVs to collaboratively train a globally shared machine learning model while locally preserving their private data. Recently, the FL in edge-aided unmanned aerial vehicle (UAV) networks has drawn an upsurge of research interest due to a bursting increase in heterogeneous data acquired by UAVs and the need to build the global model with privacy; however, a critical issue is how to deal with the non-independent and identically distributed (non-i.i.d.) nature of heterogeneous data while ensuring the convergence of learning. To effectively address this challenging issue, this paper proposes a novel and high-performing FL scheme, namely, the hierarchical FL algorithm, for the edge-aided UAV network, which exploits the edge servers located in base stations as intermediate aggregators with employing commonly shared data. Experiment results demonstrate that the proposed hierarchical FL algorithm outperforms several baseline FL algorithms and exhibits better convergence behavior.