Resilient synchronization of interconnected dynamical networks with DoS attacks: An adaptive event-triggered control approach
Yao Cui, Xiaohua Ge, Pei Cheng, Xin Liu · Neurocomputing · 2025
This paper is concerned with resilience and efficiency of a class of interconnected dynamical networks (IDNs) under limited communication resources and potential cyber-attacks. We first propose a novel framework for synchronization control in IDNs that integrates distributed event-triggered communication with resilience against denial-of-service (DoS) attacks. We start with developing an adaptive event-triggered mechanism (AETM), dependent on the attack parameter, to dynamically adjust threshold parameters for each node based on network conditions and severity of DoS attacks. Then, we design DoS-resilient event-triggered distributed synchronization controllers for each node to achieve the desired synchronization objective, even under sporadic event-based communication and disruptions caused by DoS attacks. Third, we establish co-design criteria to jointly determine the gain matrices of both the AETMs and synchronization controllers. These criteria provide valuable insights into the interplay between communication quality (e.g., triggering frequency and DoS severity) and control quality (e.g., convergence rate). Finally, we validate the effectiveness of the proposed approach through two illustrative examples.