网络拓扑
同步(交流)
计算机科学
马尔可夫链
马尔可夫过程
事件(粒子物理)
复杂网络
拓扑(电路)
计算机网络
数学
工程类
物理
电气工程
频道(广播)
统计
量子力学
机器学习
万维网
作者
Haoyu Zhou,Jian Liu,Yongbao Wu,Lei Xue,Changyin Sun
出处
期刊:IEEE Transactions on Circuits and Systems I-regular Papers
[Institute of Electrical and Electronics Engineers]
日期:2025-05-20
卷期号:72 (11): 7319-7331
被引量:3
标识
DOI:10.1109/tcsi.2025.3569082
摘要
This article adopts the event-triggered control strategy (E-TCS) to achieve the practical predefined-time synchronization (PPTS) for dynamic complex networks (DCNs) with Markov switching topologies under stochastic denial-of-service (SDoS) attacks. We consider the Markov switching topologies, and the coupling weight of the complex networks between nodes is dynamic. For the proposed E-TCS, the minimum inter-event interval can be directly obtained, thereby eliminating the Zeno phenomenon. By employing the time-varying function, all states of the DCNs can achieve PPTS within the predefined time. Concretely, in contrast to finite/fixed-time synchronization, utilizing PPTS enables the arbitrary setting of convergence time, independent of initial values and controller parameters. Notably, the SDoS attacks occur with a certain probability within the attack intervals. Moreover, the intermittent attacks and the average non-attack rate are considered, and this approach leads to less conservative results. Additionally, we prove that a higher average non-attack rate makes it easier for all states of the DCNs to achieve PPTS. Finally, the validity of the proposed E-TCS is verified by the examples of Chua’s circuit and the Kuramoto oscillator network.
科研通智能强力驱动
Strongly Powered by AbleSci AI