当前位置: X-MOL 学术IEEE Trans. Syst. Man Cybern. Syst. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Input-Based Event-Triggering Consensus of Multiagent Systems Under Denial-of-Service Attacks
IEEE Transactions on Systems, Man, and Cybernetics: Systems ( IF 8.7 ) Pub Date : 2020-04-01 , DOI: 10.1109/tsmc.2018.2875250
Yong Xu , Mei Fang , Zheng-Guang Wu , Ya-Jun Pan , Mohammed Chadli , Tingwen Huang

This paper applies an input-based triggering approach to investigate the secure consensus problem in multiagent systems under denial-of-service (DoS) attacks. The DoS attacks are based on the time-sequence fashion and occur aperiodically in an unknown attack strategy, which can usually damage the control channels executed by an intelligent adversary. A novel event-triggered control scheme on the basis of the relative interagent state is developed under the DoS attacks, by designing a link-based estimator to estimate the relative interagent state between intermitted communication instead of the absolute state. Compared with most of the existing work on the design of the triggering condition related to the state measurement error, the proposed triggering condition is designed based on the control input signal from the view of privacy protection, which can avoid continuous sampling for every agent. Besides, the attack frequency and attack duration of DoS attacks are analyzed and the secure consensus is reachable provided that the attack frequency and attack duration satisfy some certain conditions under the proposed control algorithm. “Zeno phenomenon” does not exhibit by proving that there exist different positive lower bounds corresponding to different link-based triggering conditions. Finally, the effectiveness of the proposed algorithm is verified by a numerical example.

中文翻译:

拒绝服务攻击下多代理系统基于输入的事件触发共识

本文应用基于输入的触发方法来研究拒绝服务 (DoS) 攻击下多代理系统中的安全共识问题。DoS 攻击基于时间序列方式,以未知的攻击策略不定期发生,通常会损坏智能对手执行的控制通道。在 DoS 攻击下,通过设计基于链路的估计器来估计间歇通信之间的相对代理状态而不是绝对状态,开发了一种基于相对代理状态的新型事件触发控制方案。与现有大部分与状态测量误差相关的触发条件设计工作相比,本文从隐私保护的角度出发,基于控制输入信号设计触发条件,这样可以避免对每个代理进行连续采样。此外,分析了DoS攻击的攻击频率和攻击持续时间,在所提出的控制算法下,只要攻击频率和攻击持续时间满足一定的条件,就可以达成安全共识。通过证明不同的基于链接的触发条件对应存在不同的正下界,“芝诺现象”不会出现。最后,通过数值算例验证了所提算法的有效性。通过证明不同的基于链接的触发条件对应存在不同的正下界,“芝诺现象”不会出现。最后,通过数值算例验证了所提算法的有效性。通过证明不同的基于链接的触发条件对应存在不同的正下界,“芝诺现象”不会出现。最后,通过数值算例验证了所提算法的有效性。
更新日期:2020-04-01
down
wechat
bug