Fixed-time dual-channel event-triggered secure quasi-synchronization of coupled memristive neural networks
Fixed-time dual-channel event-triggered secure quasi-synchronization of coupled memristive neural networks
复制标题
DOI：
10.1016/j.jfranklin.2021.10.023
复制
发表时间：
2021-10
期刊：
影响因子：
--
通讯作者：
Yuangui Bao;Yijun Zhang
中科院分区：
文献类型：
--
作者：
Yuangui Bao;Yijun Zhang
研究耦合记忆神经网络(CMNN)的固定时间准同步问题。通信信道受到伯努利随机变量描述的欺骗攻击。为了减少信号传输，提出了一种双通道事件触发机制。在传感器到控制器、控制器到执行器的每个通道中，设计了一种事件触发机制。与通信环路中的单一事件触发机制相比，主要的难点在于如何处理分组调度和网络攻击问题。利用Lyapunov方法，结合一个新的引理，给出了保证CMNN先导跟随准同步的充分条件。所设计的双通道事件触发机制排除了Zeno行为。讨论了事件触发机制对估计稳定时间的影响。给出了三个数值算例，验证了理论结果的有效性。
This paper is concerned with the fixed-time quasi-synchronization of coupled memristive neural networks (CMNNs). The communication channel is subject to the deception attack described by the Bernoulli stochastic variable. To reduce signal transmissions, a dual-channel event-triggered mechanism is proposed. In each channel of sensor to controller and controller to actuator, an event-triggered mechanism is designed. Compared with the single event-triggered mechanism in the communication loop, the main difficulties lie in how to deal with the problems of packet scheduling and network attacks. By using Lyapunov method combining with a new proposed lemma, some sufficient conditions are derived to guarantee the leader-following quasi-synchronization of CMNNs. The Zeno behavior is excluded for the designed dual-channel event-triggered mechanism. The influence of the event-triggered mechanism on the estimation of settling time is discussed. Three numerical examples are provided to show the effectiveness of the theoretical results.