This paper shows solicitude for the quantization synchronization of delayed chaotic master and slave neural networks under an dynamic event-triggered strategy. In virtue of a generalized Halanay-type inequality, a theoretical criterion for quasi-synchronization of master and slave neural networks is derived. Meanwhile, we can obtain an exact upper bound of synchronization error by using this criterion. Compared with output feedback controller with event triggering and quantization, the case where the controller only affected by quantization is also considered. Then, we exclude the Zeno behavior of the event-triggered controller. A sufficient criterion for the existence of the quantized output feedback controllers is also provided. A numerical example is cited to illustrate the efficiency of our theoretical criteria. In addition, some experiments of secure image communication are conducted under quasi-synchronization of master and slave neural networks.

本文关注动态事件触发策略下延迟混沌主从神经网络的量化同步。根据广义Halanay型不等式，推导了主从神经网络准同步的理论判据。同时，利用这个准则，我们可以得到一个准确的同步误差上界。与具有事件触发和量化的输出反馈控制器相比，还考虑了控制器仅受量化影响的情况。然后，我们排除事件触发控制器的 Zeno 行为。还提供了量化输出反馈控制器存在的充分标准。引用了一个数值例子来说明我们的理论标准的效率。此外，