不完备离散事件系统的当前状态不透明性

本文针对不完备系统模型,研究不完备离散事件系统的当前状态不透明性.根据系统的实际输出与预测输出之间的差异,构建了一个具有学习功能的学习诊断器.这种学习诊断器不仅能够模拟系统的状态转移,而且还可以将系统缺失的状态信息通过学习得到恢复.通过引入集合覆盖理论处理由学习诊断器得出的结果,提出了一种基于学习诊断器的不完备离散事件系统当前状态不透明性的验证算法.

Current-state opacity of incomplete discrete-event systems

In recent years, the opacity of discrete event systems (DESs) has received considerable attention, which has been successfully applied to many information technology areas such as digital signature, communication security, information authentication, intrusion detection and data encryption. This paper aims to propose an approach of the current-state opacity for incomplete DESs in which some information may be unavailable or even missing. According to the difference between the actual output and the predicted output of the incomplete system, a learning diagnoser is constructed. Note that the learning diagnoser not only can simulate the state transition of the system, but also can restore the absent state information from the system through learning. And the set coverage theory is introduced to deal with the results obtained by the learning diagnoser. A method to verify the current state opacity of an incomplete system is proposed based on the learning diagnoser. Moreover, the construction of the learning diagnoser and the verification of the current state opacity are illustrated by an example arising from the pressure testing process of special steel.