Abstract

During the operation of a system for which a semi-Markov model has been constructed, it is not always possible to obtain all the information contained in the state codes when the system states change, but it is possible to obtain some signal (information) associated with the states of a semi-Markov model. In this case, the states of the semi-Markov model can be viewed as hidden (unobservable). In this article, using the example of a superposition of two independent restoration processes, we consider an approach to the construction of a hidden Markov model based on a semi-Markov process with a state space of a general form and the use of such a model for analyzing the operation of the modeled system based on the signal vector obtained.

中文翻译：

---

基于两个恢复过程叠加的隐马尔可夫模型

摘要

构建了半马尔可夫模型的系统在运行过程中，当系统状态发生变化时，并不总是可以获得状态码中包含的所有信息，但可以获得一些相关的信号（信息）与半马尔可夫模型的状态。在这种情况下，半马尔可夫模型的状态可以被视为隐藏的（不可观察的）。在本文中，以两个独立恢复过程的叠加为例，我们考虑了一种基于具有一般形式状态空间的半马尔可夫过程的隐马尔可夫模型的构建方法，以及这种模型的使用用于基于获得的信号向量分析建模系统的操作。