人工免疫系统中淋巴细胞的下推自动机模型

有些淋巴细胞的构造模型，如r-邻域位线性构造模型，识别能力不够强，尤其对正则表达式缺乏识别能力。该文以下推自动机模型为原理，提出了一种自动机模型的淋巴细胞，这种淋巴细胞具有比有限自动机更强的识别能力，同时利用相近的入侵具有相似性这一点，引进了一个经验队列，配合HMM中状态转移概率的概念，对状态转移进行评判。最后给出了模型具体的设计方案。

Pushdown Automata Model Detector of Lymphocyte in Artificial Immune System

Some construction models of lymphocyte, for example r-border upon linearity model, has not enough detect ability, especially lack of the ability to detect regular expression. This paper based on pushdown automata model, suggests a new automata model lymphocyte. This kind of lymphocyte has stronger detect ability, and as the similarity of border upon attack is concerned, it imports an experience queue, cooperates with the concept of probability in HMM, to make the judgement of state change. At last, it gives the exact design blue print of the model.