基于贝叶斯网络的反舰导弹身份识别

为了解决反舰导弹身份识别所面临的实时性和信息不确定性问题,提出了采用贝叶斯网络识别反舰导弹的方法。设计了4种贝叶斯网络分类器,分别在导弹仿真数据集和UCI数据集上作了测试,比较了它们各自的分类性能。实验结果表明,朴素贝叶斯网络的分类准确率虽然比其它分类器稍低,但它简单有效,稳健性比其它分类器都好,可用于反舰导弹身份的实时识别。

Recognition for Anti-Ship Missiles Based on Bayes Network

In this paper,Bayes Network classifiers are proposed to fulfill real-time requirements and to deal with uncertainties in the identification of anti-ship missiles(ASMs).Four types of Bayes Network classifiers are built and tested by using the simulated date sets and the date sets in UCI Repository,and their classification performance are compared.The experimental results show that the Na?ve Bayes Classifier is simple and effective,especially,its robust performance is the best among the four classifiers,and it can be used in the identification of ASMs in real time,though its classification accuracy is a little lower than the other three classifiers.