基于SVM决策树的数据链识别分类  

Classification and Recognition of Data Link Based on SVM Decision Tree

作　　者：杨咪 王安丽 胡正 Yang Mi;Wang Anli;Hu Zheng(School of Aeronautics,Northwestern Polytechnical University,Xi’an 710072,China)

机构地区：[1]西北工业大学航空学院

出　　处：《兵工自动化》2019年第12期54-57,63,共5页Ordnance Industry Automation

基　　金：国家自然科学基金(61573283)

摘　　要：为解决数据链信号识别分类的问题,提出一种基于支持向量机(support vector machine,SVM)决策树的数据链识别分类方法。通过分析美军常用的数据链通信信号特征,采用小波变换法分析数据链的特征信息,得出小波系数与信号能量分布的关系,根据SVM算法原理,构建目标特征模型,对信号特征量进行识别分类,对SVM分类器的关键参数进行优化设计,并与BP神经网络算法进行对比实验仿真。结果表明:SVM决策树网络分类器在进行收敛速度和准确率表现优异,能改善分类识别效能。In order to solve the problem of data link signal identification and classification,a data link recognition and classification method based on Support Vector Machine(SVM)decision tree is proposed.By analyzing the characteristics of the data link communication signals commonly used by the US army,the wavelet transform method is used to analyze the characteristic information of the data link,and the relationship between wavelet coefficients and signal energy distribution is obtained.According to the principle of SVM algorithm,the target feature model is constructed to identify and classify the signal feature quantities.The key parameters of the SVM classifier are optimized and compared with the BP neural network algorithm.The results show that the SVM decision tree network classifier performs well in convergence speed and accuracy,and can improve classification and recognition performance.

关 键 词：数据链 小波变换 支持向量机 决策树 模式识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]