作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2013, Vol. 39 ›› Issue (8): 223-226,230. doi: 10.3969/j.issn.1000-3428.2013.08.048

• 人工智能及识别技术 • 上一篇    下一篇

一种基于支持向量机的信号调制分类方法

徐 闻,王 斌   

  1. (解放军信息工程大学信息工程学院,郑州 450002)
  • 收稿日期:2012-02-27 出版日期:2013-08-15 发布日期:2013-08-13
  • 作者简介:徐 闻(1987-),男,硕士研究生,主研方向:通信信号调制分析;王 斌,教授

A Signal Modulation Classification Method Based on SVM

XU Wen, WANG Bin   

  1. (Institute of Information Engineering, PLA Information Engineering University, Zhengzhou 450002, China)
  • Received:2012-02-27 Online:2013-08-15 Published:2013-08-13

摘要: 针对现有数字信号调制分类的问题,在人工分类的基础上,提出一种基于支持向量机(SVM)的自动分类方法。提取信号的高阶累计量特征参数用于训练与测试数据。比较已有的基于SVM的调制分类方法,采用应用混合核函数的SVM分类方法,并利用决策树二分类思想设计分类流程。经过仿真比较,验证了该混合核函数的SVM具有较好的分类性能。

关键词: 调制分类, 高阶累积量, 特征参数, 支持向量, 混合核函数, 二分类

Abstract: This paper addresses the problem of automatic modulation classification of the existed digital signal, according to the method of artificial classification, a classification method based on Support Vector Machine(SVM) is developed. The characteristic parameter of high-order cumulants of the signal is used for training data and test data. Comparing with the method of SVM for modulation classification existed, this method adopts SVM based on mixture kernel function and makes use of two classification of decision tree’s way to design classification process. By comparing the simulation results, indicates that the mixture kernel function of the SVM has a better classification performance.

Key words: modulation classification, high-order cumulant, feature parameter, Support Vector Machine(SVM), mixture kernel function, two classification

中图分类号: