期刊文献+

一种新的基于支持向量机的自动调制识别方案 被引量:9

Novel Scheme of Automatic Modulation Recognition Based on SVM
下载PDF
导出
摘要 为了解决在合作或非合作的通信应用领域中(如软件无线电,电子侦察系统等)多种调制信号之间的切换问题,提出1种基于多类别支持向量机(SVM)的模拟和数字信号的调制识别的新方案。SVM将特征向量非线性地映射到高维特征空间中,并建立1个最优超平面来实现信号调制方式的分类。这种方法避免了在人工神经网络中的过学习、欠学习以及局部最小化的问题。仿真中将应用于调制识别的sVM算法与人工神经网络算法(ANN)做了比较,结果表明SVM自动调制识别方法结构简单,识别率高,解决小样本的能力强,在信噪比SNR不低于5 dB时,正确识别率达到94%以上,适于在工程中应用。 In order to solve the conversion problem of multi-class modulation types which exists in such cooperative or non-cooperative communication applications as software radio, electronic surveillance systems, a new scheme based on multi-class support vector machine (SVM) for recognition of both analog and digital modulation signals is presented. The SVM maps input vectors nonlinearly into a high dimensional feature space and constructs the optimum separating hyperplane to realize the signal recognition. This method avoids overfitting, underfitting and local minimum in neural networks. The simulation compares the SVM with the ANN (artificial neural network) algorithm in modulation recognition. The results show that the proposed modulation recognition algorithm has the advantages of simple structure, higher success rates, small-sample-problem-solving ability. The success rates are over 94% when the signal noise rates are not lower than 5 dB. The proposed scheme is fit for engineering applications.
作者 吴丹 顾学迈
出处 《南京理工大学学报》 EI CAS CSCD 北大核心 2006年第5期569-572,591,共5页 Journal of Nanjing University of Science and Technology
基金 国家"863"项目(2004AA001210)
关键词 支持向量机 最优超平面 核函数 调制识别 support vector machine optimal hyperlane kernel function modulation recognition
  • 引文网络
  • 相关文献

参考文献9

  • 1Andi A K,Azzouz E E.Automation modulation recognition[J].Signal Processing,1995,46 (2):211-222.
  • 2Azzouz E E,Nandi A K.Automatic Modulation Recogtnition-Ⅱ[J].Journal of the Franklin Institule,1997,334 (12):275-305.
  • 3Wong M L D,Nandi A K.Automatic digital modulation recognition using spectral and statistical features with multilayer perceptions[A].Sixth International Symposium on Signal Processing and its Applications (ISSPA)[C].Kuala Lumpur:ISSPA,2001.390-393.
  • 4Vapnik V N.Statistical Learning Theory[M].New York:John Wley and Sons,1998.
  • 5Devroye L,Gyr(o)fi L,Lugosi L.A Probablistic Theory of Pattern Recognition[M].New York:Springer,1996.
  • 6Daniel J S,James A B.Support vector machine techniques for nonlinear equalization[J].IEEE Transactions on Signal Processing,2000,11 (48):3 271 -3 227.
  • 7Nandi A K,Azzouz E E.Algorithms for automatic modulation recognition of communication signals[J].IEEE Transactions on Communications,1998 (4):431-436.
  • 8刘志刚,李德仁,秦前清,史文中.支持向量机在多类分类问题中的推广[J].计算机工程与应用,2004,40(7):10-13. 被引量:150
  • 9Keerthi S S,Lin C J.Asymptotic behaviors of support vector machines with Gaussian kernel[J].Neural Computation,2003,15 (7):1 667-1 689.

共引文献149

同被引文献86

引证文献9

二级引证文献83

;
使用帮助 返回顶部