期刊文献+

混响背景中信号检测的特征核支持向量机 被引量:4

Feature kernel support vector machines for signal detection with reverberation
下载PDF
导出
摘要 针对支持向量机的核函数对检测性能的影响,分析了核函数在特征空间中的作用.利用混响和目标回波的非高斯特性上的差异设计了特征核函数,即将核函数改进为任意2个样本高阶统计量的几何均值与原核函数乘积的形式,使其自适应调整核函数值,从而提高分类性能.实验证明了采用特征核函数后扩大了2类样本间的距离,并且仍然满足Mercer定理.将特征核支持向量机应用于高斯或非高斯分布混响背景中的信号检测,结合实际应用给出了训练和检测算法.实验及仿真研究表明,当选取2类样本差异较大的高阶统计量作为特征量时,混响背景为非高斯分布时,其检测性能优于匹配滤波器以及基于传统核函数的支持向量机. Because kernel functions in a support vector machine influence detection performance, analysis of the effects of kernel features is needed to effectively design feature kernel functions. A new kernel was developed by the product of original kernel functions and the geometric mean value of the high order statistics of two arbitrary samples. The new kernel function is adaptively adjusted by using the non-Gaussian differences between reverberation and target echoes for improving classification performance. It was also proven that it is possible to enlarge the differences between two kinds of samples when using the feature kernel. The suggested kernel also satisfies the Mercer theorem. The proposed feature kernel support vector machine was used for signal detection of Gaussian and nonGaussian reverberation. The training and detecting algorithms used in testing are given. The results of experiments and simulations showed that when the data tested has significant differences in features, and the reverberation has a non-Gaussian distribution, the new algorithm's performance is better than matching filters and support vector machines based on traditional kernel functions.
出处 《哈尔滨工程大学学报》 EI CAS CSCD 北大核心 2009年第1期52-59,共8页 Journal of Harbin Engineering University
关键词 信号检测 混响 非高斯分布 高阶统计量 支持向量机 signal detection reverberation non-Gaussian distribution high order statistics support vector ma- chines
  • 引文网络
  • 相关文献

参考文献15

  • 1XU Bohou, ZENG Lingzao, LI Jianlong. Application of stochastic resonance in target detection in shallow water reverberation[ J ]. Journal of Sound and Vibration, 2007 ( 1 ) : 255-263.
  • 2赵申东,唐劲松,黄海宁,蔡志明.三维空时多波束方法在抗混响中的应用[J].声学学报,2008,33(2):124-130. 被引量:8
  • 3GUILLAUME G, JOURDAIN verse algorithm for detection in [ J ]. IEEE Journal of Oceanic 310-321. G. Principal component inthe presence of reverberation Engineering, 2002, 27 ( 2 ) :
  • 4ABRAHAM D A. Non-Rayleigh reverberation and clutter [J]. IEEE Journal of Oceanic Engineering, 2004, 29(2) : 193-195.
  • 5ABRAHAM D A. Signal excess in K distributed reverberation [ J ]. IEEE Journal of Oceanic Engineering, 2003, 28 ( 3 ) :526-536.
  • 6KHAIRNAR D G, MERCHANT S N, DESAI U B. Aneural solution for signal detection in non-Gaussian noise [ C ]//International Conference on Information Technology. Las Vegas, USA , 2007.
  • 7GANDHI P P, RAMAMURTI V. Neural networks for signal detection in non-Gaussian noise [ J ]. IEEE Transactions on Signal Processing, 1997, 45 ( 11 ) :2846-285t.
  • 8VAPNIK V N. Statistic learning theory [ M ]. New york: John Wiley & Sons, 1998.
  • 9MICCHELLI C A, PONTIL M. Learning the kernel function via regularization [ J ]. Journal of Machine Learning Research, 2005, 6 : 1099-1125.
  • 10AMARI S, WU S. Improving support vector machine classifters by modifying kernel functions [ J ]. Nenral Networks, 1999, 12(12):783-789.

二级参考文献31

共引文献37

同被引文献53

  • 1李军,侯朝焕.基于多尺度特征的匹配滤波处理[J].声学学报,2004,29(4):313-318. 被引量:5
  • 2陈振洲,李磊,姚正安.基于SVM的特征加权KNN算法[J].中山大学学报(自然科学版),2005,44(1):17-20. 被引量:52
  • 3邓兵,陶然,齐林,刘锋.基于分数阶傅里叶变换的混响抑制方法研究[J].兵工学报,2005,26(6):761-765. 被引量:24
  • 4孙大军,田坦,张殿伦.跳频脉冲信号在水下沉底小目标探测中的应用[J].哈尔滨工程大学学报,2007,28(9):1025-1029. 被引量:4
  • 5肖先赐.现代谱估计:原理与应用[M].哈尔滨:哈尔滨工业大学出版社,1989:30-35.
  • 6王玉泉.水声设备[M].哈尔滨:哈尔滨工程大学出版社,1999:234-237.
  • 7吴国清.自适应分段技术在回波分析中的应用[J].应用声学,1985(4):23-31.
  • 8D. A. Abraham, A. P. Lyons. Reverberation envelope statistics and their dependence on sonar beamwidth and bandwidth in Impact of Littoral Environmental Variability on Acoustic Predictions and Sonar Performance [ J ]. N. G. Pace and F. B. Jensen Eds. Norwell MA: Kluwer, 2002:539 - 546.
  • 9Cannillet V, Joulain G. Wideband sonar detection in reverberati on using autoregressive models [ J ]. Proc Oceans96 MTSIIEEE, Fort Laud. Erdale, 1996 ( 3 ) : 1435 - 1440.
  • 10Carmillet V, Amblard Po. Detection of phase - or frequency- modulated signals in reverberation noise [ J ]. JASA, 1999 ( 6): 3375 - 3389.

引证文献4

二级引证文献5

;
使用帮助 返回顶部