| 一种基于BP网的非线性变换信号模式识别算法 |
| |
| 引用本文： | 刘茂,全厚德.一种基于BP网的非线性变换信号模式识别算法[J].成都信息工程学院学报,2008,23(2):174-177. |
| |
| 作者姓名： | 刘茂 全厚德 |
| |
| 作者单位： | 军械工程学院光学与电子工程系,河北,石家庄,050000 |
| |
| 摘 要： | 为了获取常用超短波信号的特征参数,采取一种非线性变换的方法。根据信号参数的分布信息,对信号特征不够明显的参数进行二次变换,使所有信号的特征参数都具有较好的区分度,并利用BP网对这些特征参数进行识别。大量的仿真实验表明,该方法参数提取容易,计算量较小,在较低信噪比下能够较好地识别常用的超短波信号。识别的信号为2FSK、MSK、QAM、DPSK和FM。
|
| 关 键 词： | 模式识别 非线性变换 特征参数 |
A modulation scheme recognition arithmetic based on nonlinear transform and BP neural network |
| |
| LIU Mao,QUAN Hou-de.A modulation scheme recognition arithmetic based on nonlinear transform and BP neural network[J].Journal of Chengdu University of Information Technology,2008,23(2):174-177. |
| |
| Authors: | LIU Mao QUAN Hou-de |
| |
| Institution: | LIU Mao,QUAN Hou-de (Ordnance Engineering College,Shijiazhuang 050000,China) |
| |
| Abstract: | The nonlinear transform is used to abstract the characters of the usual ultrashort signals. Based on the character analysis the less-characteristic characters are retransformed and all the characters are recognized with the BP neural network. The characters are easy to be abstracted. Better results are obtained for the ultrashort signals with Gaussian noise whose SNR are 6-20 dB. Five kinds of modulation signals; namely 2FSK, MSK, QAM, DPSK and FM are discussed. |
| |
| Keywords: | scheme recognition nonlinear transform character |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
