| 基于卷积神经网络的调制识别新方法 |
| |
| 引用本文： | 王鹏,张君毅,赵国庆. 基于卷积神经网络的调制识别新方法[J]. 无线电工程, 2019, 0(6): 453-457 |
| |
| 作者姓名： | 王鹏 张君毅 赵国庆 |
| |
| 作者单位： | 1.中国电子科技集团公司第五十四研究所;2.西安电子科技大学电子工程学院 |
| |
| 基金项目： | 第63批中国博士后科学基金面上资助(2018M631766) |
| |
| 摘 要： | 针对通信信号调制方式识别问题,提出了一种基于卷积神经网络的通信信号调制方式识别新方法,利用深度卷积网络实现了通信信号特征的自学习,避免了传统算法中特征提取与选择问题,并设计了基于自学习特征的分类器,实现了通信信号调制方式的识别。仿真结果表明,利用卷积神经网络实现通信信号调制方式的识别是可行、有效的。
|
| 关 键 词： | 调试识别 卷积神经网络 深度学习 机器学习 |
A Novel Method Based on Convolutional Neural Networks for Modulation Recognition |
| |
| WANG Peng,ZHANG Junyi,ZHAO Guoqing. A Novel Method Based on Convolutional Neural Networks for Modulation Recognition[J]. Radio Engineering of China, 2019, 0(6): 453-457 |
| |
| Authors: | WANG Peng ZHANG Junyi ZHAO Guoqing |
| |
| Affiliation: | (The 54th Research Institute of CETC,Shijiazhuang 050081,China;College of Electronic Engineering,Xidian University,Xi ’an 710071,China) |
| |
| Abstract: | In order to realize modulation recognition,a novel method based on convolutional neural networks (CNN) is proposed.The features used for modulation recognition are generated automatically by adopting CNN,which avoids the extraction and selection of features in traditional methods based on machine learning.In order to realize the recognition of modulation style,the Classifier is designed based on the self-learning features.Simulation results demonstrate that the method based on CNN is feasible and effective for modulation recognition. |
| |
| Keywords: | modulation recognition convolutional neural networks deep learning machine learning |
| 本文献已被 维普 等数据库收录！ |
|
