《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (4): 1318-1322.DOI: 10.11772/j.issn.1001-9081.2022030425

• 前沿与综合应用 • 上一篇    

基于卷积长短时深度神经网络的信号调制方式识别方法

杨海宇, 郭文普(), 康凯   

  1. 火箭军工程大学 作战保障学院,西安 710025
  • 收稿日期:2022-04-06 修回日期:2022-08-26 接受日期:2022-09-07 发布日期:2023-01-11 出版日期:2023-04-10
  • 通讯作者: 郭文普
  • 作者简介:杨海宇(1993—),男,陕西咸阳人,硕士研究生,主要研究方向:信号处理、深度学习、雷达辐射源个体识别;
    康凯(1987—),男,陕西西安人,讲师,博士,主要研究方向:信号处理、数据挖掘、人工智能。

Signal modulation recognition method based on convolutional long short-term deep neural network

Haiyu YANG, Wenpu GUO(), Kai KANG   

  1. Academy of Combat Support,Rocket Force University of Engineering,Xi’an Shaanxi 710025,China
  • Received:2022-04-06 Revised:2022-08-26 Accepted:2022-09-07 Online:2023-01-11 Published:2023-04-10
  • Contact: Wenpu GUO
  • About author:YANG Haiyu, born in 1993, M. S. candidate. His research interests include signal processing, deep learning, radar emitter individual identification.
    KANG Kai, born in 1987, Ph. D., lecturer. His research interests include signal processing, data mining, artificial intelligence.

摘要:

针对信号调制方式识别计算复杂度高、低信噪比(SNR)条件下识别率较低、网络结构相对单一的问题,提出一种基于卷积长短时深度神经网络(CLDNN)的信号调制方式识别方法。首先,采用基准开源数据集RadioML2016.10a,对该数据集做同相正交(I/Q)数据转换,并将得到的结果作为网络输入;其次,构建CLDNN模型,模型分为三层卷积神经网络(CNN)、两层长短期记忆(LSTM)网络和两层全连接网络(FCN);最后,对所提模型进行训练及测试,得到分类结果。实验结果表明,对11种信号在不同SNR下进行调制方式识别时,与现有的单一网络结构模型如残差神经网络(RES)模型、CNN模型和残差生成对抗网络(RES-GAN)模型进行对比,随着SNR的提升,CLDNN模型的识别准确率也随之提高,且CLDNN模型的识别准确率均高于其他3种对比模型,当SNR在4 dB以上时,达到了92%。

关键词: 调制方式识别, 深度学习, 卷积神经网络, 长短期记忆网络, 深度神经网络

Abstract:

Focused on the high computational complexity, low recognition rate under the condition of low Signal-to-Noise Ratio (SNR), and relatively simple network structure, a signal modulation recognition method based on Convolutional Long short-term Deep Neural Network (CLDNN) was proposed. Firstly, the open-source benchmark dataset RadioML2016.10a was adopted, and In-phase/Quadrature (I/Q) data conversion was performed on it, then the obtained result was used as the network input. Secondly, the CLDNN model was constructed, which was divided into three parts, that is three-layer Convolutional Neural Network (CNN), two-layer Long Short-Term Memory (LSTM) network, and two-layer Fully Connected Network (FCN). Finally, the proposed model was trained and tested to obtain classification results. Experimental results show that recognition accuracy of CLDNN model increases with SNR improvement and reaches 92% with SNR bigger than 4 dB, which is higher than those of the existing single network structure models such as Residual Neural Network (RES) model, CNN model and RESidual Generative Adversarial Network (RES-GAN) model, in the modulation recognition of 11 kinds of signals at different SNR.

Key words: modulation recognition, deep learning, Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM) network, Deep Neural Network (DNN)

中图分类号: