数据驱动的RF信号深度调制识别方法  

A Data-driven Deep Modulation Identification Method for RF Signals

作　　者：徐亚军[1] 郭恩豪 陈林 司成可 XU Ya-jun;GUO En-hao;CHEN Lin;SI Cheng-ke(Aviation Engineering College,Civil Aviation Flight University of China,Guanghan 618307,China;Tongfang Electronic Science and Technology Co.Ltd.,Jiujiang 332000,China)

机构地区：[1]中国民用航空飞行学院航空工程学院,四川广汉618307 [2]同方电子科技有限公司,江西九江332000

出　　处：《计算机与现代化》2022年第6期80-86,95,共8页Computer and Modernization

基　　金：中国民用航空飞行学院重点研究基金资助项目(ZJ2020-04);大学生创新创业训练项目(S202010624089)。

摘　　要：基于卷积结构的信号调制识别神经网络的识别性能受信号调制类型种类限制。例如,在12 d B信噪比条件下,同时对24种信号调制类型进行识别,其识别准确率仅为80%。若需要进一步提高识别性能,则要求更复杂的网络模型,导致网络训练所需数据集规模和硬件资源成本增大。鉴于此,针对无线电信号特征,设计一种适用于无线电信号调制识别的紧致残差神经网络,将其作为信号调制类型特征学习和特征提取工具,实现从原始I、Q数据到信号调制类型的端到端识别。利用迁移学习降低网络重新训练所需样本数,增强在无线信道响应发生变化时的环境适应能力,降低训练阶段所需的硬件资源和训练数据集规模。研究表明,当信道脉冲响应改变时,所提的信号调制识别神经网络在信噪比为12 d B条件下的识别性能达到95%,多个对比实验验证本文所设计神经网络的识别性能具有优势。The identification performance with convolutional neural network( CNN) is limited with the types of signal modulation identification.For instance,the identification accuracy is just only 80% when 24 kinds of modulation waveforms presented at SNR = 12 d B.If a better recognition performance wants to be obtained,more complicated network must be required.It directly enlarges the requirement of the data set size and the cost of hardware calculation resources is also increased.Therefore,a compact residual neural network for radio signal modulation identification is designed in the paper,which can be used to extract the characteristics of signal modulation.The end-to-end identification is accomplished from the baseband in-phase and quadrature components.By using transfer learning,the number of samples in the network retraining stage is reduced dramatically and the adaptive ability of the proposed network is enhanced.The test results show that the identification performance with the proposed neural network approaches 95% when the SNR is 12 d B even though the wireless channel impulse response verified.Several comparative experiments illustrate the advantages of the proposed neural network.

关 键 词：深度调制识别 残差神经网络 迁移学习 数据驱动 卷积神经网络 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]