网络安全与数据治理2025，Vol.44Issue(5)：35-41,7.DOI:10.19358/j.issn.2097-1788.2025.05.006
一种基于DRSN-GAN的通信信号调制识别方法
A communication signal modulation identification method based on DRSN-GAN
摘要
Abstract
A deep learning framework based on deep residual shrinkage network and generative adversarial network(DRSN-GAN)is proposed to address the problem of low recognition rate of communication signal modulation under small samples and low signal-to-noise ratio conditions.First,the in-phase orthogonal data(I/Q data)of the signal is used as the model input,and the dataset is expanded by the generative data generated by the generator,which effectively solves the problem of scarcity of high-quality data and enhances the generalization ability of the model.The DRSN is utilized to form a discriminator,and the expanded data is fed into the DRSN for training.Meanwhile,global average pooling is executed on the input data in the spatial dimension,and the channel attention module is used to extract the contextual features of the I/Q signals,which effectively reduces the noise interference.The method solves the problem of low recognition accuracy due to the difficulty of applying fixed thresholds to all samples,and significantly improves the recognition effect in a low signal-to-noise ratio environment.The experimental results show that the model proposed in this paper has an accuracy of 92％at a signal-to-noise ratio of 0 dB,which improves the overall classi-fication accuracy by 3％compared with other models,and exhibits stronger robustness under the conditions of small samples and low signal-to-noise ratio.关键词
调制识别/残差收缩网络/生成对抗网络/深度学习Key words
modulation recognition/residual shrinkage network/generative adversarial network/deep learning分类
电子信息工程引用本文复制引用
刘高辉,顾家华..一种基于DRSN-GAN的通信信号调制识别方法[J].网络安全与数据治理,2025,44(5):35-41,7.基金项目
国家自然科学基金(61671375) （61671375）
