| 引用本文: | 周华吉,徐杰,郑仕链,等.面向通信信号调制识别的半监督生成对抗网络框架.[J].国防科技大学学报,2023,45(6):78-83.[点击复制] |
| ZHOU Huaji,XU Jie,ZHENG Shilian,et al.Semi-supervised generative adversarial network framework for modulation recognition of communication signals[J].Journal of National University of Defense Technology,2023,45(6):78-83[点击复制] |
|
| |
|
|
| 本文已被：浏览 3192次 下载 2323次 |
| 面向通信信号调制识别的半监督生成对抗网络框架 |
| 周华吉，徐杰，郑仕链，沈伟国，王巍，楼财义 |
|
(1.电磁空间安全全国重点实验室;2.西安电子科技大学 人工智能学院)
|
| 摘要: |
| 针对只有少量标签数据的弱监督条件下现有调制信号识别模型准确率较低的问题，提出基于生成对抗网络的半监督学习框架。该方法通过对通信信号进行冗余空域变换，使其在适应生成对抗网络模型的同时保留丰富的信号相邻特征；通过梯度惩罚Wasserstein生成对抗网络的引入，构建适宜电磁信号处理的半监督学习框架，实现对无标签信号样本的有效利用。为了验证所提算法的有效性，在RADIOML 2016.04C数据集上进行测试。实验结果表明，该方法在半监督条件下能训练出高效的分类器，获得优异的调制识别结果。 |
| 关键词: 生成对抗网络 半监督学习框架 通信信号 调制识别 |
| DOI：10.11887/j.cn.202306011 |
| 投稿日期：2021-11-26 |
| 基金项目:国家自然科学基金资助项目（61772401,U19B2015，U19B2016，61871398） |
|
| Semi-supervised generative adversarial network framework for modulation recognition of communication signals |
| ZHOU Huaji1,2, XU Jie1, ZHENG Shilian1, SHEN Weiguo1, WANG Wei1, LOU Caiyi1 |
|
(1.National Key Laboratory of Electromagnetic Space Security, Jiaxing 314033 , China;2.School of Artificial Intelligence, Xidian University, Xi′an 710071 , China)
|
| Abstract: |
| Aiming at the problem that the accuracy of the existing modulation signal recognition model was low under the condition of weak supervision with only a small amount of labeled data, a semi supervised learning framework based on generated countermeasure network was proposed. By performing a redundant spatial transformation on the communication signals, the method can adapt to the generative adversarial network model and retain rich signal adjacent features. Through the introduction of Wasserstein generative adversarial network-gradient penalty, a semi-supervised learning framework suitable for electromagnetic signal processing was constructed to realize the effective utilization of unlabeled signal samples. In order to verify the effectiveness of the proposed algorithm, sufficient experiments were conducted on the RADIOML 2016.04C dataset. Experimental results show that the proposed method can train an efficient classifier under semi-supervised conditions and obtain excellent modulation recognition results. |
| Keywords: generative adversarial network semi-supervised learning framework communication signal modulation recognition |
|
|
