Hangjie Zhu
ABSTRACT
Deep learning-based modulation recognition algorithms for communication signals have poor recognition performance when only a small number of labeled signal samples are available. A modulation recognition method based on Inception-V3 transfer learning is proposed to solve this problem. This method first converts eight types of modulated signal samples from 0 to 20 dB into time-frequency diagrams as input and fine-tunes the pre-trained network Inception-V3 for transfer learning, conducts test experiments on three datasets with finitely labeled samples, and finally visualizes the feature learning of the training process to ensure that the features of the signals are correctly learned. The experimental results show that the proposed method can achieve the highest recognition accuracy of 94.07% at 20 dB with insufficient training samples, and still achieve 92.50% and 91.67% recognition accuracy at 20 dB with only 4400 and 1760 samples, respectively.
- Wenshi Xiao, Zhongqiang Luo, and Qian Hu. 2022. A Review of Research on Signal Modulation Recognition Based on Deep Learning. Electronics 11, 17 (September 2022), 2764. https://doi.org/10.3390/electronics11172764Google Scholar
Cross Ref
- Xiaoguang Zou and Xiaoyong Zou. 2022. Automatic Modulation and Recognition of Robot Communication Signal Based on Deep Learning Neural Network. Journal of Sensors 2022, (July 2022), 1–7. https://doi.org/10.1155/2022/3519010Google ScholarCross Ref
- Anni Lin, Zhiyuan Ma, Zhi Huang, Yan Xia, and Wenting Yu. 2020. Unknown Radar Waveform Recognition Based on Transferred Deep Learning. IEEE Access 8, (2020), 184793–184807. https://doi.org/10.1109/ACCESS.2020.3029192Google ScholarCross Ref
- Peng Wu, Bei Sun, Shaojing Su, Junyu Wei, Jinhui Zhao, and Xudong Wen. 2020. Automatic Modulation Classification Based on Deep Learning for Software-Defined Radio. Mathematical Problems in Engineering 2020, (November 2020), 1–13. https://doi.org/10.1155/2020/2678310Google ScholarCross Ref
- Sangkyu Kim, Hae-Yong Yang, and Daeyoung Kim. 2022. Fully Complex Deep Learning Classifiers for Signal Modulation Recognition in Non-Cooperative Environment. IEEE Access 10, (2022), 20295–20311. https://doi.org/10.1109/ACCESS.2022.3151980Google ScholarCross Ref
- Fuxin Zhang, Chunbo Luo, Jialang Xu, and Yang Luo. 2021. An Efficient Deep Learning Model for Automatic Modulation Recognition Based on Parameter Estimation and Transformation. Retrieved April 12, 2023 from http://arxiv.org/abs/2110.04980Google ScholarCross Ref
- Yu Wang, Guan Gui, Haris Gacanin, Tomoaki Ohtsuki, Hikmet Sari, and Fumiyuki Adachi. 2020. Transfer Learning for Semi-Supervised Automatic Modulation Classification in ZF-MIMO Systems. IEEE J. Emerg. Sel. Topics Circuits Syst. 10, 2 (June 2020), 231–239. https://doi.org/10.1109/JETCAS.2020.2992128Google ScholarCross Ref
- Tyler Cody and Peter A. Beling. 2023. A Systems Theory of Transfer Learning. IEEE Systems Journal 17, 1 (March 2023), 26–37. https://doi.org/10.1109/JSYST.2022.3224650Google ScholarCross Ref
- Fuzhen Zhuang, Zhiyuan Qi, Keyu Duan, Dongbo Xi, Yongchun Zhu, Hengshu Zhu, Hui Xiong, and Qing He. 2020. A Comprehensive Survey on Transfer Learning. Retrieved April 12, 2023 from http://arxiv.org/abs/1911.02685Google Scholar
- Jiang Hua, Liangcai Zeng, Gongfa Li, and Zhaojie Ju. 2021. Learning for a Robot: Deep Reinforcement Learning, Imitation Learning, Transfer Learning. Sensors 21, 4 (February 2021), 1278. https://doi.org/10.3390/s21041278Google ScholarCross Ref
- Karl Weiss, Taghi M. Khoshgoftaar, and DingDing Wang. 2016. A survey of transfer learning. J Big Data 3, 1 (December 2016), 9. https://doi.org/10.1186/s40537-016-0043-6Google ScholarCross Ref
- Zhengxu Yu, Dong Shen, Zhongming Jin, Jianqiang Huang, Deng Cai, and Xian-Sheng Hua. 2020. Progressive Transfer Learning. Retrieved April 12, 2023 from http://arxiv.org/abs/1908.02492Google Scholar
- Safie-eldin Nasr Mohamed, bidaa Mortada Abuel-hassan, Anas M. Ali, Walid Elshafie, Ashraf A. M. Khaalaf, osama zahran, Moawad I. Dessouky, el sayed M. el rabie, and fathi E. Abd ABD El-Samie. 2022. Modulation Format Recognition using CNN-Based Transfer Learning Models. In Review. https://doi.org/10.21203/rs.3.rs-1634005/v1Google ScholarCross Ref
- Cheng Wang, Delei Chen, Lin Hao, Xuebo Liu, Yu Zeng, Jianwei Chen, and Guokai Zhang. 2019. Pulmonary Image Classification Based on Inception-v3 Transfer Learning Model. IEEE Access 7, (2019), 146533–146541. https://doi.org/10.1109/ACCESS.2019.2946000Google ScholarCross Ref
- Zehuan Jing, Peng Li, Bin Wu, Shibo Yuan, and Yingchao Chen. 2022. An Adaptive Focal Loss Function Based on Transfer Learning for Few-Shot Radar Signal Intra-Pulse Modulation Classification. Remote Sensing 14, 8 (April 2022), 1950. https://doi.org/10.3390/rs14081950Google ScholarCross Ref
Index Terms
- Communication Signal Modulation Recognition Based on Inception-V3 Transfer Learning
Recommendations
Discriminative transfer learning with sparsity regularization for single-sample face recognition
Discriminant analysis is an important technique for face recognition because it can extract discriminative features to classify different persons. However, most existing discriminant analysis methods fail to work for single-sample face recognition (SSFR)...
A Transfer Learning Approach to Compound Facial Expression Recognition
ICAIP '20: Proceedings of the 4th International Conference on Advances in Image ProcessingRecent advanced research shows that deep learning has great potential in facial expression recognition, specially, in the basic expression recognition field, many researchers have proposed lots of deep learning networks with excellent performance. As ...
Offline Signature Verification using Deep Learning Convolutional Neural Network (CNN) Architectures GoogLeNet Inception-v1 and Inception-v3
AbstractBiometric systems such as signature verification are highly viable in order to identify individuals in organizations or in finance divisions. Advancement in classification of images using deep learning networks has opened an opportunity for this ...
Comments