Mansi Patel
ABSTRACT
Deep learning has great potential for automatic modulation classification (AMC). However, its performance largely hinges upon the availability of sufficient high-quality labeled data. In this paper, we propose data augmentation with conditional generative adversarial network (CGAN) for convolutional neural network (CNN) based AMC, which provides an effective solution to the limited data problem. We present the design of the proposed CGAN based data augmentation method, and validate its performance with a public dataset. The experiment results show that CNN-based modulation classification can greatly benefit from the proposed data augmentation approach with greatly improved accuracy.
- Muhammad Amjad, Mubashir Husain Rehmani, and Shiwen Mao. 2018. Wireless multimedia cognitive radio networks: A comprehensive survey. IEEE Communications Surveys and Tutorials 20, 2 (Second Quarter 2018), 1056--1103.Google Scholar
Cross Ref
- Timothy J O'Shea, Johnathan Corgan, and T Charles Clancy. 2016. Convolutional radio modulation recognition networks. In Proc. 2016 Int. Conf. Engineering Appl. Neural Netw. Springer, Aberdeen, Scotland, 213--226.Google ScholarCross Ref
- Yun Lin, Jicheng Jia, Sen Wang, Bin Ge, and Shiwen Mao. 2020. Wireless device identification based on radio frequency fingerprint features. In Proc. IEEE ICC 2019. IEEE, Dublin, Ireland, 1--6.Google ScholarCross Ref
- Yun Lin, Haojun Zhao, Ya Tu, Shiwen Mao, and Zheng Dou. 2020. Threats of adversarial attacks in DNN-based modulation recognition. In Proc. IEEE TNFOCOM 2019. IEEE, Toronto, Canada, 1--10.Google ScholarDigital Library
- Kemal Davaslioglu and Yalin E Sagduyu. 2019. Trojan attacks on wireless signal classification with adversarial machine learning. In Proc. DySPAN 2019. IEEE, Newark, NJ, 1--6.Google ScholarDigital Library
- Jefferson L Xu, Wei Su, and Mengchu Zhou. 2010. Likelihood-ratio approaches to automatic modulation classification. IEEE Trans. Syst., Man, Cybern. Syst., Part C (Appl. Rev.) 41, 4 (2010), 455--469.Google ScholarDigital Library
- Domenico Grimaldi, Sergio Rapuano, and Luca De Vito. 2007. An automatic digital modulation classifier for measurement on telecommunication networks. IEEE Trans. Instrum. Meas. 56, 5 (2007), 1711--1720.Google ScholarCross Ref
- Gihan J Mendis, Jin Wei, and Arjuna Madanayake. 2016. Deep learning-based automated modulation classification for cognitive radio. In Proc. IEEE ICCS 2016. IEEE, Shenzhen, China, 1--6.Google ScholarCross Ref
- Yi Shi et al. 2019. Deep learning for RF signal classification in unknown and dynamic spectrum environments. In Proc. IEEE DySPAN 2019. IEEE, Newark, NJ,Google Scholar
- Guanhong Tao et al. 2019. Sequential convolutional recurrent neural networks for fast automatic modulation classification, https://arxiv.org/abs/1909.03050Google Scholar
- Shilian Zheng, Peihan Qi, Shichuan Chen, and Xiaoniu Yang. 2019. Fusion methods for CNN-based automatic modulation classification. IEEE Access J. 7 (2019), 66496--66504.Google ScholarCross Ref
- Yaohua Sun, Mugen Peng, Yangcheng Zhou, Yuzhe Huang, and Shiwen Mao. 2019. Application of machine learning in wireless networks: Key technologies and open issues. IEEE Communications Surveys and Tutorials 21, 4 (Fourth Quarter 2019), 3072--3108.Google ScholarDigital Library
- Liang Huang, Weijian Pan, You Zhang, LiPing Qian, Nan Gao, and Yuan Wu. 2019. Data augmentation for deep learning-based radio modulation classification. IEEE Access J. 8 (Dec. 2019), 1498--1506.Google Scholar
- Ian Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde-Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2014. Generative adversarial nets. In Proc. NIPS 2014. MIT Press, Montréal, Canada, 2672--2680.Google ScholarDigital Library
- Tugba Erpek, Yalin E. Sagduyu, and Yi Shi. 2019. Deep learning for launching and mitigating wireless jamming attacks. IEEE Trans. Cognitive Commun. Netw. 5, 1 (Mar. 2019), 2--14.Google Scholar
- Kemal Davaslioglu and Yalin E Sagduyu. 2018. Generative adversarial learning for spectrum sensing. In Proc. IEEE ICC 2018. IEEE, Kansas City, MO, 1--6.Google ScholarCross Ref
- Francesco Restuccia, Salvatore D'Oro, Amani Al-Shawabka, Bruno Costa Rendon, Kaushik Chowdhury, Stratis Ioannidis, and Tommaso Melodia. 2020. Hacking the waveform: Generalized wireless adversarial deep learning, https://arxiv.org/abs/2005.02270Google Scholar
- Tamoghna Roy, Tim O'Shea, and Nathan West. 2019. Generative adversarial radio spectrum networks. In Proc. 2019 ACM Workshop on Wireless Security and Machine Learning. ACM, Miami, FL, 12--15.Google ScholarDigital Library
- Yi Shi, Kemal Davaslioglu, and Yalin E Sagduyu. 2019. Generative adversarial network for wireless signal spoofing. In Proc. 2019 ACM Workshop on Wireless Security and Machine Learning. ACM, Miami, FL, 55--60.Google ScholarDigital Library
- Bin Tang, Ya Tu, Zhaoyue Zhang, and Yun Lin. 2018. Digital signal modulation classification with data augmentation using generative adversarial nets in cognitive radio networks. IEEE Access J. 6 (2018), 15713--15722.Google ScholarCross Ref
- Mehdi Mirza and Simon Osindero. 2014. Conditional generative adversarial nets. https://arxiv.org/abs/1411.1784Google Scholar
Index Terms
- Data augmentation with conditional GAN for automatic modulation classification
Recommendations
Probability density function based data augmentation for deep neural network automatic modulation classification with limited training data
AbstractDeep neural networks (DNN) based automatic modulation classification (AMC) has achieved high accuracy performance. However, DNNs are data‐hungry models, and training such a model requires a large volume of data. Insufficient training data will ...
This manuscript discusses the question of how to expand the data from the limited dataset if there is insufficient training data: the required minimum sampling size for data augmentation and the way of data expansion (waveform probability density function‐...
Phase sensitive masking-based single channel speech enhancement using conditional generative adversarial network
AbstractWe propose PSMGAN, an efficient phase sensitive masking-based single-channel speech enhancement technique using a conditional generative adversarial network (cGAN). The time–frequency (T-F) masking-based speech enhancement approaches ...
A novel proposed CNN–SVM architecture for ECG scalograms classification
AbstractNowadays, the number of sudden deaths due to heart disease is increasing with the coronavirus pandemic. Therefore, automatic classification of electrocardiogram (ECG) signals is crucial for diagnosis and treatment. Thanks to deep learning ...
Comments