Towards Semi-supervised Classification of Abnormal Spectrum Signals Based on Deep Learning

JIANG Tao; CHEN Wanqing; ZHOU Hangping; HE Jinyang; QI Peihan

doi:10.23919/cje.2022.00.395

Article Contents

Article Navigation > Chinese Journal of Electronics > 2024 > Uncorrected proof

Tao JIANG, Wanqing CHEN, Hangping ZHOU, et al., “Towards Semi-supervised Classification of Abnormal Spectrum Signals Based on Deep Learning,” Chinese Journal of Electronics, vol. 33, no. 3, pp. 1–11, 2024 doi: 10.23919/cje.2022.00.395

Citation:

Tao JIANG, Wanqing CHEN, Hangping ZHOU, et al., “Towards Semi-supervised Classification of Abnormal Spectrum Signals Based on Deep Learning,” Chinese Journal of Electronics, vol. 33, no. 3, pp. 1–11, 2024 doi: 10.23919/cje.2022.00.395

Citation:

PDF( 8923 KB)

Towards Semi-supervised Classification of Abnormal Spectrum Signals Based on Deep Learning

doi: 10.23919/cje.2022.00.395

JIANG Tao^{1, 2
,},
CHEN Wanqing^1
,,
ZHOU Hangping^1
,,
HE Jinyang^3
,,
QI Peihan^{2
,
,}

1.
School of Cyber Engineering, Xidian University, Xi’an 710126, China
2.
State Key Laboratory of Integrate Service Networks, Xidian University, Xi’an 71007l, China
3.
Guangzhou Institute of Technology, Xidian University, Guangzhou 510555, China

More Information

Author Bio:
Tao JIANG received the Ph.D. degree in cryptology from the School of Telecommunications Engineering, Xidian University, Xi’an, China, in 2016. He is currently an Associate Professor with the School of Cyber Engineering, Xidian University. His research interests include applied cryptography and artificial intelligence security.(Email: taojiang@xidian.edu.cn)

Wanqing CHEN was born in Hubei Province, China, in 1998. She received the B.E. degree in communication engineering from Three Gorges University, Yichang, China, in 2020. She is currently an M.S. candidate in Xidian University. Her main research interests are deep learning, image classification, and wireless spectrum identification. (Email: wqchen@stu.xidian.edu.cn)

Hangping ZHOU was born in Shandong Province, China, in 1995. He received the B.E. degree in software engineering from Shandong Jianzhu University, Jinan, China, in 2019. He is currently an M.S. candidate in Xidian University, Xi’an, China. His main research interests are deep learning, adversarial sample attack and prevention, and artificial intelligence security. (Email: hangpingzhou@stu.xidian.edu.cn)

Jinyang HE was born in Shaanxi Province, China, in 1998. He received the B.E. degree in geological engineering from China University of Mining and Technology, Xuzhou, China, in 2020. He is currently an M.S. candidate in Xidian University, Xi’an, China. His main research interests are deep learning and wireless spectrum anomaly detection. (Email: yanghe3586@outlook.com)

Peihan QI was born in Henan Province, China, in 1986. He received the B.S. degree from Chang’an University, Xi’an, China, in 2006, and the M.S. degree and the Ph.D. degree from Xidian University, Xi’an, in 2011 and 2014, respectively. He is currently a Professor with the School of Telecommunications Engineering, Xidian University. His research interests include compressed sensing, spectrum sensing in cognitive radio networks, and high-speed digital signal processing. (Email: phqi@xidian.edu.cn)
Corresponding author: Email: phqi@xidian.edu.cn
Received Date: 2022-11-18
Accepted Date: 2023-06-16

Available Online: 2023-08-23

Abstract

Abstract

In order to cope with the heavy labor cost challenge of the manual abnormal spectrum classification and improve the effectiveness of existing machine learning schemes for spectral datasets with interference-to-signal ratios, we proposes a semi-supervised classification of abnormal spectrum signals (SSC-ASS), aimed at addressing some of the challenges in abnormal spectrum signal (ASS) classification tasks. A significant advantage of SSC-ASS is that it does not require manual labeling of every abnormal data, but instead achieves high-precision classification of ASSs using only a small number of labeled data. Furthermore, the method can to some extent avoid the introduction of erroneous information resulting from the complex and variable nature of abnormal signals, thereby improving classification accuracy. Specifically, SSC-ASS uses a memory AutoEncoder module to efficiently extract features from abnormal spectrum signals by learning from the reconstruction error. Additionally, SSC-ASS combines convolutional neural network and the K-means using a DeepCluster framework to fully utilize the unlabeled data. Furthermore, SSC-ASS also utilizes pre-training, category mean memory module and replaces pseudo-labels to further improve the classification accuracy of ASSs. And we verify the classification effectiveness of SSC-ASS on synthetic spectrum datasets and real on-air spectrum dataset.
- Abnormal spectrum classification,
- AutoEncoder,
- Semi-supervised,
- DeepCluster

FullText(HTML)

References(38)

References

[1]	H. T. Hayvaci and B. Tavli, “Spectrum sharing in radar and wireless communication systems: A review,” in Proceedings of 2014 International Conference on Electromagnetics in Advanced Applications, Palm Beach, Aruba, pp. 810–813, 2014.
[2]	W. L. Bai, X. H. Zou, P. X. Li, et al., “Photonic millimeter-wave joint radar communication system using spectrum-spreading phase-coding,” IEEE Transactions on Microwave Theory and Techniques, vol. 70, no. 3, pp. 1552–1561, 2022. doi: 10.1109/TMTT.2021.3138069
[3]	P. B. Si, F. R. Yu, R. Z. Yang, et al., “Dynamic spectrum management for heterogeneous UAV networks with navigation data assistance,” in Proceedings of 2015 IEEE Wireless Communications and Networking Conference, New Orleans, LA, USA, pp. 1078–1083, 2015.
[4]	X. Liu, Q. Q. Sun, W. D. Lu, et al., “Big-data-based intelligent spectrum sensing for heterogeneous spectrum communications in 5G,” IEEE Wireless Communications, vol. 27, no. 5, pp. 67–73, 2020. doi: 10.1109/MWC.001.1900493
[5]	K. K. Wong and T. O’Farrell, “Spread spectrum techniques for indoor wireless IR communications,” IEEE Wireless Communications, vol. 10, no. 2, pp. 54–63, 2003. doi: 10.1109/MWC.2003.1196403
[6]	S. L. Zheng, P. H. Qi, S. C. Chen, et al., “Fusion methods for CNN-based automatic modulation classification,” IEEE Access, vol. 7 pp. 66496–66504, 2019. doi: 10.1109/ACCESS.2019.2918136
[7]	P. H. Qi, X. Y. Zhou, S. L. Zheng, et al., “Automatic modulation classification based on deep residual networks with multimodal information,” IEEE Transactions on Cognitive Communications and Networking, vol. 7, no. 1, pp. 21–33, 2021. doi: 10.1109/TCCN.2020.3023145
[8]	P. H. Qi, X. Y. Zhou, Y. L. Ding, et al., “FedBKD: Heterogenous federated learning via bidirectional knowledge distillation for modulation classification in IoT-edge system,” IEEE Journal of Selected Topics in Signal Processing, vol. 17, no. 1, pp. 189–204, 2023. doi: 10.1109/JSTSP.2022.3224597
[9]	K. Y. Yazdandoost and R. Kohno, “Wireless communications for body implanted medical device,” in Proceedings of 2007 Asia-Pacific Microwave Conference, Bangkok, Thailand, pp. 1–4, 2007.
[10]	I. Sohn, Y. H. Jang, and S. H. Lee, “Ultra-low-power implantable medical devices: Optical wireless communication approach,” IEEE Communications Magazine, vol. 58, no. 5, pp. 77–83, 2020. doi: 10.1109/MCOM.001.1900609
[11]	H. Karvonen, M. Hämäläinen, J. Iinatti, et al., “Coexistence of wireless technologies in medical scenarios,” in Proceedings of 2017 European Conference on Networks and Communications, Oulu, Finland, pp. 1–5, 2017.
[12]	M. Tubaishat, P. Zhuang, Q. Qi, et al., “Wireless sensor networks in intelligent transportation systems,” Wireless Communications and Mobile Computing, vol. 9, no. 3, pp. 287–302, 2009. doi: 10.1002/wcm.616
[13]	H. Zhang and X. X. Lu, “Vehicle communication network in intelligent transportation system based on Internet of Things,” Computer Communications, vol. 160 pp. 799–806, 2020. doi: 10.1016/j.comcom.2020.03.041
[14]	H. S. Li and Z. Han, “Catch me if you can: An abnormality detection approach for collaborative spectrum sensing in cognitive radio networks,” IEEE Transactions on Wireless Communications, vol. 9, no. 11, pp. 3554–3565, 2010. doi: 10.1109/TWC.2010.091510.100315
[15]	S. Rajendran, W. Meert, V. Lenders, et al., “Unsupervised wireless spectrum anomaly detection with interpretable features,” IEEE Transactions on Cognitive Communications and Networking, vol. 5, no. 3, pp. 637–647, 2019. doi: 10.1109/TCCN.2019.2911524
[16]	K. Jiang, W. Y. Xie, Y. S. Li, et al., “Semisupervised spectral learning with generative adversarial network for hyperspectral anomaly detection,” IEEE Transactions on Geoscience and Remote Sensing, vol. 58, no. 7, pp. 5224–5236, 2020. doi: 10.1109/TGRS.2020.2975295
[17]	H. E. Egilmez and A. Ortega, “Spectral anomaly detection using graph-based filtering for wireless sensor networks,” in Proceedings of 2014 IEEE International Conference on Acoustics, Speech and Signal Processing, Florence, Italy, pp. 1085–1089, 2014.
[18]	N. Tandiya, A. Jauhar, V. Marojevic, et al., “Deep predictive coding neural network for RF anomaly detection in wireless networks,” in Proceedings of 2018 IEEE International Conference on Communications Workshops, Kansas City, MO, USA, pp. 1–6, 2018.
[19]	B. Jdid, K. Hassan, I. Dayoub, et al., “Machine learning based automatic modulation recognition for wireless communications: A comprehensive survey,” IEEE Access, vol. 9 pp. 57851–57873, 2021. doi: 10.1109/ACCESS.2021.3071801
[20]	R. L. Zhou, F. G. Liu, and C. W. Gravelle, “Deep learning for modulation recognition: A survey with a demonstration,” IEEE Access, vol. 8 pp. 67366–67376, 2020. doi: 10.1109/ACCESS.2020.2986330
[21]	Y. Zeng, M. Zhang, F. Han, et al., “Spectrum analysis and convolutional neural network for automatic modulation recognition,” IEEE Wireless Communications Letters, vol. 8, no. 3, pp. 929–932, 2019. doi: 10.1109/LWC.2019.2900247
[22]	J. Y. Xie, R. Girshick, and A. Farhadi, “Unsupervised deep embedding for clustering analysis,” in Proceedings of the 33rd International Conference on Machine Learning, New York, NY, USA, pp. 478–487, 2016.
[23]	M. Caron, P. Bojanowski, A. Joulin, et al., “Deep clustering for unsupervised learning of visual features,” in Proceedings of the 15th European Conference on Computer Vision, Munich, Germany, pp. 139–156, 2018.
[24]	D. H. Lee, “Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks,” in Proceedings of the Workshop on Challenges in Representation Learning, Atlanta, GE, USA, vol. 3, no. 2, pp. 896-901, 2013.
[25]	D. Berthelot, N. Carlini, I. Goodfellow, et al., “MixMatch: A holistic approach to semi-supervised learning,” in Proceedings of the 33rd International Conference on Neural Information Processing Systems, Vancouver, BC, Canada, article no. 454, 2019.
[26]	A. Tarvainen and H. Valpola, “Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results,” in Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, USA, pp. 1195–1204, 2017.
[27]	T. Miyato, S. I. Maeda, M. Koyama, et al., “Virtual adversarial training: A regularization method for supervised and semi-supervised learning,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 41, no. 8, pp. 1979–1993, 2019. doi: 10.1109/TPAMI.2018.2858821
[28]	S. Laine and T. Aila, “Temporal ensembling for semi-supervised learning,” ArXiv e-Print, arXiv:1610.02242, 2016.
[29]	L. X. Chen, W. T. Ruan, X. Y. Liu, et al., “SeqVAT: Virtual adversarial training for semi-supervised sequence labeling,” in Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, Online, pp. 8801–8811, 2020.
[30]	Y. L. Ou, Y. Xue, Y. Yuan, et al., “Semi-supervised cervical dysplasia classification with learnable graph convolutional network,” in Proceedings of 2020 IEEE 17th International Symposium on Biomedical Imaging, Iowa City, IA, USA, pp. 1720–1724, 2020.
[31]	K. Sohn, D. Berthelot, C. L. Li, et al., “FixMatch: Simplifying semi-supervised learning with consistency and confidence,” in Proceedings of the 34th International Conference on Neural Information Processing Systems, Vancouver, BC, Canada, vol. 33, article no. 51, 2020.
[32]	K. H. Wang, C. H. Y. Yang, and M. Betke, “Consistency regularization with high-dimensional non-adversarial source-guided perturbation for unsupervised domain adaptation in segmentation,” in Proceedings of the 35th AAAI Conference on Artificial Intelligence, Online, pp. 10138–10146, 2021.
[33]	Z. J. Hu, Z. Y. Yang, X. F. Hu, et al., “SimPLE: Similar pseudo label exploitation for semi-supervised classification,” in Proceedings of 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville, TN, USA, pp. 15099–15108, 2021.
[34]	Y. Grandvalet and Y. Bengio, “Semi-supervised learning by entropy minimization,” in Proceedings of the 17th International Conference on Neural Information Processing Systems, Vancouver, British Columbia, Canada, pp. 529–536, 2004.
[35]	D. Berthelot, N. Carlini, E. D. Cubuk, et al., “ReMixMatch: Semi-supervised learning with distribution matching and augmentation anchoring,” in Proceedings of the 8th International Conference on Learning Representations, Addis Ababa, Ethiopia, 2020.
[36]	B. W. Zhang, Y. D. Wang, W. X. Hou, et al., “FlexMatch: Boosting semi-supervised learning with curriculum pseudo labeling,” in Proceedings of the 35th International Conference on Neural Information Processing Systems, Online, pp. 18408–18419, 2021.
[37]	S. Pourahmad, A. Basirat, A. Rahimi, et al., “Does determination of initial cluster centroids improve the performance of K-means clustering algorithm? Comparison of three hybrid methods by genetic algorithm, minimum spanning tree, and hierarchical clustering in an applied study,” Computational and Mathematical Methods in Medicine, vol. 2020, article no. 7636857, 2020. doi: 10.1155/2020/7636857
[38]	S. Ioffe and C. Szegedy, “Batch normalization: Accelerating deep network training by reducing internal covariate shift,” in Proceedings of the 32nd International Conference on Machine Learning, Lille, France, pp. 448–456, 2015.