TimeGAN and Coordinated Attention Prototype Network Based Prediction Model for Infrasound Signal

Downloads

Authors

  • Quanbo Lu College of Communication Engineering, Chongqing Polytechnic University of Electronic Technology, China
  • Xiaojuan Huang School of Mechanical Engineering, Chongqing Three Gorges University, China
  • Rao Li College of Communication Engineering, Chongqing Polytechnic University of Electronic Technology, China
  • Mei Li School of Information Engineering, China University of Geosciences, China
  • Dong Zhu Sevnce Robotics Co, Ltd, China

Abstract

Due to the complexity of the infrasound environment and the high costs associated with data collection, frequent acquisition of infrasound data is often impractical, resulting in a limited amount of labeled data. To address the challenge of low classification prediction accuracy caused by data scarcity, this paper proposes an infrasound prediction model based on a time-series generative adversarial network (TimeGAN) and coordinated attention prototype network (CAPN) (TimeGAN-CAPN). The model begins by introducing TimeGAN, where the generative network is trained using a combination of unsupervised and supervised learning. This approach enables the network to operate within the latent space of temporal features and generate time-series data that closely aligns with the distribution of the original data. These generated samples are then combined with the original data to form an augmented dataset. Subsequently, the augmented data is input into the CAPN, which enhances the sample size per class, allowing for more precise class prototypes and improving the prediction accuracy of the model. Furthermore, the quality and diversity of the data generated by TimeGAN are quantitatively and qualitatively assessed using maximum mean discrepancy (MMD) and t-istributed stochastic neighbor embedding (t-SNE), facilitating a comparison and verification of the generated data’s performance. Experimental results show that TimeGAN-CAPN significantly outperforms the CAPN model in classification tasks with limited infrasound data, achieving an increase in accuracy of 7.15 %. This demonstrates that the proposed method is highly effective for predicting infrasound-related disasters, particularly in scenarios with small sample sizes.

Keywords:

infrasound signal, time-series generative adversarial network, coordinated attention prototype network, maximum mean discrepancy

References


  1. Baeza Moyano D., Gonzalez Lezcano R.A. (2022), Effects of infrasound on health: Looking for improvements in housing conditions, International Journal of Occupational Safety and Ergonomics, 28(2): 809–823, https://doi.org/10.1080/10803548.2020.1831787

  2. Dong H. et al. (2024), Enhanced infrasound denoising for debris flow analysis: Integrating empirical mode decomposition with an improved wavelet threshold algorithm, Measurement, 235: 114961, https://doi.org/10.1016/j.measurement.2024.114961

  3. Friedrich B., Joost H., Fedtke T., Verhey J.L. (2023), Effects of infrasound on the perception of a low-frequency sound, Acta Acustica, 7: 60, https://doi.org/10.1051/aacus/2023061

  4. Hou Q., Zhou D., Feng J. (2021), Coordinate attention for efficient mobile network design, [in:] 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 13713–13722, https://doi.org/10.1109/CVPR46437.2021.01350

  5. Hupe P., Ceranna L., Le Pichon A., Matoza R.S., Mialle P. (2022), International monitoring system infrasound data products for atmospheric studies and civilian applications, Earth System Science Data Discussions, 14(9): 4201–4230, https://doi.org/10.5194/essd-14-4201-2022

  6. Ji Z., Liu X., Pang Y., Ouyang W., Li X. (2020), Few-shot human-object interaction recognition with semantic-guided attentive prototypes network, IEEE Transactions on Image Processing, 30: 1648–1661, https://doi.org/10.1109/TIP.2020.3046861

  7. Jiang Y.H. et al. (2025), Recursive prototypical network with coordinate attention: A model for few-shot cross-condition bearing fault diagnosis, Applied Acoustics, 231: 110442, https://doi.org/10.1016/j.apacoust.2024.110442

  8. Listowski C. et al. (2022), Remote monitoring of Mediterranean hurricanes using infrasound, Remote Sensing, 14(23): 6162, https://doi.org/10.3390/rs14236162

  9. Lu Q.B., Li M. (2023), VMD and CNN-based classification model for infrasound signal, Archives of Acoustics, 48(3): 403–412, https://doi.org/10.24425/aoa.2023.145247

  10. Macpherson K.A., Fee D., Colwell J.R., Witsil A.J. (2023), Using local infrasound to estimate seismic velocity and earthquake magnitudes, Bulletin of the Seismological Society of America, 113(4): 1434–1456, https://doi.org/10.1785/0120220237

  11. Mitropoulos S., Toulas V., Douligeris C. (2022), A prototype network monitoring information system: modelling, design, implementation and evaluation, International Journal of Information and Communication Technology, 21(2): 111–136, https://doi.org/10.1504/IJICT.2022.124807

  12. Ruddick K.G. et al. (2024), WATERHYPERNET: A prototype network of automated in situ measurements of hyperspectral water reflectance for satellite validation and water quality monitoring, Frontiers in Remote Sensing, 5: 1347520, https://doi.org/10.3389/frsen.2024.1347520

  13. Sehar U., Xiong J., Xia Z. (2025), Automatic tooth labeling after segmentation using prototype-based meta-learning, Machine Intelligence Research, 22: 539–552, https://doi.org/10.1007/s11633-024-1520-6

  14. Sharma S.K., Alenizi A., Kumar M., Alfarraj O., Alowaidi M. (2024), Detection of real-time deep fakes and face forgery in video conferencing employing generative adversarial networks, Heliyon, 10(17): e37163, https://doi.org/10.1016/j.heliyon.2024.e37163

  15. Sovilla B. et al. (2025), The dominant source mechanism of infrasound generation in powder snow avalanches, Geophysical Research Letters, 52(2): e2024GL112886, https://doi.org/10.1029/2024GL112886

  16. Tang T. et al. (2023), An improved prototypical network with L2 prototype correction for few-shot cross-domain fault diagnosis, Measurement, 217: 113065, https://doi.org/10.1016/j.measurement.2023.113065

  17. Vuletić M., Prenzel F., Cucuringu M. (2024), Fin-GAN: Forecasting and classifying financial time series via generative adversarial networks, Quantitative Finance, 24(2): 175–199, https://doi.org/10.1080/14697688.2023.2299466

  18. Wang W. et al. (2021), Rethinking maximum mean discrepancy for visual domain adaptation, IEEE Transactions on Neural Networks and Learning Systems, 34(1): 264–277, https://doi.org/10.1109/TNNLS.2021.3093468

  19. Watson L.M. et al. (2022), Volcano infrasound: Progress and future directions, Bulletin of Volcanology, 84: 44, https://doi.org/10.1007/s00445-022-01544-w

  20. Wilson T.C., Petrin C.E., Elbing B.R. (2023), Infrasound and low-audible acoustic detections from a long-term microphone array deployment in Oklahoma, Remote Sensing, 15(5): 1455, https://doi.org/10.3390/rs15051455

  21. Yang S. et al. (2025), Correlation between and mechanisms of gas desorption and infrasound signals, Natural Resources Research, 34: 515–537, https://doi.org/10.1007/s11053-024-10417-2

  22. Yoon J., Jarrett D., van der Schaar M. (2019), Time-series generative adversarial networks, [in:] Advances in Neural Information Processing Systems 32 (NeurIPS 2019), 32.

  23. Zajamsek B., Hansen K.L., Nguyen P.D., Lechat B., Micic G., Catcheside P. (2023), Effect of infrasound on the detectability of amplitude-modulated tonal noise, Applied Acoustics, 207: 109361, https://doi.org/10.1016/j.apacoust.2023.109361

  24. Zhang B., Xu M., Zhang Y., Ye S., Chen Y. (2024), Attention-ProNet: A prototype network with hybrid attention mechanisms applied to zero calibration in rapid serial visual presentation-based brain–computer interface, Bioengineering, 11(4): 347, https://doi.org/10.3390/bioengineering11040347