EUVIMEDEUVIMED entdecken
Crossref

A Hybrid Deep Learning Framework for IoT Traffic Generation Based on Channel–Spatial Residual Attention and VAE-WGAN

Xuanrui Xiong, Xingyou Guo, Yu Wu, Wei Peng, Li Zhou

IEEE Internet of Things Journal · 01.06.2026 · 13 (11) · 24575-24589

DOI: 10.1109/jiot.2026.3673327 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Internet of Things Journal, erschienen am 01.06.2026. Verfasst von Xuanrui Xiong und 4 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Science and Technology Research Program of Chongqing Municipal Education Commission.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Data paucity in Internet of Things (IoT) attack/anomaly traffic samples inhibits accurate threat identification, limiting defensive modeling capabilities. Existing generative models, such as WGAN-GP and CTGAN, often fail to preserve fine-grained temporal dynamics and multidimensional feature correlations in packet-level traffic, leading to low fidelity in synthesized data and poor generalization in downstream tasks. To address these limitations, this article proposes VWRAM, a novel traffic generation framework combining a variational autoencoder (VAE) with a Wasserstein generative adversarial network (WGAN) and a residual attention mechanism. It first utilizes Gramian angular summation field (GASF) to transform 1-D time-series network traffic data into 2-D images, and then performs gamma correction using a power–law expression to enhance image quality. The framework employs a hybrid architecture combining VAE and WGAN to synthesize network traffic feature maps. To enhance the extraction of channel and spatial characteristics, a bidirectional multiscale channel fusion residual module and a dual attention mechanism (accounting for both channel and spatial dimensions) are incorporated into the generator of the GAN. This design effectively bridges the gap in modeling complex, multiscale IoT traffic patterns with high structural and statistical fidelity. Finally, theoretical analysis and experimental results on the MedBIoT dataset demonstrate that the proposed framework exhibits significant advantages over representative generative models. Specifically, VWRAM achieves an anomaly detection accuracy of 93%, outperforming state-of-the-art models like SyNIG and CTGAN by 6% and 15%, respectively. Furthermore, it demonstrates superior distributional fidelity, achieving the lowest Jensen–Shannon Divergence (JSD) of 0.0918 in key traffic features.

Förderung

  • National Natural Science Foundation of China
  • Science and Technology Research Program of Chongqing Municipal Education Commission

Zitieren

APA

Xiong, X., Guo, X., Wu, Y., Peng, W., & Zhou, L. (2026). A Hybrid Deep Learning Framework for IoT Traffic Generation Based on Channel–Spatial Residual Attention and VAE-WGAN. IEEE Internet of Things Journal, 13(11), 24575-24589. https://doi.org/10.1109/jiot.2026.3673327

Vancouver

Xiong X, Guo X, Wu Y, Peng W, Zhou L. A Hybrid Deep Learning Framework for IoT Traffic Generation Based on Channel–Spatial Residual Attention and VAE-WGAN. IEEE Internet of Things Journal. 2026;13(11):24575-24589. doi:10.1109/jiot.2026.3673327

BibTeX
@article{Xiong2026,
  author = {Xiong, Xuanrui and Guo, Xingyou and Wu, Yu and Peng, Wei and Zhou, Li},
  title = {A Hybrid Deep Learning Framework for IoT Traffic Generation Based on Channel–Spatial Residual Attention and VAE-WGAN},
  journal = {IEEE Internet of Things Journal},
  year = {2026},
  volume = {13},
  number = {11},
  pages = {24575--24589},
  doi = {10.1109/jiot.2026.3673327}
}
Kostenlos starten →