EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

CSI-4CAST : A Hybrid Deep Learning Model for CSI Prediction With Comprehensive Robustness and Generalization Testing

Sikai Cheng, Reza Zandehshahvar, Haoruo Zhao, Daniel A. Garcia-Ulloa, Alejandro Villena-Rodriguez, Carles Navarro Manchón, Pascal Van Hentenryck

IEEE Transactions on Machine Learning in Communications and Networking · 2026 · 4 · 882-912

DOI: 10.1109/tmlcn.2026.3694830 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Machine Learning in Communications and Networking, 2026. Verfasst von Sikai Cheng und 6 weiteren Autor:innen. Die Arbeit nennt 75 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NSF, Research Gift from Keysight Technologies.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Channel state information (CSI) prediction is a promising strategy for ensuring reliable and efficient operation of massive multiple-input multiple-output (mMIMO) systems by providing timely downlink (DL) CSI. While deep learning–based methods have advanced beyond conventional model-driven and statistical approaches, they remain limited in robustness to practical non-Gaussian noise, generalization across diverse channel conditions, and computational efficiency. This paper introduces CSI-4CAST, a hybrid deep learning architecture that integrates 4 key components, i.e., Convolutional neural network residuals, Adaptive correction layers, ShuffleNet blocks, and Transformers, to efficiently capture both local and long-range dependencies in CSI prediction. To enable rigorous evaluation, this work further presents a comprehensive benchmark, CSI-RRG for Regular, Robustness and Generalization testing, which includes more than 300,000 samples across 3,060 realistic scenarios for both TDD and FDD systems. The dataset spans multiple channel models, a wide range of delay spreads and user velocities, and diverse noise types and intensity degrees. Experimental results show that CSI-4CAST achieves superior prediction accuracy with substantially lower computational cost, outperforming baselines in 81.5% of TDD scenarios and 44.4% of FDD scenarios—the best performance among all evaluated models—while reducing FLOPs by ${5}\times $ and ${3}\times $ compared to LLM4CP, the strongest baseline. In addition, evaluation over CSI-RRG provides valuable insights into how different channel factors affect the performance and generalization capability of deep learning models. Both the dataset (Hugging Face) and evaluation protocols (GitHub) are publicly released to establish a standardized benchmark and to encourage further research on robust and efficient CSI prediction.

Förderung

  • NSF
  • Research Gift from Keysight Technologies

Zitieren

APA

Cheng, S., Zandehshahvar, R., Zhao, H., Garcia-Ulloa, D. A., Villena-Rodriguez, A., Manchón, C. N., & Hentenryck, P. V. (2026). CSI-4CAST : A Hybrid Deep Learning Model for CSI Prediction With Comprehensive Robustness and Generalization Testing. IEEE Transactions on Machine Learning in Communications and Networking, 4, 882-912. https://doi.org/10.1109/tmlcn.2026.3694830

Vancouver

Cheng S, Zandehshahvar R, Zhao H, Garcia-Ulloa DA, Villena-Rodriguez A, Manchón CN, et al. CSI-4CAST : A Hybrid Deep Learning Model for CSI Prediction With Comprehensive Robustness and Generalization Testing. IEEE Transactions on Machine Learning in Communications and Networking. 2026;4:882-912. doi:10.1109/tmlcn.2026.3694830

BibTeX
@article{Cheng2026,
  author = {Cheng, Sikai and Zandehshahvar, Reza and Zhao, Haoruo and Garcia-Ulloa, Daniel A. and Villena-Rodriguez, Alejandro and Manchón, Carles Navarro and Hentenryck, Pascal Van},
  title = {CSI-4CAST : A Hybrid Deep Learning Model for CSI Prediction With Comprehensive Robustness and Generalization Testing},
  journal = {IEEE Transactions on Machine Learning in Communications and Networking},
  year = {2026},
  volume = {4},
  pages = {882--912},
  doi = {10.1109/tmlcn.2026.3694830}
}
Kostenlos starten →