CSI Prediction Using Diffusion Models
IEEE Transactions on Wireless Communications · 2026 · 25 · 21418-21432
DOI: 10.1109/twc.2026.3719116 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Wireless Communications, 2026. Verfasst von Mehdi Sattari und 3 weiteren Autor:innen. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: SEMANTIC Project, European Union (EU) Horizon 2020 Research and Innovation Program through the Marie Skłodowska-Curie, Hexa-X-II Project funded by the Smart Networks and Services Joint Undertaking through EU’s Horizon Europe Research and Innovation Program.
Abstract (Originalsprache)
Acquiring accurate channel state information (CSI) is critical for reliable and efficient wireless communication, but challenges such as high pilot overhead and channel aging hinder timely and accurate CSI acquisition. CSI prediction, which forecasts future CSI from historical observations, offers a promising solution. Recent deep learning approaches, including recurrent neural networks and Transformers, have achieved notable success but typically learn deterministic mappings, limiting their ability to capture the stochastic and multimodal nature of wireless channels. In this paper, we introduce a novel probabilistic framework for CSI prediction based on diffusion models, offering a flexible design that supports integration of diverse prediction schemes. We decompose the CSI prediction task into two components: a temporal encoder, which extracts channel dynamics, and a diffusion-based generator, which produces future CSI samples. We investigate two inference schemes—autoregressive and sequence-to-sequence—and explore multiple diffusion backbones, including U-Net and Transformer-based architectures. Furthermore, we examine a diffusion-based approach without an explicit temporal encoder and utilize the DDIM scheduling to reduce model complexity. Extensive simulations demonstrate that our diffusion-based models significantly outperform state-of-the-art baselines.
Förderung
- SEMANTIC Project
- European Union (EU) Horizon 2020 Research and Innovation Program through the Marie Skłodowska-Curie
- Hexa-X-II Project funded by the Smart Networks and Services Joint Undertaking through EU’s Horizon Europe Research and Innovation Program
- Swedish Research Council
- Wallenberg Artificial Intelligence (AI), Autonomous Systems and Software Program
- Knut och Alice Wallenbergs Stiftelse
- Swedish Innovation Agency
- National Academic Infrastructure for Supercomputing in Sweden
Zitieren
APA
Sattari, M., Aliakbari, J., Amat, A. G. I., & Svensson, T. (2026). CSI Prediction Using Diffusion Models. IEEE Transactions on Wireless Communications, 25, 21418-21432. https://doi.org/10.1109/twc.2026.3719116
Vancouver
Sattari M, Aliakbari J, Amat AGI, Svensson T. CSI Prediction Using Diffusion Models. IEEE Transactions on Wireless Communications. 2026;25:21418-21432. doi:10.1109/twc.2026.3719116
BibTeX
@article{Sattari2026,
author = {Sattari, Mehdi and Aliakbari, Javad and Amat, Alexandre Graell i and Svensson, Tommy},
title = {CSI Prediction Using Diffusion Models},
journal = {IEEE Transactions on Wireless Communications},
year = {2026},
volume = {25},
pages = {21418--21432},
doi = {10.1109/twc.2026.3719116}
}
EUVIMED