High-Accuracy Predictive Channel Modeling for 6G Wireless Communications With an Improved Diffusion-Driven Learning Framework

Высокоточное прогнозное моделирование каналов для беспроводной связи 6G с использованием усовершенствованной структуры обучения на основе диффузии
Cheng‐Xiang Wang, El‐Hadi M. Aggoune, Chen Huang, Tong Wu, Junling Li, Xiaoyu Chen, Mingchuan Yao
2026-01-01

6G wireless communicationschannel state informationdiffusion-driven learningpredictive channel modelingspace-time-frequency diffusion network
To address sparse channel measurement data and inadequate predictive capabilities in conventional channel models, predictive channel modeling employs joint generative and predictive architectures to enhance robustness. In this paper, we propose an enhanced diffusion-driven predictive framework that integrates generative augmentation and prior-aware prediction into a unified learning pipeline. We first introduce a space-time-frequency (STF) coupled diffusion network based on transformers that generates synthetic channel data preserving critical channel statistical properties. Additionally, we compress measured channel state information into a low-dimensional manifold via a latent encoder and introduce an innovative composite training scheme that couples diffusion-driven prior generation with prediction, equipping the predictive module with rich latent features that lift its performance ceiling and markedly improve generalization across diverse scenarios. Extensive experiments confirm the superiority of our algorithm, and its performance is further validated using channel measurement data, thereby demonstrating its robustness for advanced wireless communications in real-world deployment scenarios.
1
A composite training scheme couples diffusion-based prior generation with prediction, supplying richer latent features and improving generalization across diverse scenarios.
2
A transformer-based space-time-frequency coupled diffusion network generates synthetic channel data while preserving critical channel statistical properties.
3
An enhanced diffusion-driven predictive framework unifies generative augmentation and prior-aware prediction to improve wireless channel modeling from sparse measurements.
4
Experiments and real channel measurement validation demonstrate superior performance and robustness for advanced wireless communication deployment.
5
Measured channel state information is compressed into a low-dimensional manifold using a latent encoder, providing compact representations for prediction.

6G wireless communication channels and their measured channel state information across diverse propagation scenarios

High-accuracy predictive channel modeling, including channel-data generation, latent-feature representation, prediction performance, and cross-scenario generalization

Publication Details
Publication Date
2026-01-01
Journal
Publisher
ISSN
Cited by
4
Access Type
Author Information
Authors
Cheng‐Xiang Wang
El‐Hadi M. Aggoune
Chen Huang
Tong Wu
Junling Li
Xiaoyu Chen
Mingchuan Yao
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%