Cross-Band Channel Impulse Response Prediction: Leveraging 3.5 GHz Channels for Upper Mid-Band

2025-10-31

AI Summary

To address the challenges of high penetration loss and costly field measurements in channel modeling for 6G high-frequency bands (FR3, 7–24 GHz), this paper proposes a cross-band channel impulse response (CIR) prediction method that accurately infers 7 GHz CIR from readily available 3.5 GHz low-band channel data. We introduce CIR-UNext, a novel deep learning framework integrating attention mechanisms and auxiliary-task learning (AU-Net-Aux), and extend it into Channel2ComMap—a general-purpose foundation model supporting MIMO-OFDM throughput prediction. Leveraging ray-tracing data augmentation and joint gain-phase modeling, our approach achieves a median gain error of 0.58 dB and phase error of 0.27 rad under complex propagation scenarios—substantially outperforming state-of-the-art methods. Furthermore, Channel2ComMap demonstrates strong generalization capabilities in beam management, localization, and radio resource scheduling.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Accurate cross-band channel prediction is essential for 6G networks, particularly in the upper mid-band (FR3, 7--24 GHz), where penetration loss and blockage are severe. Although ray tracing (RT) provides high-fidelity modeling, it remains computationally intensive, and high-frequency data acquisition is costly. To address these challenges, we propose CIR-UNext, a deep learning framework designed to predict 7 GHz channel impulse responses (CIRs) by leveraging abundant 3.5 GHz CIRs. The framework integrates an RT-based dataset pipeline with attention U-Net (AU-Net) variants for gain and phase prediction. The proposed AU-Net-Aux model achieves a median gain error of 0.58 dB and a phase prediction error of 0.27 rad on unseen complex environments. Furthermore, we extend CIR-UNext into a foundation model, Channel2ComMap, for throughput prediction in MIMO-OFDM systems, demonstrating superior performance compared with existing approaches. Overall, CIR-UNext provides an efficient and scalable solution for cross-band prediction, enabling applications such as localization, beam management, digital twins, and intelligent resource allocation in 6G networks.

Problem

• Predicting 7 GHz channels using 3.5 GHz data
• Reducing computational cost of ray tracing methods
• Enabling efficient cross-band prediction for 6G networks

Innovation

• Deep learning predicts 7 GHz channels from 3.5 GHz data
• Attention U-Net variants model gain and phase accurately
• Framework extends to throughput prediction in MIMO-OFDM systems