Deep Learning-based mmWave MIMO Channel Estimation using sub-6 GHz Channel Information: CNN and UNet Approaches

2025-06-13

AI Summary

Millimeter-wave (mmWave) MIMO channel estimation suffers from low accuracy under low signal-to-noise ratio (SNR) conditions. Method: This paper proposes a novel paradigm leveraging out-of-band sub-6 GHz channel information to assist mmWave channel estimation. It introduces heterogeneous-band knowledge transfer into an end-to-end deep learning framework for the first time, designing two lightweight architectures—CNN and U-Net—to model the nonlinear mapping between sub-6 GHz and mmWave channels. Cross-band feature alignment and joint supervised training enable efficient channel state information (CSI) reconstruction. Contribution/Results: Simulation results demonstrate that the proposed method significantly improves spectral efficiency, outperforming both standalone mmWave deep learning approaches and existing cross-frequency assistance schemes. It provides a scalable, low-overhead solution for dual-band cooperative communication, establishing a new direction for practical mmWave channel estimation.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Multi-modal Vision
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Future wireless multiple-input multiple-output (MIMO) systems will integrate both sub-6 GHz and millimeter wave (mmWave) frequency bands to meet the growing demands for high data rates. MIMO link establishment typically requires accurate channel estimation, which is particularly challenging at mmWave frequencies due to the low signal-to-noise ratio (SNR). In this paper, we propose two novel deep learning-based methods for estimating mmWave MIMO channels by leveraging out-of-band information from the sub-6 GHz band. The first method employs a convolutional neural network (CNN), while the second method utilizes a UNet architecture. We compare these proposed methods against deep-learning methods that rely solely on in-band information and with other state-of-the-art out-of-band aided methods. Simulation results show that our proposed out-of-band aided deep-learning methods outperform existing alternatives in terms of achievable spectral efficiency.

Problem

• Estimating mmWave MIMO channels using sub-6 GHz data
• Overcoming low SNR challenges in mmWave frequencies
• Comparing CNN and UNet for channel estimation

Innovation

• CNN for mmWave MIMO channel estimation
• UNet architecture for channel estimation
• Leveraging sub-6 GHz out-of-band information