Multi-Block Attention for Efficient Channel Estimation in IRS-Assisted mmWave MIMO

2026-05-14

AI Summary

This work addresses the high pilot overhead and cascaded channel estimation challenges in intelligent reflecting surface (IRS)-assisted millimeter-wave MIMO systems, which stem from the IRS’s passive nature and large-scale deployment. To tackle these issues, a deep learning-based multi-block attention (MBA) framework is proposed. The approach integrates optimal DFT/Hadamard phase configurations with a selective IRS deactivation strategy and employs a two-stage network architecture: a convolutional attention network (CAN) recovers spatial correlations, while a complex-valued multi-convolutional network (CMN) suppresses noise to refine features effectively. Compared to conventional least-squares estimators, the proposed method reduces pilot overhead by up to 87% and achieves approximately 51% lower normalized mean square error than state-of-the-art techniques at 10 dB SNR, offering both low computational complexity and strong adaptability to varying environments.

Abstract

Intelligent Reflecting Surfaces (IRSs) are a promising technology for enhancing the spectral and energy efficiency of millimeter-wave (mmWave) multiple-input multiple-output (MIMO) systems. In these systems, accurate channel estimation remains challenging due to the passive nature of IRS elements and the high pilot overhead in large-scale deployments. This paper presents a deep learning-based Multi-Block Attention (MBA) framework for efficient cascaded channel estimation in IRS-assisted mmWave MIMO systems that utilize orthogonal frequency division multiplexing (OFDM). First, we show the optimality of the discrete Fourier transform (DFT) and Hadamard matrices as phase configurations for least squares (LS) estimation. To reduce training overhead, we selectively deactivate IRS elements and compensate for induced feature loss using a two-stage architecture: (i) a Convolutional Attention Network (CAN) for spatial correlation recovery and (ii) a Complex Multi-Convolutional Network (CMN) for noise suppression. The MBA architecture mitigates error propagation through attention-guided feature refinement and denoising. Simulation results indicate that the MBA method reduces pilot overhead by up to 87% compared to the LS estimator. Additionally, at signal-to-noise ratios of 10 dB, our proposed method achieves approximately 51% lower normalized mean squared error (NMSE) than leading methods. It also maintains low computational complexity and adapts effectively to various propagation environments.

Problem

• Intelligent Reflecting Surface
• mmWave MIMO
• Channel Estimation
• Pilot Overhead
• Cascaded Channel

Innovation

• Multi-Block Attention
• Intelligent Reflecting Surface
• Channel Estimation
• mmWave MIMO
• Deep Learning