Recurrent Transformer-Based Near- and Far-Field THz Wideband Channel Estimation for UM-MIMO

2026-05-12

AI Summary

In 6G integrated terahertz and extremely large-scale MIMO systems, the high carrier frequency and large antenna aperture cause users to simultaneously reside in both near-field and far-field regions, rendering conventional hybrid beamforming architectures ineffective for accurate channel estimation in such mixed-field scenarios. To address this challenge, this work proposes, for the first time, a block-recurrent Transformer model with state memory that iteratively processes wideband terahertz channels through a single Transformer block, enabling generalization across varying scatterer distances, multipath counts, and bandwidth conditions. The proposed method supports one-time training followed by multiple iterative inferences, achieving approximately 5 dB and 7.5 dB improvements in normalized mean square error (NMSE) over the current state-of-the-art in narrowband and wideband settings, respectively.

Abstract

The integration of terahertz communications and ultra-massive multiple-input multiple-output (UM-MIMO) systems in 6G networks is motivated by their ability to enable unprecedented data rates, mitigate spectrum congestion, and enhance overall network performance. However, the enlarged antenna apertures and higher carrier frequencies in these systems increase the Rayleigh distance, causing users to span both the near-field and conventional far-field regions. Accurate spatial precoding thus requires exact channel estimation at the base station - a task made more challenging by the hybrid coexistence of near- and far-field effects and the limited number of digital chains available in hybrid beamforming architectures. In this paper, we propose a block recurrent transformer model to address this challenge. We demonstrate that a single transformer block equipped with state memory can be trained once and then iteratively applied for hybrid-field channel estimation. Furthermore, we train the model such that it generalizes to wireless channels with varying scatterer distances, different numbers of propagation paths, and wideband operation. Simulation results show that the proposed method achieves performance gains of approximately 5 dB and 7.5 dB in normalized mean squared error (NMSE) over state-of-the-art solutions in narrowband and wideband scenarios, respectively.

Problem

• THz communications
• UM-MIMO
• near-field
• far-field
• channel estimation

Innovation

• Recurrent Transformer
• Near- and Far-Field Channel Estimation
• Terahertz Communications
• UM-MIMO
• Wideband Channel Estimation