Physics-Informed Spatial-Temporal Transformer for Terahertz Near-Field Beam Tracking

TL;DR

This paper introduces PAST-TT, a physics-informed spatial-temporal transformer that significantly improves near-field beam tracking accuracy in terahertz UM-MIMO systems by leveraging parallax and lightweight neural architectures.

Contribution

The paper proposes a novel low-overhead near-field tracking method combining parallax exploitation with a lightweight transformer-based model, enhancing accuracy and reducing latency.

Findings

Achieves 7.81 mm distance RMSE at 15 dB SNR.

Reduces RMSE by over 23% compared to existing methods.

Maintains a critical path latency of 0.61 ms.

Abstract

Terahertz (THz) ultra-massive multiple-input multiple-output (UM-MIMO) promises ultra-high throughput, while its highly directional beams demand rapid and accurate beam tracking driven by precise user-state estimation. Moreover, large array apertures at high frequencies induce near-field propagation effects, where far-field modeling becomes inaccurate and near-field parametric channel estimation is costly. Bypassing near-field codebook, PAST-TT is proposed to bridge near-field tracking with low-overhead far-field codebook probing by exploiting parallax, amplified by widely spaced subarrays. With comb-type frequency-division multiplexing pilots, each subarray yields frequency-affine phase signatures whose frequency and temporal increments encode propagation delay and its variation between frames. Building on these signatures, a Parallax-Aware Spatial Transformer (PAST) compresses them and outputs per-frame position estimates with token reliability to downweight bad frames, regularized by a physics-in-the-loop consistency loss. A causal Temporal Transformer (TT) then performs reliability-aware filtering and prediction over a sliding window to initialize the beam of the next frame. Acting on short token sequences, PAST-TT avoids a monolithic spatial-temporal network over raw pilots, which keeps the model lightweight with a critical path latency of 0.61 ms. Simulations show that at 15 dB signal-to-noise ratio, PAST achieves 7.81 mm distance RMSE and 0.0588{\deg} angle RMSE. Even with a bad-frame rate of 0.1, TT reduces the distance and angle prediction RMSE by 23.1% and 32.8% compared with the best competing tracker.