Learned Score-Based Turbo Message Passing for Channel Recovery From Spatially Compressed Measurements
Abstract
Hybrid beamforming (HBF) reduces the hardware cost of massive MIMO base stations, but its channel acquisition requires recovering the full spatial-domain channel from highly compressed beam-domain observations. This paper proposes a learned score-based turbo message passing (L-STMP) framework for uplink HBF channel recovery under extreme spatial compression. By deeply unfolding the turbo message-passing loop and introducing trainable calibration parameters and learnable score priors in each layer, L-STMP liberates STMP from the idealized compression-matrix requirements of right-rotational invariance, and reduces inference complexity to only several unfolded layers. In addition, to preserve dominant channel components that random discrete Fourier transform (DFT) selection may miss, we introduce precoding matrix indicator (PMI)-assisted beam selection as a low-overhead directional prior. Simulation results demonstrate that L-STMP consistently outperforms existing baselines and obtains substantial reconstruction gains under PMI-assisted compression in channel reconstruction.
Related
- Recovery Guarantees for Posterior Sampling of One-Bit Compressed Sensing
- Temporal Degeneracy Limits Message Passing Decoding of QLDPC Codes Under Repeated Noisy Syndrome Measurements
- Deep-learning side-channel analysis of embedded ML-KEM: Single-trace message recovery, cross-device generalization, and frequency-band leakage characterization
- Identifying unknown time delay and spatially varying coefficients in a reaction-diffusion equation from boundary measurements
- Simultaneous recovery of multiple parameters in nonlocal diffusion equations from internal measurements
BibTeX
@misc{wei2026learned,
title = {{Learned Score-Based Turbo Message Passing for Channel Recovery From Spatially Compressed Measurements}},
author = {Yiming Wei and Xiaojun Yuan},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.17785},
eprint = {2610.17785},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.17785},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Learned Score-Based Turbo Message Passing for Channel Recovery From Spatially Compressed Measurements
AU - Yiming Wei
AU - Xiaojun Yuan
PY - 2026
DA - 2026/10/13
PB - arXiv
DO - 10.48550/arxiv.2610.17785
UR - https://arxiv.org/abs/2610.17785
AB - Hybrid beamforming (HBF) reduces the hardware cost of massive MIMO base stations, but its channel acquisition requires recovering the full spatial-domain channel from highly compressed beam-domain observations. This paper proposes a learned score-based turbo message passing (L-STMP) framework for uplink HBF channel recovery under extreme spatial compression. By deeply unfolding the turbo message-passing loop and introducing trainable calibration parameters and learnable score priors in each layer, L-STMP liberates STMP from the idealized compression-matrix requirements of right-rotational invariance, and reduces inference complexity to only several unfolded layers. In addition, to preserve dominant channel components that random discrete Fourier transform (DFT) selection may miss, we introduce precoding matrix indicator (PMI)-assisted beam selection as a low-overhead directional prior. Simulation results demonstrate that L-STMP consistently outperforms existing baselines and obtains substantial reconstruction gains under PMI-assisted compression in channel reconstruction.
ER -