Conditional Prior-based Non-stationary Channel Estimation Using Accelerated Diffusion Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Mohsin, Muhammad Ahmed, Bilal, Ahsan, Umer, Muhammad, Aali, Asad, Jamshed, Muhammad Ali, Hougen, Dean F., Cioffi, John M.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916956560424960
author Mohsin, Muhammad Ahmed
Bilal, Ahsan
Umer, Muhammad
Aali, Asad
Jamshed, Muhammad Ali
Hougen, Dean F.
Cioffi, John M.
author_facet Mohsin, Muhammad Ahmed
Bilal, Ahsan
Umer, Muhammad
Aali, Asad
Jamshed, Muhammad Ali
Hougen, Dean F.
Cioffi, John M.
contents Wireless channels in motion-rich urban microcell (UMi) settings are non-stationary; mobility and scatterer dynamics shift the distribution over time, degrading classical and deep estimators. This work proposes conditional prior diffusion for channel estimation, which learns a history-conditioned score to denoise noisy channel snapshots. A temporal encoder with cross-time attention compresses a short observation window into a context vector, which captures the channel's instantaneous coherence and steers the denoiser via feature-wise modulation. In inference, an SNR-matched initialization selects the diffusion step whose marginal aligns with the measured input SNR, and the process follows a shortened, geometrically spaced schedule, preserving the signal-to-noise trajectory with far fewer iterations. Temporal self-conditioning with the previous channel estimate and a training-only smoothness penalty further stabilizes evolution without biasing the test-time estimator. Evaluations on a 3GPP benchmark show lower NMSE across all SNRs than LMMSE, GMM, LSTM, and LDAMP baselines, demonstrating stable performance and strong high SNR fidelity.
format Preprint
id arxiv_https___arxiv_org_abs_2509_15182
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Conditional Prior-based Non-stationary Channel Estimation Using Accelerated Diffusion Models
Mohsin, Muhammad Ahmed
Bilal, Ahsan
Umer, Muhammad
Aali, Asad
Jamshed, Muhammad Ali
Hougen, Dean F.
Cioffi, John M.
Distributed, Parallel, and Cluster Computing
Wireless channels in motion-rich urban microcell (UMi) settings are non-stationary; mobility and scatterer dynamics shift the distribution over time, degrading classical and deep estimators. This work proposes conditional prior diffusion for channel estimation, which learns a history-conditioned score to denoise noisy channel snapshots. A temporal encoder with cross-time attention compresses a short observation window into a context vector, which captures the channel's instantaneous coherence and steers the denoiser via feature-wise modulation. In inference, an SNR-matched initialization selects the diffusion step whose marginal aligns with the measured input SNR, and the process follows a shortened, geometrically spaced schedule, preserving the signal-to-noise trajectory with far fewer iterations. Temporal self-conditioning with the previous channel estimate and a training-only smoothness penalty further stabilizes evolution without biasing the test-time estimator. Evaluations on a 3GPP benchmark show lower NMSE across all SNRs than LMMSE, GMM, LSTM, and LDAMP baselines, demonstrating stable performance and strong high SNR fidelity.
title Conditional Prior-based Non-stationary Channel Estimation Using Accelerated Diffusion Models
topic Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2509.15182