arXiv AI By Yuzhi Yang, Omar Alhussein, M\'erouane Debbah

Non-Identical Diffusion Models in MIMO-OFDM Channel Generation

Read the original on arXiv AI →

arXiv:2509. 01641v3 Announce Type: replace-cross Abstract: We propose a novel diffusion model, termed the non-identical diffusion model, and investigate its application to wireless orthogonal frequency division multiplexing (OFDM) channel generation.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv AI
6d ago

JEPA-Inspired Latent Field Prediction for Sparse-Pilot MIMO-OFDM Channel Reconstruction

The paper introduces a JEPA-inspired latent field prediction model for reconstructing MIMO‑OFDM channels from sparse pilots. It maps observed CSI into latent time‑frequency tokens, predicts masked or future latent states using historical multi‑band context, and fuses these predictions with current noisy pilots to recover the full channel. Ray‑tracing experiments demonstrate that this approach reduces normalized mean square error compared to pilot‑only and direct reconstruction baselines.

By Yuzhi Yang, Brahim Mefgouda, Hang Zou, Lina Bariah, Anis Bara, Yuhuan Lu, Hao Zhang, Merouane Debbah
arXiv Machine Learning
Oct 8

SNR-Gated LSTM-Conditioned Diffusion Model for MIMO Channel Estimation

The paper introduces a diffusion-based channel estimation framework for MIMO systems that incorporates temporal dynamics via an LSTM-conditioned denoiser. It employs a learnable SNR-gated fusion mechanism to balance observation fidelity with generative priors across a wide SNR range, and uses DDIM-style reverse updates with SNR-adaptive truncation to reduce inference latency. Simulations on standardized time-evolving channel models show consistent performance gains over existing diffusion-based baselines while maintaining low latency.

By Jixing Zhou, Xinming Huang