← Back to all news
arXiv Machine Learning By Zhibin Zhang, Robert Potekhin, Ziwei Wan, Vladimir Lyashev, Zhen Gao

ReFLEX: Length-Generalizable CSI Denoising for MIMO-OFDM via Relative-Frequency Bias

Read the original on arXiv Machine Learning →

arXiv:2606. 00263v1 Announce Type: cross Abstract: This letter studies CSI denoising for MIMO--OFDM with variable NR resource block (RB) allocations.

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 Machine Learning.

Related stories

arXiv AI
Sep 17

Semantic CSI Feedback for Beam Selection: When Task-Aware Embeddings from Sparse Pilots Outperform Full-Bandwidth Reconstruction

The paper introduces a semantic communication approach for CSI feedback in FDD massive MIMO, where the UE sends a learned embedding optimized for beam selection rather than reconstructing the full channel. Experiments show that an 8‑dimensional embedding derived from just 43 NR CSI‑RS pilots in the angular‑delay domain achieves the best beam prediction accuracy, surpassing methods that use the entire 512‑subcarrier channel. This demonstrates that beam‑relevant information is inherently low‑dimensional, allowing the semantic encoder to discard irrelevant details and transmit only the intent needed for beam selection.

By Cristian J. Vaca-Rubio, Konstantinos Vandikas, Aneta Vulgarakis Feljan
rag
More like this →
arXiv AI
Jul 21

Hybrid Mamba-Attention Neural Architecture for Channel Estimation

arXiv:2601. 17108v2 Announce Type: replace-cross Abstract: This paper proposes a hybrid Mamba-attention neural architecture to achieve improved channel estimation for orthogonal frequency-division multiplexing (OFDM) waveforms, particularly for configurations with a large number of subcarriers.

By Dianxin Luan, Chengsi Liang, Jie Huang, Zheng Lin, Kaitao Meng, John Thompson, Cheng-Xiang Wang
llms
More like this →