HELENA: High-Efficiency Learning-based channel Estimation using dual Neural Attention

TL;DR

HELENA is a compact deep learning model for channel estimation in 5G systems, combining attention mechanisms to achieve high accuracy with reduced inference time and fewer parameters.

Contribution

It introduces a novel lightweight neural network with dual attention mechanisms that outperforms existing models in speed and efficiency for channel estimation.

Findings

Reduces inference time by 45% compared to CEViT.

Achieves comparable accuracy to state-of-the-art methods.

Uses 8 times fewer parameters than CEViT.

Abstract

Accurate channel estimation is critical for high-performance Orthogonal Frequency-Division Multiplexing systems such as 5G New Radio, particularly under low signal-to-noise ratio and stringent latency constraints. This letter presents HELENA, a compact deep learning model that combines a lightweight convolutional backbone with two efficient attention mechanisms: patch-wise multi-head self-attention for capturing global dependencies and a squeeze-and-excitation block for local feature refinement. Compared to CEViT, a state-of-the-art vision transformer-based estimator, HELENA reduces inference time by 45.0\% (0.175\,ms vs.\ 0.318\,ms), achieves comparable accuracy ($-16.78$\,dB vs.\ $-17.30$\,dB), and requires $8\times$ fewer parameters (0.11M vs.\ 0.88M), demonstrating its suitability for low-latency, real-time deployment.