Saved in:
Bibliographic Details
Main Authors: Yellapragada, SaiKrishna Saketh, Kocharlakota, Atchutaram K., Costa, Mário, Ollila, Esa, Vorobyov, Sergiy A.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.12941
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914380412616704
author Yellapragada, SaiKrishna Saketh
Kocharlakota, Atchutaram K.
Costa, Mário
Ollila, Esa
Vorobyov, Sergiy A.
author_facet Yellapragada, SaiKrishna Saketh
Kocharlakota, Atchutaram K.
Costa, Mário
Ollila, Esa
Vorobyov, Sergiy A.
contents Deep learning-based neural receivers offer promising physical-layer solutions for next-generation wireless systems. We propose an axial self-attention transformer neural receiver that achieves state-of-the-art Block Error Rate (BLER) performance with significantly improved computational efficiency during inference and large-scale training. By factorizing attention operations along temporal and spectral axes, the proposed architecture reduces computational complexity from $O((TF)^2)$ to $O(T^2F+TF^2)$, yielding substantially fewer floating-point operations and attention matrix multiplications per transformer block. Experimental validation under 3GPP Clustered Delay Line (CDL) channels demonstrates consistent performance gains across varying mobility scenarios. Under non-line-of-sight conditions, our proposed axial neural receiver outperforms global self-attention and convolutional neural receiver baselines at 10% BLER and 1% BLER respectively, with reduced computational complexity.
format Preprint
id arxiv_https___arxiv_org_abs_2510_12941
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Computationally Efficient Neural Receivers via Axial Self-Attention
Yellapragada, SaiKrishna Saketh
Kocharlakota, Atchutaram K.
Costa, Mário
Ollila, Esa
Vorobyov, Sergiy A.
Signal Processing
Deep learning-based neural receivers offer promising physical-layer solutions for next-generation wireless systems. We propose an axial self-attention transformer neural receiver that achieves state-of-the-art Block Error Rate (BLER) performance with significantly improved computational efficiency during inference and large-scale training. By factorizing attention operations along temporal and spectral axes, the proposed architecture reduces computational complexity from $O((TF)^2)$ to $O(T^2F+TF^2)$, yielding substantially fewer floating-point operations and attention matrix multiplications per transformer block. Experimental validation under 3GPP Clustered Delay Line (CDL) channels demonstrates consistent performance gains across varying mobility scenarios. Under non-line-of-sight conditions, our proposed axial neural receiver outperforms global self-attention and convolutional neural receiver baselines at 10% BLER and 1% BLER respectively, with reduced computational complexity.
title Computationally Efficient Neural Receivers via Axial Self-Attention
topic Signal Processing
url https://arxiv.org/abs/2510.12941