{"ID":2855944,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.12941","arxiv_id":"2510.12941","title":"Computationally Efficient Neural Receivers via Axial Self-Attention","abstract":"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.","short_abstract":"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...","url_abs":"https://arxiv.org/abs/2510.12941","url_pdf":"https://arxiv.org/pdf/2510.12941v2","authors":"[\"SaiKrishna Saketh Yellapragada\",\"Atchutaram K. Kocharlakota\",\"Mário Costa\",\"Esa Ollila\",\"Sergiy A. Vorobyov\"]","published":"2025-10-14T19:39:24Z","proceeding":"eess.SP","tasks":"[\"eess.SP\"]","methods":"[\"Transformer\"]","has_code":false}