arXiv cs.AI· Zihang Song, Matteo Zecchin, Bipin Rajendran, Petar Popovski, Osvaldo Simeone·· 13 小时前AI 评分18
EXT-ICLNet：用上下文学习替代信道估计的 MIMO Turbo 接收机
In-Context Learning for MIMO Turbo Receivers
AI 导读
研究者提出基于 Transformer 的检测器 EXT-ICLNet，将导频作为上下文示例，无需显式信道估计即可逼近符号外信息后验分布。该网络通过对数域先验跳连只学习外信息，可直接兼容现有 turbo 接收机。在 4×4 MIMO、16-QAM 与量化前端设置下，EXT-ICLNet 在低至中等 SNR 优于基于显式 CSI 估计的基线，在高 SNR 优于具备完美 CSI 的 EP 基线。
正文 · 原文
Abstract:Multiple-input multiple-output (MIMO) turbo receivers iteratively exchange information between a soft-input soft-output detector and a decoder. Traditional implementations rely on pilot-based channel estimates, which can be poor in the case of short pilot sequences or mismatched channel models, resulting in significant performance degradation. To address this limitation, we introduce the EXTrinsic In-Context Learning Network (EXT-ICLNet), a Transformer-based detector that takes the pilots as in-context examples and is trained to approximate the symbol extrinsic posterior distribution given the pilots, without explicit channel estimation. EXT-ICLNet is thus compatible with existing turbo receivers, as it implements a log-domain prior skip connection, ensuring that the network learns only the extrinsic information. Numerical results for a $4\times4$ MIMO system with 16-QAM and a quantized front end show that EXT-ICLNet outperforms baselines based on explicit channel state information (CSI) estimation at low-to-medium signal-to-noise ratios (SNRs), as well as expectation-propagation (EP) baselines with perfect CSI at high SNRs. The results are further validated by comparing the extrinsic information transfer (EXIT) charts of EXT-ICLNet, a neural EP baseline, and EXT-ICLNet without skip connection, demonstrating how EXT-ICLNet provides a more favorable transfer of a priori and extrinsic information.
| Subjects: | Signal Processing (eess.SP); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2505.06175 [eess.SP] |
| (or arXiv:2505.06175v2 [eess.SP] for this version) | |
| https://doi.org/10.48550/arXiv.2505.06175 arXiv-issued DOI via DataCite |
Submission history
From: Zihang Song [view email]
[v1]
Fri, 9 May 2025 16:29:29 UTC (65 KB)
[v2]
Tue, 27 Oct 2026 11:42:48 UTC (2,866 KB)
来源：arXiv cs.AI · arxiv.org