发表机构
Nokia; Nokia Bell Labs; Aalto University（诺基亚; 诺基亚贝尔实验室; 阿尔托大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究基于5.88 GHz多站点实测数据，评估神经接收器跨站点迁移性能，发现联合训练无损失而微调导致空间遗忘。
AI 中文摘要
大多数已发表的神经接收器研究都是在合成信道模型上进行训练的，尽管真实的空中测量数据可以提升性能。我们考察了这种收益中有多少能从其他站点迁移到一个截然不同的部署站点，以及对该站点的适配在其他地方会付出何种代价。本研究使用了来自5.88 GHz测量活动的数据集，涵盖八个站点和四种环境类型。一个完全在车辆上测量的工业站点被保留作为故意的迁移压力测试。一种接收器架构在四种训练模式下进行训练：仅仿真；其他七个站点；所有站点联合训练；以及先在其他站点训练再在部署站点进行微调。后两种模式使用相同的测量数据但采用不同的优化协议。在五个独立训练的随机种子上，我们在未编码比特错误率（BER）目标为$10^{-1}$和$10^{-2}$（分别代表中等速率和较高速率的编码链路）下评估灵敏度增益。在部署站点上，使用其他站点数据而非仿真数据进行训练，在$10^{-1}$时灵敏度提升0.3 dB，在$10^{-2}$时提升0.1 dB。加入部署站点数据在$10^{-1}$时贡献小于0.1 dB；在$10^{-2}$时，相对于仿真的总增益在联合训练下达到0.4 dB，在微调下达到0.6 dB。联合训练在其他站点测试集上没有表现出可测量的总体损失，而微调虽然带来了更大的部署站点增益，但在$10^{-2}$时使其他站点的灵敏度降低了约0.1 dB，这表明存在空间遗忘。
英文摘要
Most published neural-receiver studies train on synthetic channel models, although real over-the-air measurements can improve performance. We examine how much of this benefit transfers from other sites to a distinctly different deployment site and what adaptation to that site costs elsewhere. The study uses a dataset from a 5.88 GHz measurement campaign covering eight sites and four environment types. An industrial site measured entirely from a vehicle is withheld as a deliberate transfer stress test. One receiver architecture is trained under four regimes: simulation alone; seven other sites; all sites jointly; and other-site training followed by deployment-site finetuning. The last two use identical measurements through different optimization protocols. Across five independently trained seeds, sensitivity gains are evaluated at uncoded bit error rate (BER) targets of $10^{-1}$ and $10^{-2}$, representative of moderate- and higher-rate coded links, respectively. On the deployment site, training with other-site data rather than simulation improves sensitivity by 0.3 dB at $10^{-1}$ and 0.1 dB at $10^{-2}$. Adding deployment-site data contributes less than 0.1 dB at $10^{-1}$; at $10^{-2}$, total gains over simulation reach 0.4 dB with joint training and 0.6 dB with finetuning. Joint training shows no measurable aggregate loss on the other-site test set, whereas finetuning yields the larger deployment-site gain but reduces other-site sensitivity by about 0.1 dB at $10^{-2}$, indicating spatial forgetting.
