发表机构
Southeast University; Nanjing University; National Sun Yat-sen University（东南大学; 南京大学; 国立中山大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对无线电图盲预测中环境与配置表示不完整的问题，提出RadioDecomp方法，通过先验引导加确定性残差细化（实例化为RadioLSR），在跨配置和跨环境场景下提升泛化性能。
AI 中文摘要
无线电图盲预测从传播环境和基站（BS）配置的可观测表示中推断无线电图，无需现场测量。这些表示本质上是不完整的，无法唯一确定目标无线电图。在平方损失下，我们将条件均值无线电图确定为总体最优的确定性目标，并将域风险分解为目标逼近误差和不可约不确定性。训练-测试风险差距促使传播先验作为跨域指导，尽管其部分或简化形式可能使可达到的预测器产生偏差。因此，我们提出RadioDecomp，将先验引导的预测器视为可修正的基础，并使用确定性残差细化来学习其剩余的可预测差异。我们将RadioDecomp实例化为RadioLSR（视距-阴影-残差）。在跨配置和跨环境设置下的实验表明，RadioLSR在跨配置泛化方面尤为有效，并在跨环境泛化下相对于受控的整体对应方法提供了总体增益。
英文摘要
Radiomap blind prediction infers radiomaps from observable representations of the propagation environment and base station (BS) configuration without field measurements. These representations are inherently incomplete and cannot uniquely determine the target radiomap. Under squared loss, we identify the conditional-mean radiomap as the population-optimal deterministic target and decompose domain risk into target-approximation error and irreducible uncertainty. The train-test risk gap motivates propagation priors as cross-domain guidance, although their partial or simplified forms may bias the attainable predictor. We therefore propose RadioDecomp, which treats a prior-guided predictor as a correctable base and uses deterministic residual refinement to learn its remaining predictable discrepancy. We instantiate RadioDecomp as RadioLSR (LoS-Shadow-Residual). Experiments under cross-configuration and cross-environment settings show that RadioLSR is especially effective for cross-configuration generalization and provides overall gains over a controlled monolithic counterpart under cross-environment generalization.
CommentsThis paper has been accepted for presentation at IEEE Globecom 2026
