---
格式版本: 2
标题: "Physics-Unrolled Neural Operator for Wireless Field Modeling"
原文链接: "https://arxiv.org/abs/2608.18495"
发布日期: "2026-08-19"
发布时间校准状态: "found"
发布时间需复核: "否"
发布时间来源: "rule:local:strict_original_body"
发布时间证据: "**\\[v1\\]** Wed, 19 Aug 2026 03:39:55 UTC (7,778 KB)"
发布时间校准原因: "规则确认唯一严格发布时间，来源 local:strict_original_body"
发布时间校准置信度: "high"
发布时间候选数量: 18
发布时间严格候选数量: 6
发布时间原页读取状态: "source template page reused from URL open"
发布时间未找到原因: ""
发布时间校准时间: "2026-08-20T15:25:41+08:00"
发布时间仲裁状态: "skipped"
发布时间仲裁尝试次数: 0
发布时间仲裁耗时毫秒: 0
发现时间: "2026-08-20T15:18:01+08:00"
入库时间: "2026-08-20T07:25:41.677Z"
来源平台: "arXiv 学术论文搜索"
搜索渠道: "source_template"
搜索词: "https://arxiv.org/search/?query=deployment&searchtype=all"
匹配关键词:
  - "deployment"
  - "AI"
相关厂家:
  []
相关专家:
  []
内容类型: "网页"
抓取工具: "Free Fetch + Defuddle"
清洗工具: "Defuddle Markdown + Defuddle/Readability 正文提取"
原始附件:
  []
AI优质: "否"
AI打分: 12
AI分档: "非优质"
AI质检状态: "不通过"
AI打分理由: "该资料为arXiv机器学习论文，主题是无线场建模的神经算子，与超节点、AI Rack、机柜级AI基础设施等完全无关，仅命中deployment搜索词，无相关技术或商业信息。"
AI质检模型: "ali-deepseek-v4-flash"
AI质检时间: "2026-08-20T15:29:38+08:00"
AI主题相关性: 0
AI来源权威性: 5
AI新颖性: 3
AI技术细节: 2
AI商业部署信号: 0
AI完整性: 2
AI摘要: "作者提出物理展开混合神经算子（PU-HNO），一种三阶段级联模型，利用低保真射线追踪输出和场景先验预测高保真室内无线电地图，逐步建模反射、衍射和散射效应；"
AI摘要模型: "ali-deepseek-v4-flash"
AI摘要时间: "2026-09-07T03:17:12.498Z"
采集批次: "2026年8月20日14点19分32秒"
采集批次ID: "20260820-141932-079"
去重键: "https://arxiv.org/abs/2608.18495"
---

## Computer Science > Machine Learning

## Title:Physics-Unrolled Neural Operator for Wireless Field Modeling

[View PDF](https://arxiv.org/pdf/2608.18495) [HTML (experimental)](https://arxiv.org/html/2608.18495v1)

> Abstract:Radio maps are essential for wireless decision-making tasks such as access-point placement, coverage planning, and localization, but their fine spatial details are governed by complex propagation effects and are costly to simulate accurately. Machine learning offers a path to high-fidelity radio-map prediction without running expensive high-fidelity simulations for every scene. However, generating high-quality training labels at scale is also difficult: the affordable labels come from finite-ray simulations, which are richer than low-fidelity inputs but carry residual Monte Carlo noise. We address this challenge with Physics-Unrolled Hybrid Neural Operator (PU-HNO), a three-stage cascade that predicts high-fidelity indoor radio maps from low-fidelity ray-tracing outputs and scene priors by progressively capturing reflection, diffraction, and scattering effects, rather than treating radio maps as generic images. We prove that, under conditionally unbiased label noise, the model can learn stable propagation structure and outperform its own training labels. Experiments across diverse floorplans show that PU-HNO outperforms image-to-image baselines, wireless learning models, and monolithic neural operators across both image-quality and wireless deployment metrics.

| Comments: |  |
| --- | --- |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | [arXiv:2608.18495](https://arxiv.org/abs/2608.18495) \[cs.LG\] |
|  | (or [arXiv:2608.18495v1](https://arxiv.org/abs/2608.18495v1) \[cs.LG\] for this version) |
|  | [https://doi.org/10.48550/arXiv.2608.18495](https://doi.org/10.48550/arXiv.2608.18495) |

## Submission history

From: Rafid Umayer Murshed \[[view email](https://arxiv.org/show-email/0d1f501e/2608.18495)\]  
**\[v1\]** Wed, 19 Aug 2026 03:39:55 UTC (7,778 KB)

[Which authors of this paper are endorsers?](https://arxiv.org/auth/show-endorsers/2608.18495) | Disable MathJax ([What is MathJax?](https://info.arxiv.org/help/mathjax.html))