发表机构
Tel Aviv University（特拉维夫大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出物理增强图变换器框架，通过预测表面电流的中间监督和可微辐射损失，实现贴片天线高效正演与逆向设计，在8万样本基准上达到MSE 0.17，逆向设计相对误差降低32%。
AI 中文摘要
全波电磁（EM）仿真能够实现精确的贴片天线分析，但对于大规模正演预测和逆向设计而言计算成本过高。我们提出了一种基于网格原生、物理增强的图学习框架，将辐射方向图预测视为不规则表面网格上的信号重建。对于正演问题，采用物理增强中间监督（PAIS）训练GPS图变换器，这是一种辅助的节点级目标，用于预测连接几何与辐射的物理中间量——表面复电流。PAIS在不增加推理成本的情况下提升了多种GNN骨干网络的性能，而打乱电流和非物理对照实验表明，性能提升来源于物理对应关系。方向条件解码和可微的辐射积分一致性损失进一步利用了这种结构。在80,000样本的CST基准上，GPS+PAIS达到MSE 0.17 / PSNR 19.67，泛化到PCA划分，并零样本迁移到典型贴片。对于逆向设计，代理过滤扩散相比最近邻检索相对MSE降低了32%。
英文摘要
Full-wave electromagnetic (EM) simulation enables accurate patch-antenna analysis but is computationally expensive for large-scale forward prediction and inverse design. We present a mesh-native, physics-augmented graph-learning framework that treats radiation-pattern prediction as signal reconstruction on an irregular surface mesh. For the forward problem, a GPS graph transformer is trained with Physics-Augmented Intermediate Supervision (PAIS), an auxiliary node-level objective that predicts complex surface currents, the physical intermediate linking geometry to radiation. PAIS improves multiple GNN backbones at no inference-time cost, while shuffled-current and non-physical controls show the gain comes from physical correspondence. Direction-conditioned decoding and a differentiable radiation-integral consistency loss further exploit this structure. On an 80,000-sample CST benchmark, GPS+PAIS reaches MSE 0.17 / PSNR 19.67, generalizes to a PCA split, and transfers zero-shot to canonical patches. For inverse design, surrogate-filtered diffusion beats nearest-neighbor retrieval by 32% relative MSE.
Comments6 pages, 3 figures, 3 tables. Accepted for oral presentation at the 2026 IEEE International Workshop on Machine Learning for Signal Processing (MLSP 2026), Atlanta, USA. Code, dataset and Colab demo: https://github.com/AviEpstein/GNN-for-Antenna-design
