Learning stable radiation boundaries for wave simulations via passive neural state-space models
Aotu Li, Xiaolong Wang, Yuchen Wang, Shaoqiang Wang · 2026
Read Learning stable radiation boundaries for wave simulations via passive neural state-space models by Aotu Li, Xiaolong Wang, Yuchen Wang, Shaoqiang Wang (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13266139.txt.
Description
Aotu Li and the team propose a neural boundary method for wave simulations (2026). Simulations must cut endless space down to a finite grid without fake reflections. Standard absorbing layers need many grid rows and extra variables. A single-layer learned boundary could cut those costs.
The method uses a neural state-space model to predict boundary cell answers directly. Physical constraints keep it stable over long runs. Tests showed low reflection with substantially reduced computing overhead. No auxiliary fields need step-by-step evolution, unlike conventional matched layers. The authors plan extension to larger three-dimensional problems and complex boundary shapes.
Catalog record
| CREATOR | Aotu Li, Xiaolong Wang, Yuchen Wang, Shaoqiang Wang |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 82 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |