Physics-Informed Neural Networks with Architectural Physics Embedding for Large-Scale Wave Field Reconstruction

Researchers developed Physics-Embedded PINNs (PE-PINN) that achieve 10x faster convergence than standard physics-informed neural networks and orders of magnitude memory reduction compared to traditional methods for large-scale wave field reconstruction. The breakthrough enables high-fidelity electromagnetic wave modeling for wireless communications, sensing, and room acoustics applications.

• →PE-PINN integrates physical principles directly into neural network architecture, not just loss functions, improving convergence and stability.
• →The method achieves 10x speedup in convergence compared to standard PINNs and several orders of magnitude memory reduction versus FEM.
• →New envelope transformation layer mitigates spectral bias using kernels parameterized by source properties and wave physics.
• →Enables large-scale 2D/3D electromagnetic wave reconstruction for room-scale domains with complex wave interactions.
• →Applications span wireless communications, sensing, room acoustics, and other fields requiring wave field analysis.