Physics-Informed Deep Neural Network Design of Reactively Loaded Metasurfaces

TL;DR

This paper introduces a physics-informed deep neural network method for the inverse design of reactively loaded metasurfaces, enabling efficient synthesis of desired radiation patterns with reduced computational cost.

Contribution

It combines a DNN with a physics-based forward solver for fast, accurate metasurface design, which is a novel integration for inverse electromagnetic design tasks.

Findings

Accurately synthesizes shaped and steered radiation patterns.

Reduces computational cost of metasurface design.

Validated with full-wave electromagnetic simulations.

Abstract

A tandem deep neural network approach is presented for the inverse design of reactively loaded metasurfaces with prescribed far-field radiation characteristics. The proposed approach integrates a deep neural network (DNN) with a physics-based microwave network forward solver. The DNN maps target far-field patterns to distributions of reactive loads across the metasurface unit cells. The predicted distribution of reactive loads is evaluated by the forward solver to compute the resulting radiation pattern and guide the learning process through a cosine-similarity loss function. The forward solver enables a fast evaluation of the metasurface's electromagnetic response, significantly reducing the computational cost required for training. The proposed approach is applied to a metasurface with aperture-coupled unit cells loaded with reactances. Several design examples are presented to demonstrate the accurate synthesis of shaped and steered radiation patterns. Full-wave electromagnetic simulations are performed to validate the accuracy of the designed beamforming metasurfaces.