A Combined Machine-Learning/Optimization-Based Approach for Inverse Design of Nonuniform Bianisotropic Metasurfaces

Naseri, Parinaz; Pearson, Stewart; Wang, Zhengzheng; Hum, Sean V.

doi:10.1109/tap.2021.3137496

“…In the microscopic design step, the optimized set of (Z se , Y sm , K em ) are first converted to scattering parameters and then are realized using physical unit cells. To find the optimized unit cell, a previously proposed approach is employed [32].…”

Section: Microscopic Optimizer Using Dnn Surrogate Modelsmentioning

confidence: 99%

“…In order to make the EMMS design process more streamlined, we aim to use the integrated macroscopic and microscopic optimizers the authors previously presented [32] to solve for an EMMS that forms two beams. Although the results were previously verified in simulation using Ansys HFSS, we will go one step further to experimentally verify them using a fabricated EMMS illuminated by a standard gain horn (SGH) in a near-field chamber.…”

Section: Introductionmentioning

confidence: 99%

Inverse Design and Experimental Verification of a Bianisotropic Metasurface Using Optimization and Machine Learning

Pearson¹,

Naseri²,

Hum³

2022

Preprint

Self Cite

0

View full text Add to dashboard Cite

Electromagnetic metasurfaces have attracted significant interest recently due to their low profile and advantageous applications. Practically, many metasurface designs start with a set of constraints for the radiated far-field, such as main-beam direction(s) and side lobe levels, and end with a non-uniform physical structure for the surface. This problem is quite challenging, since the required tangential field transformations are not completely known when only constraints are placed on the scattered fields. Hence, the required surface properties cannot be solved for analytically. Moreover, the translation of the desired surface properties to the physical unit cells can be time-consuming and difficult, as it is often a one-tomany mapping in a large solution space. Here, we divide the inverse design process into two steps: a macroscopic and microscopic design step. In the former, we use an iterative optimization process to find the surface properties that radiate a far-field pattern that complies with specified constraints. This iterative process exploits non-radiating currents to ensure a passive and lossless design. In the microscopic step, these optimized surface properties are realized with physical unit cells using machine learning surrogate models. The effectiveness of this end-to-end synthesis process is demonstrated through measurement results of a beam-splitting prototype.

show abstract

“…The macroscopic optimizer is similar to the ones reported previously [32], [33]. It is formulated using a homogenized two-dimensional EMMS model constructed with the MoM.…”

Section: Admm-based Macroscopic Optimizermentioning

confidence: 99%

“…As a result, the surface parameters in the two-dimensional problem reduce to a scalars jX se , jB sm , and K em . This yields the matrix equations [32], [33]…”

Section: Admm-based Macroscopic Optimizermentioning

confidence: 99%

See 1 more Smart Citation

A Beam-Splitting Bianisotropic Metasurface Designed by Optimization and Machine Learning

Pearson

¹

,

Naseri

²

,

Hum

³

2022

IEEE Open J. Antennas Propag.

Self Cite

5

0

View full text Add to dashboard Cite

Electromagnetic metasurfaces have attracted significant interest recently due to their low profile and advantageous applications. Practically, many metasurface designs start with a set of constraints for the radiated far-field, such as main-beam direction(s) and side lobe levels, and end with a non-uniform physical structure for the surface. This problem is quite challenging, since the required tangential field transformations are not completely known when only constraints are placed on the scattered fields. Hence, the required surface properties cannot be solved for analytically. Moreover, the translation of the desired surface properties to the physical unit cells can be time-consuming and difficult, as it is often a one-tomany mapping in a large solution space. Here, we divide the inverse design process into two steps: a macroscopic and microscopic design step. In the former, we use an iterative optimization process to find the surface properties that radiate a far-field pattern that complies with specified constraints. This iterative process exploits non-radiating currents to ensure a passive and lossless design. In the microscopic step, these optimized surface properties are realized with physical unit cells using machine learning surrogate models. The effectiveness of this end-to-end synthesis process is demonstrated through measurement results of a beam-splitting prototype.

show abstract

A Dual-Path Generative Adversarial Network-based inverse design method for broadband RCS reduction metasurface element patterns

Liu,

Bai,

Qiu

et al. 2024

Optics and Lasers in Engineering

0

View full text Add to dashboard Cite

A Combined Machine-Learning/Optimization-Based Approach for Inverse Design of Nonuniform Bianisotropic Metasurfaces

Cited by 19 publications

References 41 publications

Inverse Design and Experimental Verification of a Bianisotropic Metasurface Using Optimization and Machine Learning

Inverse Design and Experimental Verification of a Bianisotropic Metasurface Using Optimization and Machine Learning

A Beam-Splitting Bianisotropic Metasurface Designed by Optimization and Machine Learning

A Dual-Path Generative Adversarial Network-based inverse design method for broadband RCS reduction metasurface element patterns

Contact Info

Product

Resources

About