Rapid inverse design of microwave devices with scattering matrices

Inverse design (InvDes) combines simulation and optimization to algorithmically discover physical device layouts beyond conventional intuition. However, typical InvDes approaches require hundreds or thousands of simulations and numerous initializations to adequately explore complex design spaces. This computational burden is especially pronounced in metallic microwave-frequency devices, which involve highly nonconvex optimization landscapes and demand computationally expensive, high-resolution full-wave electromagnetic simulations. Here, we introduce a significantly accelerated inverse design method tailored for microwave-frequency devices. Our approach leverages rigorous full-wave simulations to precompute scattering parameters of small, simple elements, which are subsequently combined to form larger devices during optimization. This scattering-matrix-based approach enables simulations that are approximately $10\phantom{\rule{0.1em}{0ex}}000$ times faster and use approximately $100$--$10\phantom{\rule{0.1em}{0ex}}000$ times less memory than conventional full-wave methods, while maintaining high accuracy. Coupling this accelerated simulator with an adaptive random search optimization algorithm allows rapid, thorough exploration of highly nonconvex design spaces. We demonstrate the effectiveness and versatility of our method by efficiently designing a range of commonly used microwave devices, including couplers, power splitters, phase shifters, impedance-matching circuits, and demultiplexers.

Authors

Institutions

Publication Details

Journal
Physical Review Applied
Published
2025-09-29
DOI
https://doi.org/10.1103/qlfy-4qhw
Primary Topic
Metamaterials and Metasurfaces Applications
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Rapid inverse design of microwave devices with scattering matrices

Nathaniel Morrison, Yue Ma, X. T. He, Tianhang Xie
Physical Review Applied
Metamaterials and Metasurfaces Applications
article

Rapid inverse design of microwave devices with scattering matrices

Nathaniel Morrison, Yue Ma, X. T. He, Tianhang Xie
article en

Abstract

Inverse design (InvDes) combines simulation and optimization to algorithmically discover physical device layouts beyond conventional intuition. However, typical InvDes approaches require hundreds or thousands of simulations and numerous initializations to adequately explore complex design spaces. This computational burden is especially pronounced in metallic microwave-frequency devices, which involve highly nonconvex optimization landscapes and demand computationally expensive, high-resolution full-wave electromagnetic simulations. Here, we introduce a significantly accelerated inverse design method tailored for microwave-frequency devices. Our approach leverages rigorous full-wave simulations to precompute scattering parameters of small, simple elements, which are subsequently combined to form larger devices during optimization. This scattering-matrix-based approach enables simulations that are approximately $10\phantom{\rule{0.1em}{0ex}}000$ times faster and use approximately $100$--$10\phantom{\rule{0.1em}{0ex}}000$ times less memory than conventional full-wave methods, while maintaining high accuracy. Coupling this accelerated simulator with an adaptive random search optimization algorithm allows rapid, thorough exploration of highly nonconvex design spaces. We demonstrate the effectiveness and versatility of our method by efficiently designing a range of commonly used microwave devices, including couplers, power splitters, phase shifters, impedance-matching circuits, and demultiplexers.

Physical Review AppliedVol. 24(4)
Lawrence Berkeley National Laboratory (US), University of California, Berkeley (US)
Openalex Percentile: Top 78%
Metamaterials and Metasurfaces Applications
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.