TY - JOUR
T1 - Hybrid Inverse Design Framework for Microwave Ultrabroadband Achromatic Metalens
AU - Bai, Yihao
AU - Gu, Zhen
AU - Wang, Yifeng
AU - Wang, Xiong
AU - Zhang, Hualiang
AU - An, Sensong
AU - Ding, Jun
N1 - Publisher Copyright:
© 1963-2012 IEEE.
PY - 2025
Y1 - 2025
N2 - A hybrid inverse design framework is proposed for the efficient design of ultra broadband achromatic metalenses with very simple resonant structures. The framework combines a data-driven inverse dispersion engineering approach along with an optimal random spatial multiplexing (ORSM) method. The two-stage framework is based on a design strategy that segments the entire operating band into multiple subbands and performs continuous achromatic design within each subband. As a result, the framework alleviates the phase dispersion requirements imposed on meta-atoms by conventional dispersion engineering for ultra broadband operation, which typically necessitates complex and time-consuming meta-atom design. The inverse dispersion engineering, trained on a compact dataset of only 792 samples, accurately predicts meta-atom structures with desired phase and amplitude responses. Subsequently, the ORSM method optimizes the spatial arrangement of the inverse-designed meta-atoms, ultimately generating the final metalens layout. To validate the performance of the proposed framework, a reflective microwave ultra broadband metalens is designed, simulated, and characterized. The experimental results demonstrate achromatic focusing with a focal length of 215 mm and an average focusing efficiency of 62.73% over the operating band of 8–16 GHz. This method could pave a new way for the development of achromatic meta-devices for wideband microwave applications.
AB - A hybrid inverse design framework is proposed for the efficient design of ultra broadband achromatic metalenses with very simple resonant structures. The framework combines a data-driven inverse dispersion engineering approach along with an optimal random spatial multiplexing (ORSM) method. The two-stage framework is based on a design strategy that segments the entire operating band into multiple subbands and performs continuous achromatic design within each subband. As a result, the framework alleviates the phase dispersion requirements imposed on meta-atoms by conventional dispersion engineering for ultra broadband operation, which typically necessitates complex and time-consuming meta-atom design. The inverse dispersion engineering, trained on a compact dataset of only 792 samples, accurately predicts meta-atom structures with desired phase and amplitude responses. Subsequently, the ORSM method optimizes the spatial arrangement of the inverse-designed meta-atoms, ultimately generating the final metalens layout. To validate the performance of the proposed framework, a reflective microwave ultra broadband metalens is designed, simulated, and characterized. The experimental results demonstrate achromatic focusing with a focal length of 215 mm and an average focusing efficiency of 62.73% over the operating band of 8–16 GHz. This method could pave a new way for the development of achromatic meta-devices for wideband microwave applications.
KW - Deep neural networks (DNNs)
KW - dispersion engineering
KW - inverse design
KW - spatial multiplexing
KW - ultra broadband achromatic metalenses
UR - https://www.scopus.com/pages/publications/105010304455
U2 - 10.1109/TAP.2025.3584622
DO - 10.1109/TAP.2025.3584622
M3 - 文章
AN - SCOPUS:105010304455
SN - 0018-926X
VL - 73
SP - 7816
EP - 7829
JO - IEEE Transactions on Antennas and Propagation
JF - IEEE Transactions on Antennas and Propagation
IS - 10
ER -