Robust Large-scale Metasurface Inverse Design Using Phaseless Data with Invertible Neural Network

Zhedong Wang,Chao Qian,Yasir Saifullah,Yuan Fang,Er-Ping Li,Hongsheng Chen
DOI: https://doi.org/10.1109/tap.2024.3484014
IF: 5.7
2024-01-01
IEEE Transactions on Antennas and Propagation
Abstract:On-demand manipulation of electromagnetic (EM) far-fields has greatly accelerated the extension of intelligent metasurfaces in diverse applications. An important prerequisite for most of the effective utilizations boils down to the precise and expeditious determination of far-field. However, the existence of one-to-many mapping between the far-field space and metasurface space renders the inverse-design neural network hard to converge and ill-posed, especially in the case of phaseless far-field. In the work, we introduce an invertible neural network (INN) based generation-elimination framework. By introducing extra trainable latent variables in inverse training, the model can largely mitigate the impact by the phaseless far-field, which is the chief culprit of the one-to-many issue. The INN also provides a powerful stochastically generative capability based on trained latent space. An elimination network is cascaded to INN, where groups of nominated patterns from INN will be screen and eliminate the inferior metasurface patterns. Both simulated and experimental results demonstrate that the far-field match degrees are improved by up to 50%, compared with the case of general deep neural network (DNN), meanwhile retaining a competitive process cycle within 1 ms. This cost-effective and rapid scheme can be deployed on metasurfaces with larger scale to significantly improve signal strength and the quality of wireless communication for Internet of Things (IoT) devices.
What problem does this paper attempt to address?