Deep neural network-based automatic metasurface design with a wide frequency range

Abstract

Beyond the scope of conventional metasurface which necessitates plenty of computational resources and time, an inverse design approach using machine learning algorithms promises an effective way for metasurfaces design. In this paper, benefiting from Deep Neural Network (DNN), an inverse design procedure of a metasurface in an ultra-wide working frequency band is presented where the output unit cell structure can be directly computed by a specified design target. To reach the highest working frequency, for training the DNN, we consider 8 ring-shaped patterns to generate resonant notches at a wide range of working frequencies from 4 to 45 GHz. We propose two network architectures. In one architecture, we restricted the output of the DNN, so the network can only generate the metasurface structure from the input of 8 ring-shaped patterns. This approach drastically reduces the computational time, while keeping the network's accuracy above 91%. We show that our model based on DNN can satisfactorily generate the output metasurface structure with an average accuracy of over 90% in both network architectures. Determination of the metasurface structure directly without

Deep Learning Approach for Target Locating in Through-the-Wall Radar under Electromagnetic Complex Wall

“…In machine learning, instead of programming everything, data is given to a general algorithm, and the algorithm builds its logic based on the data given to it [14]. With the increasing development of machine learning and the potential of this science to solve various problems, we are witnessing the penetration of this science into electromagnetic problems [15]- [17]. Most machine learning algorithms have a high ability to find hidden patterns and classification of different signals [18].…”