Deep-learning-assisted reconfigurable metasurface antenna for real-time holographic beam steering

Hyunjun Ma,Jin-soo Kim,Jong-Ho Choe,Q-Han Park

DOI: https://doi.org/10.1515/nanoph-2022-0789

2024-06-19

Abstract:We propose a metasurface antenna capable of real time holographic beam steering. An array of reconfigurable dipoeles can generate on demand far field patterns of radiation through the specific encoding of meta atomic states. i.e., the configuration of each dipole. Suitable states for the generation of the desired patterns can be identified using iteartion, but this is very slow and needs to be done for each far field pattern. Here, we present a deep learning based method for the control of a metasurface antenna with point dipole elements that vary in their state using dipole polarizability. Instead of iteration, we adopt a deep learning algorithm that combines an autoencoder with an electromagnetic scattering equation to determin the states required for a target far field pattern in real time. The scattering equation from Born approximation is used as the decoder in training the neural network, and analytic Green's function calculation is used to check the validity of Born approximation. Our learning based algorithm requires a computing time of within in 200 microseconds to determine the meta atomic states, thus enabling the real time opeartion of a holographic antenna.

Computational Physics,Machine Learning,Data Analysis, Statistics and Probability,Optics

What problem does this paper attempt to address?

The problem this paper attempts to address is: how to achieve real-time holographic beam steering with reconfigurable metasurface antennas without mechanically moving the antenna elements.

Specifically, traditional beam steering methods such as phased array antennas require complex electronic equipment and high power consumption, while reconfigurable metasurface antennas control the wavefront by adjusting the states of unit elements to achieve beam steering. However, the methods for determining the states of these unit elements usually rely on iterative optimization algorithms (such as genetic algorithms, particle swarm optimization methods, etc.), which are time-consuming and require recalculation each time.

To solve this problem, the authors propose a deep learning-based method that combines autoencoder neural networks and electromagnetic scattering equations to determine the required states of unit elements in real-time, thereby achieving efficient and fast holographic beam steering. This method not only significantly improves the computation speed but also generates high-quality holographic far-field patterns within milliseconds.