Physics-Informed Machine Learning for Efficient Reconfigurable Intelligent Surface Design
This work addresses the efficiency bottleneck in RIS design for wireless communication and radar applications, representing an incremental improvement by integrating existing machine learning methods into the simulation process.
The paper tackles the time-consuming full-wave electromagnetic simulations required for reconfigurable intelligent surface (RIS) design by proposing a machine-learning-assisted approach that combines a multi-layer perceptron neural network with a dual-port network to predict reflection coefficients efficiently, and validates it through fabrication and measurement showing good agreement with simulation results.
Reconfigurable intelligent surface (RIS) is a two-dimensional periodic structure integrated with a large number of reflective elements, which can manipulate electromagnetic waves in a digital way, offering great potentials for wireless communication and radar detection applications. However, conventional RIS designs highly rely on extensive full-wave EM simulations that are extremely time-consuming. To address this challenge, we propose a machine-learning-assisted approach for efficient RIS design. An accurate and fast model to predict the reflection coefficient of RIS element is developed by combining a multi-layer perceptron neural network (MLP) and a dual-port network, which can significantly reduce tedious EM simulations in the network training. A RIS has been practically designed based on the proposed method. To verify the proposed method, the RIS has also been fabricated and measured. The experimental results are in good agreement with the simulation results, which validates the efficacy of the proposed method in RIS design.