Machine learning and regression models for Reconfigurable Intelligent Surface optimization

With the introduction of Reconfigurable Intelligent Surfaces (RIS) there is an increasing need for their optimization so that their full potential can be exploited. This study aims to generate an artificial intelligence model that can optimally configure a (RIS). Four regression models were produced, they were derived using data generated from Finite Difference Time Domain (FDTD) simulations as input and output for 'machine learning. In conclusion it is possible to produce RIS configurations by regression models with values very close to the desired ones, thus laying the foundation for the introduction of artificial intelligence in the approach to optimization and configuration of RIS.