32nd IEEE Conference on Signal Processing and Communications Applications, SIU 2024, Mersin, Türkiye, 15 - 18 Mayıs 2024
Frequency selective surfaces are electromagnetic filters widely used in various industries. The design, optimization, and realization of these filters typically take a long time. Instead of classical optimization methods, the use of deep learning algorithms accelerates the design process and enables obtaining optimal parameters. In this study, a frequency selective surface design with 4 variable parameters operating in the 8-16 GHz range was conducted using the CST Microwave Studio program. Initially, simulations of this design were carried out using the CST Microwave Studio program. However, due to the lengthy simulation processes, an Ensemble Learning algorithm was employed by using the MATLAB program for the optimization processes.