Enhanced multi-layer perceptron model training with ant lion optimization for transmitarray unit cell design
Seckin Karasu, Sibel Unaldi, Muhammed Malkoc
article (Uluslararası, Hakemli) | Yazar(lar): Seckin Karasu, Sibel Unaldi, Muhammed Malkoc | Yıl: 2026 | Dergi/Kurum: PAMUKKALE UNIVERSITY JOURNAL OF ENGINEERING SCIENCES-PAMUKKALE UNIVERSITESI MUHENDISLIK BILIMLERI DERGISI | İndeks: WOS.ESCI | DOI: 10.65206/pajes.90467 | Anahtar Kelimeler: Antenna, Phase, Transmitarra
In the first step of the presented study, a unit cell of the transmitarray antenna is designed, and the transmission phase characteristics are obtained by considering the design parameters. In the second step of the study, two different models, the Multilayer Perceptron (MLP) and AntLion Optimizer (ALO), an enhanced MLP neural network, are developed and trained using the simulated data of the designed unit cell element of the transmitarray. In order to evaluate the performance of the models, root mean square error (RMSE) values were calculated, and the performance of both models was compared. Thus, a high accuracy model was developed for the design parameters and transmission phase of the designed unit cell. The study presents an innovative and effective approach to the design of the unit cell of transmitarray, especially in 5G beyond communication systems, and highlights the potential of optimization and machine learning integration.