Optimized Artificial Neural Network for reflectarray cell modelling

V. Richard (1) , Renaud Loison (1) , R. Gillard (1) , H. Legay , M. Romier

1 IETR - Institut d'Électronique et des Technologies du numéRique

This paper proposes a design methodology to optimize Artificial Neural Networks (ANN) modelling reflectarray cell. It is applied to a Phoenix cell with 5 inputs parameters. The results demonstrate that the final ANN model is reliable and accurate with an average phase error of 1,4°. © 2016 IEEE.

Domaines: Réseaux et télécommunications [cs.NI]; Electronique

V. Richard, Renaud Loison, R. Gillard, H. Legay, M. Romier. Optimized Artificial Neural Network for reflectarray cell modelling. 2016 IEEE Antennas and Propagation Society International Symposium, APSURSI 2016, Jun 2016, Puerto Rico, United States. pp.1211--1212, ⟨10.1109/APS.2016.7696313⟩. ⟨hal-01416324⟩