On the use of the angle of incidence in support vector regression surrogate models for practical reflectarray design

A common approach in the literature when obtaining surrogate models of reflectarray unit cells is to include, among other variables, the angles of incidence as input variables to the model. In this work, we use support vector regression (SVR) to compare this approach with a new strategy which consis...

Descripción completa

Detalles Bibliográficos
Autores: Rodríguez Prado, Daniel|||0000-0002-2774-7572, López Fernández, Jesús Alberto|||0000-0001-7603-9591, Arrebola Baena, Manuel|||0000-0002-2487-121X, Goussetis, George
Tipo de recurso: artículo
Fecha de publicación:2021
País:España
Institución:Universidad de Oviedo (UNIOVI)
Repositorio:RUO. Repositorio Institucional de la Universidad de Oviedo
Idioma:inglés
OAI Identifier:oai:digibuo.uniovi.es:10651/57773
Acceso en línea:http://hdl.handle.net/10651/57773
Access Level:acceso abierto
Palabra clave:Machine learning
surrogate model
support vector regression (SVR)
angle of incidence
reflectarray antenna
Descripción
Sumario:A common approach in the literature when obtaining surrogate models of reflectarray unit cells is to include, among other variables, the angles of incidence as input variables to the model. In this work, we use support vector regression (SVR) to compare this approach with a new strategy which consists in grouping the refletarray elements under a small set of angles of incidence and train surrogate models per angle of incidence pair. In this case, the dimensionality of the SVR decreases in two with regard to the former approach. In both cases, two geometrical variables are considered for reflectarray design, obtaining 4D and 2D SVRs, respectively. In contrast to the common approach in the literature, the comparison between the 4D and 2D SVRs shows that a proper discretization of the angles of incidence is more competitive than introducing the angles as input variables in the SVR. The 2D SVR offers a shorter training time, faster reflectarray analysis, and a similar accuracy than the 4D SVR, making it more suitable for design and optimization procedures.