Numerical Investigations to Improve the Reduced-Order Model Approach for Antenna Measurements - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Numerical Investigations to Improve the Reduced-Order Model Approach for Antenna Measurements

Résumé

The determination and validation of the radiation properties of antennas are necessary tasks in their development. It often calls for the sophisticated and costly evaluation of the 3D radiation pattern. The Reduced-Order Model (ROM) has been successfully applied in antenna measurements to reduce the required number of sampling points over the scan surface, leading to faster field acquisitions. It is constructed by the Singular Value Decomposition (SVD) of the radiation matrix computed from the equivalent, or Huygens, surface principle, using the method of moments. In this article, we propose several studies aiming for a better and easier use of this technique as it deeply relies on numerical computations. In particular, the evaluated number of degrees of freedoms is investigated relatively to the dimensions of the equivalent surface, the area of the observation region and the stability with respect to various parameters. Also, the use of the Randomized Singular Value Decomposition (RSVD) to speed up the computation time of the ROM is studied. Validations based on both simulation and experimental data are shown.
Fichier principal
Vignette du fichier
AMTA22_Numerical Investigations to Improve the Reduced-Order Model Approach for Antenna Measurements.pdf (10.18 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04643189 , version 1 (10-07-2024)

Identifiants

  • HAL Id : hal-04643189 , version 1

Citer

Samuel Corre, Nicolas Mézières, L. Le Coq, Benjamin Fuchs, Michael Mattes. Numerical Investigations to Improve the Reduced-Order Model Approach for Antenna Measurements. AMTA 2022, Oct 2022, Denver (Colorado, USA), United States. ⟨hal-04643189⟩
5 Consultations
4 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More