Dimensionality Reduction of a Dual-Polarized Antenna Simulation Data Using Multi-Dimensional Scaling
International Journal of Advanced Research in Science and Technology · 2026 · 15 (10) · 153-162
DOI: 10.62226/ijarst20262851 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advanced Research in Science and Technology, 2026. Verfasst von Ajasa Abiodun Afis und 5 weiteren Autor:innen.
Abstract (Originalsprache)
This study applies multidimensional scaling (MDS) to reduce and visualize a six-dimensional synthetic electromagnetic dataset generated from a dual-polarized antenna with two orthogonally excited ports. The dataset comprised 1,000 samples containing electric-field strength, power-flow density, and surface-current density for each port. Min–max normalization and Euclidean dissimilarity were employed before applying classical MDS. The analysis produced 15 visualizations to examine feature relationships, excitation patterns, clustering, and dimensional structure. The results showed that Port 1 exhibited an average power-flow density approximately four times higher than Port 2, indicating a substantial difference in electromagnetic response between the two excitations. MDS stress decreased markedly when moving from one to two dimensions, while further dimensional reduction provided only marginal improvement, supporting two-dimensional representation as an effective compromise between information preservation and visualization simplicity. The resulting projections revealed distinct data structures, dense regions, gradients, and potential outliers. The findings demonstrate that MDS provides an interpretable approach for exploring complex multidimensional antenna simulation data and identifying dominant electromagnetic patterns. However, validation using experimentally measured antenna data is required to establish the generalizability of the results.
Zitieren
APA
Afis, A. A., Femi, O. S., Ademola, M. Q., O, J. K., Seun, A. O., & Oluwaseun, O. E. (2026). Dimensionality Reduction of a Dual-Polarized Antenna Simulation Data Using Multi-Dimensional Scaling. International Journal of Advanced Research in Science and Technology, 15(10), 153-162. https://doi.org/10.62226/ijarst20262851
Vancouver
Afis AA, Femi OS, Ademola MQ, O JK, Seun AO, Oluwaseun OE. Dimensionality Reduction of a Dual-Polarized Antenna Simulation Data Using Multi-Dimensional Scaling. International Journal of Advanced Research in Science and Technology. 2026;15(10):153-162. doi:10.62226/ijarst20262851
BibTeX
@article{Afis2026,
author = {Afis, Ajasa Abiodun and Femi, Olorunfemi Silas and Ademola, Mumuni Quadri and O, Jimoh Kudirat and Seun, Adeleke Oluwafemi and Oluwaseun, Ogunjobi Elijah},
title = {Dimensionality Reduction of a Dual-Polarized Antenna Simulation Data Using Multi-Dimensional Scaling},
journal = {International Journal of Advanced Research in Science and Technology},
year = {2026},
volume = {15},
number = {10},
pages = {153--162},
doi = {10.62226/ijarst20262851}
}
EUVIMED