Intelligent Optimization of Antenna Parameters Using Machine Learning Algorithms

Parameter optimization of antennas is an extremely resource-intensive task in contemporary wireless systems design, especially considering the increasing complexity of antenna geometry and tightening constraints in terms of performance at millimeter-wave and terahertz frequencies. In this paper, we consider a systematic study of the application of machine learning algorithms in antenna parameter optimization with a particular focus on surrogate modeling and regression-based ensemble approaches. Our approach involves development of a closed-loop methodology based on generation of electromagnetic simulation data, feature engineering, model selection, and iterative training of a machine learning-based optimizer. We evaluate three algorithm families: tree-based ensembles, neural network surrogates, and kernel methods as possible candidates for predicting antenna metrics based on its geometry. The analysis of optimization process reveals that ensemble algorithms, such as random forest regressor, have very promising prospects for significantly lowering the number of simulations needed during optimization while preserving design accuracy within given parameter spaces. The comparison of algorithms is carried out by means of evaluating their performance according to several criteria: convergence rate, efficiency of data use, robustness to variations in parameters.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-17
DOI
https://doi.org/10.5281/zenodo.24304463
Primary Topic
Antenna Design and Optimization
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Intelligent Optimization of Antenna Parameters Using Machine Learning Algorithms

R.Ananthakrishnan, Aarthi Ananthakirupa V P M B
2 links
Zenodo (CERN European Organization for Nuclear Research)
Antenna Design and Optimization
article

Intelligent Optimization of Antenna Parameters Using Machine Learning Algorithms

R.Ananthakrishnan, Aarthi Ananthakirupa V P M B
article en

Abstract

Parameter optimization of antennas is an extremely resource-intensive task in contemporary wireless systems design, especially considering the increasing complexity of antenna geometry and tightening constraints in terms of performance at millimeter-wave and terahertz frequencies. In this paper, we consider a systematic study of the application of machine learning algorithms in antenna parameter optimization with a particular focus on surrogate modeling and regression-based ensemble approaches. Our approach involves development of a closed-loop methodology based on generation of electromagnetic simulation data, feature engineering, model selection, and iterative training of a machine learning-based optimizer. We evaluate three algorithm families: tree-based ensembles, neural network surrogates, and kernel methods as possible candidates for predicting antenna metrics based on its geometry. The analysis of optimization process reveals that ensemble algorithms, such as random forest regressor, have very promising prospects for significantly lowering the number of simulations needed during optimization while preserving design accuracy within given parameter spaces. The comparison of algorithms is carried out by means of evaluating their performance according to several criteria: convergence rate, efficiency of data use, robustness to variations in parameters.

Zenodo (CERN European Organization for Nuclear Research)
Antenna Design and Optimization
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon