.png){kind=icon}
.png){kind=icon}
Return
Modeling, Characterization, and Machine Learning Algorithm for Rectangular Choke Horn Antennas
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2024.3394790.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In this work, we present the design and modeling of a new type of choke horn antenna. It incorporates a rectangular waveguide and a rectangular choke acting as a parasitic element. The four-sided geometry of the antenna is applicable to systems that utilize rectangular waveguides. Also, it can overcome the need for rectangular-to-circular transition of transmission line or mode conversion. The main objective of this paper is to develop a model that calculates the far field radiation characteristics of the proposed antenna (analytical part) and to incorporate a finite element method (FEM) solver that adds to the theoretical solution (empirical part), which finally leads to obtaining a hybrid model. The omnidirectional radiation property of the choke is demonstrated, which gives an insight into the influence of this parasitic element on the total radiated power. The same observations made on the rectangular choke can be translated to the circular choke as well. At an operating frequency of 2.45 GHz, the design is numerically and experimentally validated. Also, the demonstrated hybrid model can leverage the integration of supervised machine learning (ML) models by exporting radiation variables such as gain and half-power beamwidth (HPBW) and performing predictions based on the training data from the model. Therefore, we incorporate gradient boosting and neural network ML algorithms, which are tailored to the desired radiation pattern parameters.
Keywords:
Inductors
Antenna radiation patterns
Horn antennas
Electromagnetic waveguides
Vectors
Waveguide components
Machine learning algorithms
Analytical model
beamwidth enhancement
choke horn antenna
empirical model
gain enhancement
gradient boosting
neural network