Machine Learning Enabled Al2O3Ceramic Based Dual Band Frequency Reconfigurable Dielectric Antenna for Wireless Application

A ceramic (Al2O3) material-based dual-band high-tuning range (TR) frequency reconfigurable (FR) dielectric antenna for wireless applications with a machine learning (ML) algorithm is presented in this article. The proposed antenna is a hybrid structure in which the antenna radiator is designed with a dielectric resonator (DR) (alumina (Al2O3) ceramic material with a relative dielectric constant ( $\in _{r}$ ) =9.8. This work offers dual-band, compactness, and frequency reconfigurability (FR). FR is obtained through two p-i-n diode switches, operating in on-on, on-off, off-on, and off-off configurations. It offers a total spectrum (TS) and a maximum wide TR of 71.49% and 44.44%, respectively. Dual-band is generated through the excitation of ${\text {HEM}}_{{11}\delta }$ , and ${\text {HEM}}_{{12}\delta }$ mode in cylindrical dielectric resonator (CDR). In contrast, compactness is obtained through the higher-order mode excitation and hybrid structure. The proposed antenna is designed on the ANSYS HFSS software and optimized through various ML algorithms such as K-nearest neighbor (KNN), artificial neural network (ANN), decision tree (DT), extreme gradient boosting (XGB), and random forest (RF). In all configurations, KNN achieved more than 99% accuracy for the prediction of the reflection coefficient ( $\text {S}_{{11}}$ ). The proposed antenna is used for WiMAX, WLAN, and 5G wireless applications.