Microstrip line fed dielectric resonator antenna optimization using machine learning algorithms

In this communication, a microstrip line fed dielectric resonator antenna is optimized using various Machine learning-based models. Different ML algorithms such as ANN (artificial neural network), KNN (K-Nearest Neighbors), XG Boost (extreme gradient boosting), Random Forest, and Decision Tree are used to optimize the proposed antenna design within the frequency band 3.3-3.65 GHz. |S-11| of the proposed antenna is predicted by using various ML algorithms. Dataset for the same is created through HFSS EM (Electromagnetic) simulator by varying the radius, height of DRA (Dielectric Resonator Antenna) as well as the width of microstrip line and conformal strip. Predicted results from all these models are quite close to the actual one except ANN. To overcome the problem of ANN, Knowledge-Based Neural Network techniques (KBNN) are implemented. All these ML algorithms are authenticated by practically constructing and measuring the proposed antenna. Fabricated antenna results are in good agreement with the values predicted by ML algorithms.