Machine learning-based technique for resonance and directivity prediction of UMTS LTE band quasi Yagi antenna

• DOI: 10.1016/j.heliyon.2023.e19548
• Publication Date: 2023-09-01T07:00:29Z
• Authors: Md. Ashraful Haque; Dipon Saha; Samir Salem Al-Bawri; Liton Chandra Paul; Md Afzalur Rahman; Faisal Alshanketi; Ali Alhazmi; Ali Hanafiah Rambe; M.A. Zakariya; Saeed S. Ba Hashwan
• Keywords: Directivity

In this study, we have presented our findings on the deployment of a machine learning (ML) technique to enhance performance LTE applications employing quasi-Yagi-Uda antennas at 2100 MHz UMTS band. A number techniques, including simulation, measurement, and model an RLC-equivalent circuit, are discussed in article as ways assess antenna's suitability for intended applications. The CST simulation gives suggested antenna reflection coefficient -38.40 dB 2.1 GHz bandwidth 357 (1.95 GHz-2.31 GHz) -10 level. With dimension 0.535λ0×0.714λ0, it is not only compact but also features maximum gain 6.9 dB, directivity 7.67, VSWR 1.001 center frequency efficiency 89.9%. made low-cost substrate, FR4. RLC sometimes referred lumped element model, exhibits characteristics that sufficiently similar those proposed Yagi antenna. We use yet another supervised regression create exact forecast directivity. models can be evaluated using variety metrics, variance score, R square, mean square error (MSE), absolute (MAE), root (RMSE), squared logarithmic (MSLE). Out seven ML models, linear (LR) has lowest accuracy when predicting directivity, whereas ridge (RR) performs best frequency. strong candidate applications, shown by modeling results from ADS, well measured forecasted outcomes techniques.