Regresar

The Optimization of a 5G Inset-fed Patch Antenna Using the Machine Learning Algorithm Surrogate Model Assisted Differential Evolution for Antenna Synthesis

Abstract:

5G is one of the emerging technologies nowadays. An efficient antenna design is a vital component for this technology. This paper presents the optimization of a 5G antenna that operates in 28 GHz frequency band. The rectangular inset-fed patch microstrip antenna that used substrate RogerRT5880 was previously designed in a CST software. Using the design, it was simulated in MATLAB and optimized using the Surrogate model Assisted Differential Evolution for Antenna synthesis (SADEA). To optimize the directivity gain and improve the return loss, the length of the patch, the width of the patch and the width of the notch were modified. The results showed an improvement in directivity gain, return loss and bandwidth. The directivity gain improved to 8.66dBi. The return loss became-19dB. The reflection coefficient was found below-10 dB from GHz to 30.6 GHz which showed an improved impedance bandwidth from 1.1 GHz to 2 GHz. Along with the presentation of the improved results, this paper highlighted the importance of optimization techniques in the field of communications.