Comparative analysis of machine learning and physics-based optimizations of a dual circularly polarized antenna for V2X applications

In this paper, a comparative analysis of Machine learning (ML) and physics-based (PB) optimizations of the monopole fence surrounding a dual circular polarized (CP) antenna, operating in the X-band, is presented and discussed. The aim is to enhance the antenna performances in terms of achieved co-polar gain and cross-polarization rejection at grazing angles, by preserving good impedance matching and input CP port isolation. The results are found to be satisfactory for the specific V2X application, since an acceptable co-polar (LHCP) gain is achieved on a broad angular range with a good cross-polarization (RHCP) rejection, without increasing dramatically the final antenna complexity.