Designing broadband cross-polarization conversion metasurfaces using binary particle swarm optimization algorithm

[Display omitted] •The proposed linear polarization discrimination relating to β′ overcomes defects of the traditional methods.•β′ and BPSO algorithm drive the automatic design of cross-polarization conversion metasurfaces.•The designed cross-polarization conversion metasurface is characterized by broadband.•The metasurface is easily fabricated.•The measured results are in excellent agreement with the simulations, which validates the proposed design method. Owing to their disadvantages of broadband discontinuity, infinite value, or incapability of distinguishing cross-polarization conversions from co-polarization conversions, the traditional linear polarization discrimination methods relating to the phase difference Δφ, the ellipticity angle β, or the axial ratio are inapplicable to the software-based optimal design of broadband cross-polarization conversion metasurfaces (CPMs) consisting of subblocks. Therefore, we introduce the pseudo ellipticity angle β′ to characterize the cross-polarization conversion. And based upon the coding metasurfaces and β′, this work first employs the Python-programmed binary particle swarm optimization algorithm to accomplish the optimal design of reflective broadband CPMs. Then, the obtained CPM is further simplified by reprocessing the relatively isolated subblocks, which results in the final CPM. The simulated results show that the final CPM realizes the broadband cross-polarization conversion with β′ ≥ 0.4π and a relative bandwidth of 36.1 % from 11.66 GHz to 16.79 GHz. Moreover, the measured results are in excellent agreements with the simulated ones, which validates the proposed design method. It advances the broadband CPMs design from manual to automated method.