A novel computational intelligence method for broadband dual-polarized base station antenna design

摘要

Broadband dual-polarized base station antenna requires wide impedancematching bandwidth, high-polarization isolation, stable radiation patterns,small gain variations, and high-cross polarization discrimination level. Forantenna designers, it usually takes much computational time to design broadbandbase station antenna because there are extensive structural parameteranalysis and multiple performance characteristics to be synthesized one byone. In this paper, a novel population-based computational intelligence optimizationmethod is proposed for broadband base station antenna design. Theproposed optimization method is based on self-adaptive weight vector strategyand could speedup searching through adjusting weight vectors and decreasingthe number of full-time objective function evaluations. To verify its high efficiencyof the proposed optimization method, a compact broadband dualpolarizedbase station antenna operating at 3.3–5.0 GHz is designed. Frommeasurement results, it is observed that polarization isolation is higher than27 dB, gain variation is smaller than 0.4 dB, and half-power beamwidth(HPBW) is within 65° ± 3°. In addition, XPD level is better than 24 dB atboresight direction and better than 13 dB within a sector of ±60°. Therefore,the designed antenna is promising for 5G broadband cellular base stationapplications.KEYWORDS