Patch Antenna Array Design through Bottom-Up and Bayesian Optimizations

This paper presents the design and the optimization of a 2 × 2 antenna array through two optimization techniques, namely bottom-up optimization (BUO) and Bayesian optimization (BO). These sequential optimizations tackle the difficulty problem of antenna array design by predicting the suitable number of single radiators and by sizing the configuration of feeding circuit using BUO and BO methods, respectively. The proposed method leads to the design of high performance antenna arrays in terms of bandwidth, gain and radiated patterns where all the overall optimization process is performed with the combination of electronic design automation tool and a numerical analyzer. To validate the proposed method, a patch antenna array is designed and optimized in a frequency band from 12.9 GHz to 13.7 GHz with the maximum gain of 17.9 dB in the considered band.