Design of high performance compact linear ultra-wideband arrays with the CMA evolutionary strategy

Several small to moderate ultra-wideband antenna arrays with impressive performances have been created by directly optimizing the element spacings using the covariance matrix adaption evolutionary strategy. Because of the capabilities of CMA-ES, these problems with large parameter sets (e.g. 49 parameters for a 50 element array) are able to be successfully tackled to produce fruitful designs. Even though these arrays were optimized at d_min = 4.0 λ for an 8:1 frequency bandwidth, their useful bandwidth often extends well beyond this. These results demonstrate that the CMA-ES method represents a powerful new addition to the optimization toolbox for the antenna array synthesis community.