Linear sparse array synthesis using Modified Bayesian Optimization Algorithm

In this paper, a new approach for sparse array synthesis is proposed. It is based on the use of a recently introduced improved version of the Bayesian Optimization Algorithm (BOA), named Modified BOA (M-BOA), that has already proved its outperforming capabilities with respect to the standard BOA as well as other well-know optimization approaches. Moreover, in opposite to what is generally done relatively to sparse arrays, here an asymmetric configuration is considered, so that the number of control parameters that could be managed by the optimizer is doubled. The results on several configurations show that the sparse arrays designed with the approach here introduced outperform those synthesized with other techniques. As a proof of concept, the results relative to a 37 element linear sparse array are here presented and discussed.