Linear antenna array synthesis using gradient-based optimization with analytical derivatives

Synthesis of linear antenna arrays using gradient-based optimization is described. Our approach utilizes standard sequential-quadratic programming algorithm exploiting analytical derivatives of the antenna pattern with respect to the designable parameters, as well as optional initialization through smart random search. This allows us to greatly reduce the computational cost of the synthesis process as compared to population-based techniques such as particle swarm optimization or genetic algorithms. Two linear antenna array design cases are demonstrated.