Taguchi-RBF Neural networks Based Optimization of Phased Array Antenna With Coupling Effects

In the antenna array synthesis problems, most of the works in literature utilize isotropic elements. Thus, the mutual coupling effects between the array elements are neglected. It is obvious that an array antenna synthesized by neglecting the coupling effects cannot be used in the real world applications due to the possible mismatch between the desired and realized radiation patterns. In this paper, a novel method based on neural network algorithm RBF (Radial Basis Function ) for the synthesis and model of Antipodal Vivaldi antenna with mutual coupling effect is presented. The synthesis in implementation’s method for this type of array permits to approach the appropriated radiation pattern while considering the mutual coupling between its elements. The neural network is used to estimate the array elements’ excitations. The architecture of the neural network based on the radial basis functions (RBFs) is introduced and simulation results are presented. Results show that there is an agreement between the desired specifications and the synthesized one. The proposed optimization approach offers an efficient and robust synthesis procedure.