Linear Array Synthesis Based on Artificial Neural Network Agent Model with Fast Fourier Transform

This paper introduces a new array synthesis method based on back propagation (BP) neural network agent model with a training set quickly generated by fast Fourier transform. Moreover, it can be conveniently generalized to array models on different scenarios. A test example for synthesizing uniformly spaced linear array with non-uniform excitations is conducted. Synthesis results prove the effectiveness and superiority of the proposed method.