A fast adaptive low sidelobe wide nulling method for arbitrary DBF array

Antenna array low sidelobe wide null (LSWN) forming and steering are very important in some critical wireless communication and radar applications, especially when used in modern digital beamforming adaptive array systems. In this study, we propose a flexible and fast wide nulling approach based on a neural network (NN) and genetical algorithm (GA). We use the effective Emperor-Selective GA (ENIS-GA) for off-line arbitrary beamforming and we use a radial basis function neural network (RBF-NN) as a fast nonlinear approximate function for on-line fast beamforming. The neural network, which is trained by a finite number of off-line GA solutions, can be used for an infinite number of real time solutions.