A fast neural beamformer for antenna arrays

A fast neuro-beamformer is presented in this paper. The new algorithm is based on the radial-basis function network. To meet real-time requirement, we customize the basis function for fast computation, and apply a recursive least square learning rule to speed up the network training. By comparing the effects of center location and distribution, we can achieve a minimum network for recalling. The network recalling does not require the knowledge of direction-of-arrival, and thus the method is a blind method.