An approach to blind adaptive beamforming is presented based on the property of Doppler signals. The weights of the beamformer are Selected according to a new optimization Objective

which consists of minimizing the mean square error between the array output and a unit complex exponential. It is demonstrated via analysis and simulations that this method is close to MVDR with array manifold known exactly

which has maximum SINR

and has the advantage of small computation and rapid convergence rate.