Interference subspace approximation based adaptive beamforming in the presence of a desired signal

A novel subspace-based adaptive beamforming method is proposed which is very efficient in computation and can be applied to arrays of arbitrary geometry. In the proposed method, an approximate basis of the interference subspace is obtained from a matrix in which the desired signal component has been removed using the information on its direction. The matrix is updated recursively with data samples and so there is no need for the estimation of the array covariance matrix. A simple direction correction scheme is readily incorporated into the beamformer, so that it is robust to errors in both the look direction and the detected number of the incident signals. The theoretical analysis and simulation results demonstrate the effectiveness of the simple beamforming method.