High-resolution beamforming in the presence of coherent interferences and unknown correlated noise

In many applications of adaptive beamformers, performance of beamformers is degraded due to the presence of correlated interferences. Further, the additive noise is non-white with unknown covariance matrix whereas the beamforming algorithms assume that the noise is white. This situation leads to further deterioration in the performance of the beamformers. Whereas the former problem has been addressed by many authors, the latter has received very little attention. In the paper, a high-resolution beamforming algorithm is presented which is applicable when the interferences are fully/highly correlated with the desired signal and when the noise is spatially correlated with unknown covariance matrix. This algorithm makes use of an invariance property satisfied by persymmetric matrices to cancel out the noise covariance matrix while forming a difference matrix. The columns of this difference matrix are used to form the optimum weight vector. Two types of beamformers are considered. They are a) optimum interference notch beamformer and b) optimal combiner. Simulation results are also presented to support the theory