Beamforming when direction of arrival is random with applications to space diversity processing

This paper presents a beamforming technique using a probabilistic characterization of the direction of arrival (DOA). The underlying distribution of the desired signal DOA can be known a priori or can be completely random with a finite support. The beamforming processor iteratively computes the "a posteriori" probabilities for each assumed DOA and weights each optimum beamformer steered at the assumed direction to arrive at a composite weight vector. The theoretical background is derived and simulations are carried out to illustrate the effectiveness and robustness of this novel beamformer. Finally, an application to diversity processing in CDMA systems is briefly discussed.