Error analysis of the pruned multi-angle resolution beamformer

Recently, we proposed the pruned fast beamformer that is able to detect coherent signals during beamforming. The fast beamformer is a multistage approximation to delay and sum beamforming. Using a coherent signal detector, the pruned fast beamformer is able to reduce the computational load by eliminating processing to resolve steering directions that do not contain a target-like source. This paper analyzes the performance of the detector and discusses how approximation errors in the fast beamformer effects detection performance. The results indicate that the coherence detectors is effective under the approximation errors.