The applications of the observer-based beamforming algorithm for aeroacoustics

Beamforming based on microphone arrays is widely used in aerospace industry for localizing the distribution of noise sources of an aircraft. However, the conventional beamforming computations can only be conducted off-line after the preparations of all of the sampling data, thus cannot be used as a real-time algorithm. In addition, the performance of conventional beamforming deteriorates when there exists coherent background noise. An alternative algorithm is proposed to solve these problems. The algorithm has a similar form to a state observer in control theory, while operates in the frequency domain. The derivations and various validations of this algorithm is included in this work. It is shown that this so-called observer algorithm not only holds the real-time capability, but can also resolve the coherent restriction between background noise and the signal of interest during the iterations. With the help of this algorithm, any testing defects can be revealed instantaneously and corrected on site.