Sparse Bayesian learning for direction-of-arrival estimation with a turning observation platform

When the observation platform turns,the changes of the array orientations will cause spatial spectral peak broadening.To resolve this problem,a Sparse Bayesian Learning(SBL) method is proposed to estimate the source Direction-Of-Arrivals(DOAs).Considering that the spatial-domain sparse signals from different snapshots in geodetic coordinates have the same prior distribution,the proposed method combines multiple received snapshots corresponding to different array orientations in the SBL framework for DOA estimation.Simulation analysis and sea trials data processing show that the proposed method can obtain sharper spatial spectral peaks,and is superior to the existing methods in estimation precision and angle resolution.Besides,it suppresses the fake peaks caused by the left/right ambiguity more effectively.The results indicate that the proposed method can effectively solve the problem of spectral peak broadening,enhanced the DOA estimation performance during the turn,and improve the anti-left/right ambiguity ability by utilizing the changes of the array orientations.