Calibration of intersubarray displacement errors based on Sparse Bayesian Inference

The self calibration of intersubarray displacements errors is addressed in this paper. The array model for a linear array with intersubarray displacement errors is expressed using its first order approximation. The displacement errors and the azimuths of the sources can be estimated through alternating maximizing the log likelihood functions of these parameters, while it needs the prior information of the source number. When the source number is not known, the approximation model is used in the Sparse Bayesian Inference framework, assuming the sources are sparse and the variances of the signal and noise are controlled by hyper-parameters. Numeric simulations illustrate performances of the displacement-error estimation and the azimuth estimation.