Blind deconvolution of backscattered ultrasound using sparse sensor arrays

A method for blind deconvolution of ultrasound and other inverse-scattering problems is presented. The method can be applied to one, two or three dimensional backscattered data. This paper focuses on the one and two dimensional cases. The method is based on a subspace-based blind deconvolution algorithm. In the two dimensional case, the blind deconvolution produces an unknown linear mixture of the reflectivity profile. An entropy minimization algorithm is used to retrieve the reflectivity data from the mixture. This method improves the lateral and axial resolution compared to conventional ultrasound B-scans.