Key points are not available for this paper at this time.
The DAMAS algorithm is compared with the DAMAS2 method and other Fourier-based deconvolution approaches. is the Richardson-Lucy method, is widely used for the deconvolution of images. second is a gradient-type NNLS approach, where procedures are implemented to accelerate the computations. is described how the DAMAS2 and the Fourier-based NNLS can be embedded in outer iteration loop to take the variation of the -spread function into account. methods are tested with synthetic data. first an example with a simple linear array and a small opening angle considered, where the variation of the point-spread function the source region is negligible. this test case the results of the DAMAS2, the Fourier-based NNLS, Richardson-Lucy, and the original DAMAS algorithm are compared. is shown that all algorithms or less introduce oscillations in the reconstructed source. effect of background noise on the results is investigated. second test case with a planar array and large opening angle is presented, to demonstrate the influence a strong variation of the point-spread function. is shown that the methods which neglect the variation of the point-spread lead to distorted results, while embedded Fourier-based methods and the DAMAS algorithm a significantly better reconstruction of the source distribution.
Ehrenfried et al. (Tue,) studied this question.