Computationally feasible high-resolution minimum-distance procedures which extend the maximum-entropy method
Lee K. Jones, V Trutzer · Inverse Problems · 1989
Maximum-entropy techniques, which have been successfully used to reconstruct a function based on knowledge of a limited number of linear functional values, are extended to a family of geometric reconstruction algorithms which use directed distances and obey a directed orthogonality property analogous to that satisfied by best approximation in Hilbert space. A class of directed distances is introduced leading to numerically efficient reconstruction algorithms with high-resolution capability. Several examples of reconstruction with higher resolution than that for the maximum-entropy method are presented. Some comparisons to, and conclusions about, L p spectrum estimation are also given.