Nonlinear deconvolution approaches
B. Roy Frieden · The Journal of the Acoustical Society of America · 2005
A review of past methods of digital signal restoration is given. Emphasis is upon nonlinear methods, owing to their extra precision in allowing inequalities and other forms of prior knowledge to be entered as constraints. Some specific methods of this type are maximum entropy, maximum bounded entropy, maximum shannon information, recursive median window filtering, Jansson recursion, and minimum Fisher Information. Algorithms and demonstrations will be given.