A modified version of the Cornwell and Evans (1985) maximum entropy deconvolution algorithm is presented. In terms of implementation, the Cornwell/Evans and modified algorithms are very similar. They differ only in the determination of some control parameters. The difference in floating point operations per iteration is insignificant. Nevertheless, the modification couples better the adjustment of the Lagrangian multipliers into the optimization process and results in faster convergence.
No takes yet. Share an insight, caveat, or question.
R. J. Sault (1990) studied this question.