Conference paper

Estimation in the presence of multiple sources of uncertainties with applications

We develop an estimation technique for problems that involve multiple sources of uncertainties or errors in the data. The method allows the designer to explicitly incorporate into the problem formulation bounds on the sizes of the uncertainties; thus leading to solutions that will not over-emphasize the effects of the uncertainties beyond what is assumed by the prior information. Applications in array signal processing and image processing are considered.


Related material