Bounded-uncertainty estimation for correlated signal and noise
D. Lelescu, Frank Bossen · 2006
In this paper we present a class of bounded-uncertainty estimators as the solution of a classic estimation problem involving unknown statistics. The estimators are derived under the non-typical assumption of correlated signal and noise. The boundeduncertainty framework gives an additional degree of freedom for estimator design that can benefit its performance. It also provides an indirect way of verifying hypotheses regarding unknown variable statistics in a particular application domain by examining the behavior of the estimators as a function of the bound(s). If the unknown statistics are within a lower bound than the worst-case limit assumed by a classic minimax estimator, the quality of the estimation is increased by this new approach. I.