A new classs of GNSS ambiguity estimators.

Abstract

In Teunissen (1999) we introduced the class of admissible integer estimators. Members from this class are defined by their so-called pull-in regions. These pull-in regions satisfy the following three conditions. They are integer translational invariant and cover the whole ambiguity space without gaps and overlaps. Examples of such integer estimators are integer rounding, integer bootstrapping and integer least-squares. In the present contribution we will introduce a new class of GNSS ambiguity estimators. This class is referred to as the class of integer equivariant (IE) estimators since is still obeys the important integer remove-restore principle of integer estimation. It is shown that the IE-class is larger than the class of integer estimators as well as larger than the class of linear unbiased estimators. We will also give a useful representation of IE-estimators. This representation reveals the structure of IE-estimators and shows how they operate on the ambiguity 'float' solution.