An analytical study of ambiguity decorrelation using dual frequency code and carrier phase

In this contribution the integer least-squares estimation of the double differenced L(1) and L(2) ambiguities is analyzed, under the provision that the relative receiver-satellite geometry is dispensed with. The variance-covariance matrix of the ambiguities is instrumental for gaining insight into the characteristics of the ambiguity fixing process. A qualitative geometric description in detail is therefore given of the ambiguity search space. The elongation, the correlation coefficient and the areas of the ambiguity search space and its enclosing boxes are all given as function of the ratio or product of the carrier phase and code standard deviations. It is shown that the ambiguity search space is very elongated and that the ambiguities are highly correlated. It is also shown how the high correlation between the ambiguities can be used to ones advantage for the transformation to new ambiguities. This is done by means of the least-squares ambiguity decorrelation adjustment method. The improvements in terms of decorrelation, elongation and precision are shown and the corresponding optimal time-invariant ambiguity transformations are given for a practical range of code and measurement precisions.