When monitoring deformations by means of different sensors, one has to be sure that the various observations do see the same variations in time of the earth surface. As an example one can think of a deformation as seen by the SAR technique and the deformation of the same surface as seen by GPS. To this aim a hypothesis testing procedure has to be set up (Koch 1999). The first question is how to compare the different data sets, which usually do not refer neither to the same positions in space nor to the same time. The standard prediction of one set of variables from the other, for instance, is not always the best solution. It is better to use both observation sets to predict one and the same functional of the “random field” describing the deformation pattern and to evaluate the difference between the two predictions. This difference has to be small on condition that the signal we try to estimate has a fixed amplitude in mean quadratic sense. The problem is formally solved and a few examples are illustrated.
Optimal cross-validation of different surveying techniques
GILARDONI, MADDALENA;SANSO', FERNANDO;VENUTI, GIOVANNA
2012-01-01
Abstract
When monitoring deformations by means of different sensors, one has to be sure that the various observations do see the same variations in time of the earth surface. As an example one can think of a deformation as seen by the SAR technique and the deformation of the same surface as seen by GPS. To this aim a hypothesis testing procedure has to be set up (Koch 1999). The first question is how to compare the different data sets, which usually do not refer neither to the same positions in space nor to the same time. The standard prediction of one set of variables from the other, for instance, is not always the best solution. It is better to use both observation sets to predict one and the same functional of the “random field” describing the deformation pattern and to evaluate the difference between the two predictions. This difference has to be small on condition that the signal we try to estimate has a fixed amplitude in mean quadratic sense. The problem is formally solved and a few examples are illustrated.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
