The study reported in this paper is focused on the determination of a proper criterion for the characterization of the level of multipath affecting a particular station. The paper starts with the description of new receiver structures allowing multiple correlator outputs inside a same tracking channel, and highlights the benefits of such designs. From the availability of this data, a new technique is proposed to quantify the deformations affecting the correlation peak. Two approaches are presented to solve this problem. One of them is based on a temporal criterion reflecting the noise variance affecting the correlation function. The second one is based on a spectral analysis and determines the unexpected spectral components in the correlation function. Specific algorithms are presented for the practical evaluation of these two metrics. The paper describes the calibration of these criteria that is performed using a GPS RF signal generator delivering known multipath signals. Validation of the process was carried out through an extensive data collection campaign involving several sites and a number of antennas. Results with both methods enable to establish a consistent ranking of site/antenna pairs. A further refinement of the techniques is considered using finer multipath identification based on observed and predicted multiple correlator outputs.