I.V. Bezmenov, I.Yu. Blinov, A.V. Naumov, S.L. Pasynok

Al’manac of Modern Metrology № 3 (15) 2018, pages 42–91

This article provides a detailed description of well-known method of smoothing the data of code measurements received from dualfrequency GNSS-receivers in the form of RINEX observation files (Receiver INdependent EXchange format). This method is widely used in processing programs for solving the applied problems in the field of geodynamics, time positioning and navigation service, in particular, in well-known Bernese GNSS Software. In the article the noise level evaluation for smoothed code data in the assumption that random components of the measurements are independent and centered values is presented. The method is based on solving the range of sub problems: 1) detection of receiver time jumps, 2) detection and removal of rough measurements (outliers) from data processing and 3) resolution of the carrier phase ambiguities. To solve these problems the techniques based on linear combinations from observation data are used. Conditions under which a receiver time jump is reliably detected against the background of measurement noise and possible jumps in phase ambiguities are formulated. So-called millisecond jumps are considered separately. For outlier detection a new algorithm minimizing a number of unreasonably rejected measurements is proposed. For detection of jumps in the so-called wide-lane phase ambiguity included in Melbourne-Wubbena combination a new algorithm based on formation of so-called clusters is proposed. The results of numerical experiments on comparison of the new algorithms with the algorithms applied in the Bernese GNSS Software are given.

Key words: Global Navigation Satellite Systems (GNSS), smoothing method, code and phase measurement data, linear combinations, phase ambiguities, Melbourne-Wubbena combination, dual-frequency GNSS-receiver, jump.

Full texts of articles are available only in Russian in printed issues of the magazine.