New Methods for Achieving Advanced Accuracy in Time and Coordinate Measurements

I. V. Bezmenov, I. Yu. Ignatenko, S. L. Pasynok

Transactions of IAA RAS, issue 60, 12–20 (2022)

DOI: 10.32876/ApplAstron.60.12-20

Keywords: GNSS, SLR, preprocessing, normal points

Abstract

Within project GGOS, IAG proclaimed the goal of achieving millimeter accuracy in determining the coordinates of reference points. To achieve such accuracy, it is necessary to equip measurement points with new generation RSDB, GNSS and SLR. However, since the accuracy targets were at the boundary of the capabilities of modern measuring instruments, it was also necessary to improve models and methods for both post processing of measurements and their preprocessing. During the preprocessing of measurements, the task of rejecting outliers — the results of rough measurements — arises. This problem is closely related to the problem of finding an unknown trend (usually polynomial) in the data measured. Obviously, an incorrect trend definition can lead to incorrect outlier detection, and good measurements can be rejected and inaccurate ones can be retained, what ultimately negatively affects the accuracy of the final result. Another problem is the detection of coarse measurements in data with trend removed. One object in creating detection algorithms in this case is to minimize the amount of the data rejected. Moreover, in the case of GNSS measurements, the loss of some part of the data at the preprocessing stage may not have a significant effect on the final result due to the huge number of navigation measurements. At the same time, when processing satellite laser rangefinder (SLR) measurements, each individual measurement is important. To solve the problem of automatic detection of the rough measurements at the stage of their preprocessing work, some algorithms are developed by I. V. Bezmenov, S. L. Pasynok, and others were involved. Initially, they were designed to reject coarse GNSS measurements and were more effective in comparison with other known methods (fewer computational operations, fewer rejected points with the same rejection threshold). As a result of using these algorithms for the preprocessing of laser measurements, I. Yu. Ignatenko managed to achieve greater accuracy of formed normal points with less amount of data rejected. The effectiveness of the proposed algorithm in preprocessing SLR measurements under conditions of abnormal atmospheric refraction was also shown. The description of some algorithms used in the paper and their mathematical theoretical foundation are presented in more details in Chapter 9 “Effective Algorithms for Detection Outliers and Cycle Slip Repair in GNSS Data Measurements” (Bezmenov, 2021) published in English in the collection “Satellite Systems: Design, Modeling, Simulation and Analysis”.

Citation