Multi-GNSS high-rate RTK, PPP and novel direct phase observation processing method: application to precise dynamic displacement detection

This paper provides the methodology and performance assessment of multi-GNSS signal processing for the detection of small-scale high-rate dynamic displacements. For this purpose, we used methods of relative (RTK) and absolute positioning (PPP), and a novel direct signal processing approach. The first two methods are recognized as providing accurate information on position in many navigation and surveying applications. The latter is an innovative method for dynamic displacement determination with the use of GNSS phase signal processing. This method is based on the developed functional model with parametrized epoch-wise topocentric relative coordinates derived from filtered GNSS observations. Current regular kinematic PPP positioning, as well as medium / long range RTK, may not offer coordinate estimates with subcentimeter precision. Thus, extended processing strategies of absolute and relative GNSS positioning have been developed and applied for displacement detection. The study also aimed to comparatively analyze the developed methods as well as to analyze the impact of combined GPS and BDS processing and the dependence of the results of the relative methods on the baseline length. All the methods were implemented with in-house developed software allowing for high-rate precise GNSS positioning and signal processing. The phase and pseudorange observations collected with a rate of 50 Hz during the field test served as the experiment's data set. The displacements at the rover station were triggered in the horizontal plane using a device which was designed and constructed to ensure a periodic motion of GNSS antenna with an amplitude of similar to 3 cm and a frequency of similar to 4.5 Hz. Finally, a medium range RTK, PPP, and direct phase observation processing method demonstrated the capability of providing reliable and consistent results with the precision of the determined dynamic displacements at the millimeter level. Specifically, the research shows that the standard deviation of the displacement residuals obtained as the difference between a benchmark-ultra-short baseline RTK solution and selected scenarios ranged between 1.1 and 3.4 mm. At the same time, the differences in the mean amplitude of the oscillations derived from the established scenarios did not exceed 1.3 mm, whereas the frequency of the motion detected with the use of Fourier transformation was the same.