Capturing Method of Millimeter Displacement in Centimeter Class PPP-RTK Measured Data

Abstract

There are more than 5 hundred thousand places with landslide risk in Japan. Especially in recent years, we have been experiencing more and more damages by large-scale landslide by heavy and short-term rain in spot areas, which is caused due probably to the global warming influence. The detection of landslide occurrence beforehand is one of the social problems, but it is thought very hard to solve due to the required accuracy, say millimeter-level, and its cost. The Real Time Kinematic (RTK) positioning using Global Navigation Satellite System (GNSS) is applicable to this measurement from the viewpoint of accuracy. It needs, however, the expensive reference station placed near from RTK positioning sensors, say within several hundred meters, to maintain the millimeter positioning accuracy which easily degrades proportionally to the distance among them. This system therefore thought rather unpromising to this application. Quasi-Zenith Satellite System (QZSS) [1] has a positioning augmentation service named Centimeter Level Augmentation Service (CLAS) [2] which is expected to become a candidate for this application. The service is free of charge, no reference station is needed, and many sensors can be placed in a larger area with a stand-alone configuration with no location-dependent degradation of accuracy. In order to utilize CLAS, however, we must firstly overcome the accuracy problem. Under these circumstances, we studied how to utilize CLAS and its dedicated receiver technology, Real-Time Kinematic Precise Point Positioning (PPP-RTK) [3], to the land slide proactive measurement. QZSS also provides the technical demonstration service of Multi-GNSS Advanced Demonstration tool for Orbit and Clock Analysis (MADOCA) [4].This paper describes; a) our paid attention to the measured data with centimeter-level accuracy taken by PPP-RTK from the viewpoint of “frequency components aggregate”, b) the development of purely mathematical method which expands the original data into orthogonal polynomial and applies the high pass filter technique to retrieve the occurrence timing of millimeter-level displacement which is embedded in the original data, and c) how to obtain the displacement value with a sufficiently accuracy, after having the occurrence timing.