Estimation of GPS positional accuracy under different forest conditions using signal interruption probability

Abstract

We proposed a signal interruption probability (SIP) value indicating the frequency of global positioning system (GPS) signal interruption or the fragmentation of GPS data, and used it to estimate the success probability of ambiguity resolution in carrier-phase differential GPS (DGPS) as well as the horizontal precision of code-phase DGPS under different forest conditions. In field tests, a rover was set up at four observation points, each representing a different condition within a forest. At each observation point, static surveying was conducted for four periods of 30 min, and data from the first 1, 5, 15, and 30 min from each observation point were extracted for analysis. The position of each point was calculated by baseline analysis using three types of GPS data, i.e., dual-frequency (L1 and L2), single-frequency (L1), and code-phase DGPS data. The results showed that ambiguity resolution was a key factor in determining the horizontal accuracy of carrier-phase DGPS, and that the success probability of ambiguity resolution differed greatly according to forest conditions, which were well explained not by the index of the canopy opening but by the SIP. Furthermore, estimation of the success probability of ambiguity resolution was made based on the SIP and the observation period using a logistic regression model. The horizontal precision of code-phase DGPS was also estimated based on the SIP and the observation period. In conclusion, the SIP was found to be an effective indicator of GPS positional accuracy under different forest conditions.