A Comparative Analysis on Inter-frequency Data Quality of Quad-constellation GNSS on a Smartphone

Abstract

This paper comprehensively analyzes the inter-frequency data quality of the quad-constellation Global Navigation Satellite System (GNSS) of GPS, GLONASS, BDS and Galileo on a smartphone. A series of indices, i.e. the number of visible satellites, data integrity rate, multipath, carrier-to-noise ratio (C/No), cycle-slip ratio and observation residuals, are employed to evaluate the data quality with a comparison between different constellations and frequencies. Experiments were conducted using the firstly released dual-frequency smartphone of Xiaomi Mi8. The results show that the GPS and BDS exhibit the best tracking performance in an open-sky environment with an average of 7 observed satellites at each epoch, which is 3 or 4 satellites more than the Galileo and GLONASS. In addition, the GPS data integrity rate is higher than the other constellations by about 20%-25%. The GPS suffers a multipath effect two times larger than the Galileo on the L1/E1 frequencies, but they are almost equal on the L5/E5a frequencies. For all four constellations, the C/No is mostly concentrated at 20-35 dB-Hz. Further, the C/No on the L1/E1 frequencies increases by 3-4 dB-Hz over the L5/E5a frequencies. The GLONASS observations exhibit the most serious cycle slip occurrence rate at a ratio of 100, which is significantly larger than the other constellations. Regarding the residuals, the phase RMS residuals for all four constellations are at a few millimeters, whereas the pseudorange residuals of GLONASS are the most prominent with an RMS of over 6 m, which is 3-4 times larger than the other constellations. The precise point positioning (PPP) results show that the convergence time and positioning accuracy can be effectively improved by adding GPS and Galileo data at L5/E5a.