Quality analysis of the low-cost GNSS receiver pseudorange data.

Abstract

The last decade has brought us rapid developments in low-cost GNSS receiver technology. The performance level of such receivers can be unexpectedly close to that offered by high-grade geodetic ones. Nevertheless, the further growth of applicability for these types of devices still needs a more detailed analysis of observations. Motivated by such developments, we focused on the quality of multi-GNSS pseudorange data acquired by three low-cost receivers: u-blox ZED-F9P, Septentrio Mosaic-X5, and Skytraq PX1122R and its comparison with reference values adopted from the geodetic receiver - Trimble Alloy. We investigated two main characteristics of code data – observation noise and correlation in the time domain. In contrast to a typical pseudorange data analysis based on a single-receiver scenario performed with multipath combinations, we extended our tests with data acquired at zero-baseline formed of homogeneous receiver pairs. Such an approach allowed us to analyze the quality of more realistic data, i.e., affected by multipath, and data with multipath removed through between-receiver differencing. The investigations revealed significant discrepancies in data quality between selected low-cost receivers and recorded signals. Generally, the pseudorange noise was the lowest for Septentrio Mosaic-X5, whereas the noisiest observations were found for Skytraq PX1122R. More interestingly, considering the deviation of code data, Septentrio Mosaic-X5 outperformed the geodetic receiver. On the other, we noted that the low-cost receivers' measurements are noticeably correlated in the time domain, even for between-receiver differences of multipath combinations.