Dana Park

Since the release of Android Nougat (N) version in 2016, Android smartphones have supported GNSS raw measurements. By using GNSS raw measurements, it is possible to calculate exact positions of the receiver through direct calculations. However, smartphones receive satellite signals through Planar Inverted-F Antennas (PIFA) due to their structural limitations. PIFA generally have weak signal reception performance. Furthermore, smartphones allocate limited processing resources to calculate satellite signals, resulting in frequent cycle slips for the carrier measurements of Android smartphones. Further, it is difficult to provide carrier measurement values for each smartphone manufacturer and to conduct Real-Time Kinematic (RTK) studies using carrier measurements from Android smartphones. In this study, we utilize Doppler measurements from Android smartphones to overcome these limitations. Although the quality of Doppler measurements is lower than that of carrier measurements, they are free from cycle slips and have a higher accuracy compared to the code measurements from Android smartphones. We propose that the position accuracy can be improved using a code-Doppler smoothing method that combines the Doppler measurements with code measurements. Additionally, we demonstrate that the accuracy can be enhanced by combining the code-Doppler smoothing method with the Differential Global Positioning System (DGPS) technique. To verify the proposed method, static and dynamic experiments were conducted. To compare the performances during the experiment, a NovAtel device was used as a high-performance receiver, and a Samsung device was used as a low-performance receiver.

Increasing the number of satellites in a global navigation satellite system (GNSS) improves the positioning accuracy and increases availability. However, it reduces the positioning accuracy improvement rate and increases the calculation loads, which can cause battery usage problems in mobile devices using a GNSS. An appropriate satellite selection method is required. One current method entails the use of ideal satellite placement with respect to the minimum geometric dilution of precision (GDOP). In this study, the described ideal satellite placement with the minimum GDOP were divided in terms of the horizontal dilution of precision (HDOP) and vertical dilution of precision (VDOP). HDOP and VDOP were mathematically derived and analyzed. The derived formula was verified using simulations. The analysis was performed with actual dual GNSS satellite data. The satellites adjacent to the ideal placement were selected and the DOP was calculated. Simply selecting satellites closest to the ideal placement afforded large values for HDOP and VDOP. This issue was addressed using a satellite changing algorithm considering the dual GNSS, resulting in reduced values of the HDOP and VDOP.