Raw GNSS measurements from smartphones
The goals of this Final Year Project are to (1) investigate the application of raw GNSS measurements from dual-band frequency smartphone in Singapore together, (2) to evaluate the BDS, QZSS constellations and L5 frequency impact on accuracy of computed positions and (3) to analyze the open source co...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
Nanyang Technological University
2020
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/140089 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | The goals of this Final Year Project are to (1) investigate the application of raw GNSS measurements from dual-band frequency smartphone in Singapore together, (2) to evaluate the BDS, QZSS constellations and L5 frequency impact on accuracy of computed positions and (3) to analyze the open source code and implement own algorithms of processing raw measurements. Data was collected using the GNSSLogger and analyzed with GnssAnalysisApp, both of which were developed by Google. By evaluating the processed measurements such as pseudoranges, channel-to-noise ratio, computed WLS positions errors, Mi 9 was proved to be a well-built GNSS receiver - smartphone with ability to record all current 5 constellations operating as well as dual-band signal of L1 and L5. From the results, the average errors of static experiments were in the range of 5-10 meters comparing to the ground-truth positions and 5-10 meters comparing to device A-GPS during moving receiver test. Further analysis on BDS constellation proved that it could be a contender of global GNSS provider, showing performance similar to GPS. For QZSS, the study showed that using it did not provide significant increase of accuracy during the experiments. On the other hand, L5 frequency band signal significantly improved the performance of Mi 9 from 20-40% in term of position accuracy comparing to the ground-truth references. Furthermore, in-depth research of GnssAnalysisApp open-source code provided insights to how raw measurements were processed. As a result, multi-constellations data processing and mapping algorithms were proposed to improve the source code. These algorithms can be the backbone for future research or project which utilizes multi-constellation raw measurements from modern smartphones. |
---|