Content uploaded by Reha Metin Alkan
Author content
All content in this area was uploaded by Reha Metin Alkan on Aug 22, 2023
Content may be subject to copyright.
*
Corresponding Author Cite this stud
y
*
(alkanr@itu.edu.tr) ORCID ID 0000-0002-1981-9783
(selbesoglu@itu.edu.tr) ORCID ID 0000-0002-1132-3978
(yavasoglu@itu.edu.tr) ORCID ID 0000-0002-3139-4327
(mutlubil@itu.edu.tr) ORCID ID 0000-0002-9763-0345
A
lkan, R. M, Selbesoglu, O, Yavasoglu, H. H, & Mutlu, B. (2023). Continuous decimeter
s
l
evel real-time Precise Point Positioning in Polar high latitude region. Intercontinenta
l
G
eoinformation Days (IGD), 6, 233-237, Baku, Azerbaijan
6th Intercontinental Geoinformation Days (IGD) –13-14 June 2023 –Baku, Azerbaijan
6th Intercontinental Geoinformation Days
igd.mersin.edu.tr
Continuous decimeters level real-time Precise Point Positioning in polar high latitude region
Reha Metin Alkan 1, Mahmut Oğuz Selbesoğlu 1, Hasan Hakan Yavaşoğlu 1, Bilal Mutlu 1
1Istanbul Technical University, Faculty of Civil Engineering, Department of Geomatics Engineering, Istanbul, Türkiye
Ke
y
words Abstract
Polar region
Marine survey
Real-time Kinematic GNSS
PPP
In this article, the usability and accuracy assessment of the real-time PrecisePoint Positioning
(PPP) technique for dynamic positioning in application areas with difficult atmospheric and
topographic conditions, such as polar regions, has been investigated. Within this frame,
kinematic Global Navigation Satellite System (GNSS) data collected during the ship voyage in
the Southern Ocean region for the 6th Turkish Antarctic Expedition in 2022 was used. The
GNSS data were collected over approximately 9 hours with a 1 Hz sampling rate by tracking
all available satellite constellations. The data were processed using Net_Diff GNSS in-house
processing software with a PPP solution with an ambiguity resolution (PPP-AR solution)
approach using real-time products in the real-time Precise Point Positioning (RT-PPP) mode.
The State Space Representation (SSR) products were retrieved from Centre National D’Etudes
Spatiales / Centre for Space Studies (CNES) Analysis Center. The results showed that the real-
time GNSS PPP solution provided a cm to dm level of accuracy. The overall results obtained
from the study showed that the RT-PPP technique is an alternative to the relative method in
challenging high-latitude regions. The results of the study will also contribute to many
researchers who will work in the polar regions and make a meaningful contribution to the
limited literature.
1. Introduction
Different types of human activities in the polar
regions, including the Arctic and Antarctic regions, are
increasing day by day. As a result, the positioning
performance of GNSS in these regions attracts more
attention, and its use with different approaches is
becoming widespread. The real-time centimetre-level
position information has critical importance in many
scientific and practical applications carried out in these
regions, including precise navigation, hydrographic
surveying, rig positioning, offshore platform survey,
geotechnical surveying, geohazard assessment, marine
construction, pipeline, and cable layout, glacial erosion
monitoring, environmental mapping, and assessment,
seismology, and so on. To fulfill the positioning needs in
these applications, the GNSS technique has been widely
used. The method has gained immense popularity owing
to its ease of application, provided by its high level of
accuracy. GNSS can be used at any time of the day and in
all weather conditions, even in temperatures as low as -
40°C and below. This makes it a highly reliable and
versatile method for positioning and navigation.
Furthermore, almost the 24/7/365-day measurement
capability of GNSS and its ability to operate with minimal
personnel requirements make it a highly efficient and
cost-effective tool. As a result, the GNSS has been widely
adopted across various domains, including land, sea, and
air applications, where accurate positioning and
navigation are critical.
The possible highest level of accurate positioning can
be provided with conventional relative techniques, both
in post-process or real-time modes. Indeed, this method
provides reliable, robust, and highly accurate
positioning. However, in order to apply the relative
method, an additional measurement must be made at
reference station(s) with known coordinates, and an
additional effort needs to transmit the corrections for
real-time applications. In the case of post-processing,
GNSS data processing software and an expert user are
also required. The land-based reference station
requirement is an important limitation in measurements
made more than hundreds of kilometres away from
shores, especially for remote marine areas like open seas
and oceans. As known, the accuracy of the relative GNSS
positioning is affected by the baseline length. Because in
6th Intercontinental Geoinformation Days (IGD) –13-14 June 2023 –Baku, Azerbaijan
234
remote areas such as the polar region, where the lack of
GNSS networks consisting of points with sufficient
density, sparse or low-density reference stations,
environmental constraints in the establishment of new
ones, the number of logistics problems, extreme
atmospheric and weather conditions are limited to the
usability of the conventional relative technique in high
latitude regions. It should be noted that distance-
dependent errors induced by the atmospheric and
orbital errors cannot be minimized by establishing long
baselines in relative technique. In this method, as the
distance between the rover and its reference receiver
increases, the positioning accuracy decreases.
Furthermore, more than thousands of km baseline-
length are too much in order to fix ambiguities for
obtaining cm-level accuracy. Depending on all these
issues, the relative method may turn into an ineffective
method with very low performance in this challenging
geography.
The Precise Point Positioning (PPP) technique, which
has almost become a standard GNSS positioning tool,
stands out as an important alternative with its ease of
application and accuracy close to that of the relative
method. The fact that the method needs precise orbit and
clock data together with GNSS data collected with only
one receiver has made the PPP an effective positioning
tool used in many different fields. The PPP is not affected
by baseline-dependent errors like in relative technique
due to not requiring reference station data. This provides
the usability of this technique anywhere in the world
freely. In remote areas such as the Arctic and Antarctic, it
is much more economical to log raw GNSS observations
and process them with the PPP technique for accurate 3D
positioning rather than using a reference station or a
control network.
However, despite all this, the most important
shortcoming of the PPP technique is that it needs a long
convergence time (typically 20-30 minutes or more) (Qu
et al., 2023). This requirement has been a serious
restriction for the use of the PPP technique in RT
applications (An et al., 2023). However, with the advent
of the ′IGS-RTS Project‵, real-time precise products
started to be produced by many Analysis Centres (ACs)
under the umbrella of IGS. This provides the emerging
real-time PPP (RT-PPP) technique as a combination of
PPP and Real-Time Kinematic (RTK) methods. The RT-
PPP technique has been widely used in many static and
kinematic applications with cm-dm level accuracy by
using only one receiver’s data without the need for any
additional reference station (Abdallah et al. 2016; Wang
et al. 2018; Monico et al. 2019; Di et al. 2020; Di et al.
2022; Savchyn et al. 2023; Yuet al. 2023). Although there
are many studies on the performance of RT-PPP in mid
and lower-latitude regions, there are very few studies on
the real-time positioning performance of the technique in
real marine remote area applications.
In this study, GNSS data collected during a ship
voyage in the Southern Ocean (South Polar Region) were
used. The data were processed in simulation mode (as if
it were a real-time application) using real-time precise
products (precise satellite clock and orbit corrections,
biases, and other necessary data) provided by CNES, and
real-time PPP coordinates were obtained. Finally, the
performance analysis of the method was made by
comparing the PPP-derived coordinates with the
reference trajectory.
2. Field Kinematic Test
2.1. Data collection
To assess the accuracy performance of the RT-PPP
technique, a kinematic GNSS data set collected with a
72.05 m long and 12.50-meter-wide research ship named
Betanzos sailing in the Southern Ocean within the scope
of the 6th Turkish Antarctic Expedition was used. The
GNSS data collected on February 07, 2022, has an
occupation time of approximately 9 hours in a 1-second
measurement interval by observing from all available
GNSS satellites. In the study, CHCNAV i90 Pro geodetic
GNSS receiver was used, which was mounted on the ship
deck. The i90 Pro receiver is capable of GPS, GLONASS,
Galileo, and BeiDou satellites’ observations with its 336
channels. The stated accuracy is given as 2.5 mm + 1 ppm
RMS (pos.), 5 mm + 1 ppm RMS (height) with post-
processing kinematics (PPK) under the open sky, without
multipath, optimal GNSS geometry and atmospheric
condition. In the time period when the measurements
were collected, the ship sailed at an average speed of
appr. 15 kph. The surveying area and kinematic test
measurement are given in Figure 1 (Britannica, 2023and
Google Earth).
Figure 1. The Kinematic test measurement
2.2. Data processing
The real-time PPP coordinates of each measurement
epoch (total 32,638 epochs) were calculated by
processing the kinematic GNSS observations, Ephemeris
Products (CNES/POD Products-CNT), and State Space
Representation (SSR) precise products produced by IGS-
RTS CNES Analysis Centre. The used CNES products
include the real-time orbit, clock corrections, and the
code and phase biases (CNES/POD Products-CNT). All
calculations were carried out with Net_Diff v.1.14
software developed at the GNSS Analysis Centre of
Shanghai Astronomical Observatory, Chinese Academy
of Sciences. Net_Diff supports all signals of GPS,
GLONASS, Galileo, and BeiDou satellite systems
6th Intercontinental Geoinformation Days (IGD) –13-14 June 2023 –Baku, Azerbaijan
235
operating globally, as well as QZSS and IRNSS serving on
a regional basis, for all single, dual, and triple frequency
multi-GNSS observations (Zhang et al. 2020). The
software can process the GNSS data in different ways,
including RTK, PPP, PPP-AR, and PPP-RTK approaches.
The software is open source on GitHub and can be
downloaded at the address of
http://202.127.29.4/shao_gnss_ac/Net_diff/Net_diff.ht
ml.
Although the GNSS data were collected from all
satellite constellations through the measurements, only
a combination of GPS (G), GLONASS (R), and BeiDou (C)
observations were used. As data processing strategy for
RT-PPP solution in Net_Diff v1.14 software, raw code and
phase observations with undifferenced-uncombined
version was processed based on kinematic RT-PPP with
Ambiguity Resolution (AR) in simulation mode.
According to the ambiguity fix percentage ratio results, it
was found that only a few ambiguities were resolved
with a ratio of 33%. Due to this low rate, the PPP-AR
solution was considered to make a limited contribution
to improving the results. It should be noted that the
coordinates of a kinematic trajectory were estimated
with forward and backward processing strategies. The
priori troposphere was modelled based on Saastamoinen
global model. Also, the ionosphere and wet troposphere
were modelled by software, and Vienna Mapping
Function (VMF1) model was used for troposphere
mapping. In order to obtain the highest level of accuracy,
phase wind-up, solid earth tide, and relativistic effects,
corrections were applied, and Kalman Filter was used as
an estimator. Besides, phase center offset values were
obtained from igs14.atx file. Finally, the kinematic
coordinates of the ship were obtained in the ITRF2014
reference frame.
The number of used satellites, PDOP values, and sky
plot were plotted in Figure 2.
In order to demonstrate the performance of the
method and to determine the provided accuracy, the
reference trajectory was calculated within cm-level
accuracy by resolving the carrier-phase ambiguities with
the relative GNSS technique. For this purpose, a
commercial software, CHCNAV Geomatics Office
Software 2.0 (CGO 2.0), was used. The coordinates
obtained from the RT-PPP solution were compared with
the relative solutions (i.e., reference trajectory) for the
horizontal position (2D) and height (Up) components for
each measurement epoch and were plotted as a time
series in Figure 3.
The differences in Figure 3 were summarized with
accuracy measures, i.e., Standard Deviation (Std.Dev.)
and Root Mean Square Error (RMSE). The calculated
values were given in Table 1.
Table 1. Statistical results
∆n
(
cm
)
∆e
(
cm
)
∆2D Pos.
(
cm
)
∆Up
(
cm
)
max. 110 96 140 60
min. -9 -138 0 -106
mean 4 5 16 16
Std.Dev. 7 17 11 14
RMSE 8 18 19 21
Figure 2. Satellite number (up), PDOP (middle), and sky
plot (bottom)
3. Results and Discussions
In general, multi-constellation PPP significantly
increased the number of satellites, and this improved the
availability and reliability of positioning results (Zhao et
al., 2022). This was also the case in our study. According
to Figure 2, it was seen that the number of satellites for
the multi-GNSS solution (GRC solution) increased
significantly compared to the single-constellation
systems. The mean number of observed satellites were
found as 11, 8, 8, and 27 for G-alone, R-alone, C-alone, and
GRC combination, respectively. In general, it was seen
that the number of satellites and the PDOP value were
inversely proportional; in other words, the PDOP value
6th Intercontinental Geoinformation Days (IGD) –13-14 June 2023 –Baku, Azerbaijan
236
improved as the number of satellites increased. In the
high latitude areas, the satellites were generally
observed at lower elevation angles and over a shorter
continuous period due to the GNSS orbit characteristics.
Thus, the multi-GNSS observations increased the signal
availability, and depending on this, positioning accuracy
and reliability were improved.
Figure 3. Comparison of the coordinates between RT-
PPP and relative GNSS solutions
The results in Figure 3 showed that the multi-GNSS
RT-PPP solution produced centimetre to decimetre level
coordinate differences after the convergence period. The
mean differences were found 16 cm for both 2D position
and height components. Although this performance was
generally well enough in such challenging conditions, the
RT-PPP positioning accuracy in the high-latitude polar
regions was found worse than that of medium and low-
latitude regions. The most likely reasons for this were the
configuration of the observed satellites, spatial
geometric distribution of the observed GNSS satellite
configuration, observing the satellites at low elevation
angles, severe weather conditions, and atmospheric
effects (mainly ionospheric effect).
Concerning accuracy, the RMSE of the RT-PPP
solution were found as 19 cm in 2D horizontal and 21 cm
in height components. As depicted in Table 1, the
Standard Deviations (Std.Dev.) were found slightly better
than the RMSE values as 11 cm and 14 cm for 2D
horizontal and height components. The RMSE and
Std.Dev. of the north-south components were found
better than that of the east-west.
Due to the fact that PPP does not need base station
data, the baseline length bias is not an issue like in the
relative technique. Unlike the relative positioning
method, the RT-PPP technique does not require
additional GNSS data. So that, it has become a viable
alternative to the conventional relative GNSS method.
4. Conclusion
The main goal of this paper was to evaluate the
performance of the real-time PPP (RT-PPP) technique in
high-latitude areas. For this purpose, a realistic
kinematic test was carried out in the Southern Ocean, and
9 hours of kinematic GNSS data was collected. The
dataset was processed with Net_Diff GNSS processing
software in real-time mode. The overall results obtained
from the study showed that the RT-PPP technique
provides a centimetre to decimetre level of positioning
accuracy in dynamic applications, especially for remote
marine applications, without the need for additional
GNSS reference data efficient and cost-effective way.
These attainable accuracies satisfied the accuracy
requirement of many real-time and kinematic remote
marine and related applications, including the IHO, IMO
accuracy standards, and others.
The overall results demonstrated that the RT-PPP
technique can be successfully used for dynamic
positioning under difficult atmospheric and
topographical measurement conditions of polar high-
latitude regions. It was clearly seen that the RT-PPP
technique produces 3D position faster, allowing users to
reduce the operational costs and time, and thus also the
carbon footprint of each project.
Acknowledgement
The authors gratefully acknowledge the IGS and CNES
Analysis Centre for providing real-time products. Dr. Yize
Zhang, the developer of Net_Diff software from the
Shanghai Astronomical Observatory, China, is
appreciated by the authors for providing the software.
This study was carried out under the auspices of the
Presidency of the Republic of Turkey, supported by the
Ministry of Industry and Technology, and coordinated by
TUBITAK MAM Polar Research Institute.
References
Abdallah, A. M., Abd Rabbou, M., & El-shazly, A. (2016,
April). Performance evaluation of real time precise
point positioning (RT-PPP) in static & kinematic
modes in Egypt. In Proceedings of the Arab Institute
of Navigation (AIN) 2016 Melaha International
Conference and Exhibition (pp. 25-27).
An, X., Ziebold, R., & Lass, C. (2023). From RTK to PPP-
RTK: towards real-time kinematic precise point
positioning to support autonomous driving of inland
waterway vessels. GPS Solutions, 27(2), 86.
https://doi.org/10.1007/s10291-023-01428-2
Britannica, (2023). The editors of encyclopaedia.
"Southern Ocean", Encyclopaedia Britannica,
https://www.britannica.com/place/Southern-
Ocean. Accessed 31 May 2023.
6th Intercontinental Geoinformation Days (IGD) –13-14 June 2023 –Baku, Azerbaijan
237
Di, M., Zhang, A., Guo, B., Zhang, J., Liu, R. & Li, M. (2020).
Evaluation of real-time PPP-based tide measurement
using IGS real-time service. Sensors, 20(10), 2968.
https://doi.org/10.3390/s20102968.
Di, M., Guo, B., Ren, J., Wu, X., Zhang, Z., Liu, Y., Liu, Q. &
Zhang, A. (2022). GNSS real-time precise point
positioning in Arctic Northeast Passage. Journal of
Marine Science and Engineering, 10(10), 1345.
https://doi.org/10.3390/jmse10101345
Monico, J.F.G., Marques, H.A., Tsuchiya, Í., Oyama, R.T., de
Queiroz, W.R.S., de Souza, M.C. & Wentz, J.P. (2019).
Real time PPP applied to airplane flight tests. Boletim
de Ciências Geodésicas, 25(2).
https://doi.org/10.1590/s1982-
21702019000200009.
Qu, L., Du, Y., Wang, H., Jiang, W. & Wang, L. (2023). Multi-
constellation and multi-frequency precise point
positioning with fast ambiguity resolution on a global
scale. Measurement, 211, 112642.
https://doi.org/10.1016/j.measurement.2023.1126
42
Savchyn, I., Brusak, I. & Tretyak, K. (2023). Analysis of
recent Antarctic plate kinematics based on GNSS data.
Geodesy and Geodynamics 14(2), 99-110.
https://doi.org/10.1016/j.geog.2022.08.004
Wang, Z., Li, Z., Wang, L., Wang, X. & Yuan, H. (2018).
Assessment of multiple GNSS real-time SSR products
from different analysis centers. ISPRS International
Journal of Geo-Information, 7(3), 85.
https://doi.org/10.3390/ijgi7030085
Yu, C., Zhang, Y., Chen, J., Chen, Q., Xu, K. & Wang, B.
(2023). Performance assessment of multi-GNSS real-
time products from various analysis centers. Remote
Sensing, 15(1), 140.
https://doi.org/10.3390/rs15010140
Zhang, Y., Chen, J., Gong, X. & Chen, Q. (2020). The update
of BDS-2 TGD and its impact on positioning. Advances
in Space Research, 65(11), 2645-2661.
https://doi.org/10.1016/j.asr.2020.03.011
Zhao, J., An, J., Wang, Z., Ai, S., Zhu, L., & Zhang, B. (2022).
Signal quality and positioning performance of
GPS/BDS-3/GLONASS/Galileo in polar regions.
Advances in Space Research, 69(6), 2541-2554.
https://doi.org/10.1016/j.asr.2021.12.032
