Precise Point Positioning and its implications

Ying Bo Wang¹

¹RMIT University, Melbourne VIC 3001, Australia

Unless otherwise indicated, diagrams included in this report were created by Ying Bo Wang, RMIT University 2013

Introduction

In the past, Point Positioning and Differential DGPS positioning have been predominantly used for mapping and geodetic positioning respectively. However, a relatively new technique referred to as Precise Point Positioning (PPP) has been introduced and promises to improve and replace Point Positioning (Holden 2013b). This paper aims to investigate this claim as well as PPP’s effects on DGPS usage. It will (i) go through key concepts behind Point Positioning, DGPS and PPP and (ii) compare PPP against Point Positioning and DGPS.

The techniques in this report are described in the context is in the United States Department of Defense’s Global Positioning System (GPS). However, the key concepts behind each technique can be applied to GLONASS, GALILEO or any other Global Navigation Satellite System (GNSS).

Point Positioning

Point Positioning is the basic way of determining a receiver’s position through use of the C/A (Course/Acquisition) code and the broadcast navigation message. The broadcast message contains the GPS date and time, the broadcast ephemeris and the almanac. Hofmann-Wellenhof, Lichtenegger and Walse (2008) state that the almanac is to help the receiver acquire the satellite while the broadcast ephemeris is to compute a reference orbit for the satellites. It requires a minimum of 4 satellites to generate 4 different pseudo-ranges to the receiver. This is represented by Figure 1 below. From there, the position of the receiver (X, Y, Z coordinates) will be calculated with a horizontal accuracy of approximately 10 metres.

Figure 1: Point Positioning using 4 satellites

The main limitation of Point Positioning is its poor accuracy, due to not accounting for the receiver clock error, satellite clock error, satellite orbit error and atmospheric errors. This makes it unsuitable for applications that require very accurate positions, like cadastral surveys. The advantages are that it only requires tracking a single frequency, L1 for GPS signals, and a low-cost handheld GPS unit. The position solution is also very quickly derived, making it real-time.

Differential Positioning using PRN code (DGPS)

Holden (2013a) asserts that “DGPS” is differential positioning using pseudo-random (PRN) code while “Differential Positioning” refers to differential positioning using carrier signals. The focus in this report will be DGPS.

DGPS calculates the position of one receiver, termed the rover, relative to another receiver, termed the base. The base receiver is set up on a point with known coordinates, like a survey mark. Knowing the true coordinates, the base receiver will then apply corrections to the rover receiver (Australian Maritime Safety Authority 2013). These corrections remove the receiver clock error and atmospheric errors, bringing the horizontal accuracy of measurements to 2-3 metres. Both receivers must track the same satellites during this time and must not be separated by more than 10 kilometres.

Figure 2: Key concept behind DGPS

Holden (2013) states that DGPS can be applied to either static receivers, whose data is differentially post-processed, or in-motion receivers, whose differential corrections are applied real-time. To facilitate DGPS, various companies and government agencies have set up Continually Operating Reference Systems (CORS). CORS is a network of GNSS receivers that are in a fixed location, and function as base receivers. Users would acquire the CORS data for the base station that suits their survey, and differentially process with their rover measurements (Holden 2013).

The main limitation of DGPS is that it requires 2 separate receivers at 2 different locations, where one location is a point with known coordinates. This could be an issue in areas with low number of established survey marks. These 2 receivers must also be tracking the same satellites for the same time period. If either of the 2 receivers loses the satellite signal, the rover’s position for that time epoch cannot be differentially processed.

Precise Point Positioning (PPP)

PPP uses accurate orbital data and accurate satellite clock data to build models for calculating a single receiver’s position (Witchayangkoon 2000, p2). The lone receiver can be either single- or dual-frequency, with no need for an additional receiver to be a ‘base’ station. This is illustrated below in Figure 3.

Figure 3: Key Concept behind PPP

Using precise International GPS Service (IGS) orbit and clock products, Kouba and Héroux (2001) found that global centimetre-level accuracy was possible with a dual-frequency lone receiver. The overall process can be illustrated below in Figure 4. A similar accuracy can also be achieved in real-time, with a dual-frequency lone receiver (Chen 2004).

Figure 4: PPP workflow

The main difference between single- and dual-frequency PPP is the fact that single-frequency PPP does not account for ionospheric effect. This traditionally results in single-frequency PPP to achieve an accuracy of several meters while dual-frequency having centimetre accuracy (Chen and Gao 2005). There are several ongoing research efforts into achieving centimetre accuracy using only a lone single-frequency receiver, in real-time. When this is achieved, the effect could be as significant as when United States turned off Selective Availability in May 2000 (Holden 2013).

Analysis

Currently, most asset mapping GPS surveys use the DGPS technique along with a lone single-frequency receiver. The DGPS technique will provide the required accuracy with an acceptable amount of cost. Even with PPP refined to real-time centimetre-level accuracy, the field work for asset mapping would remain the same. The user would still have to go into the field to collect data with a single-frequency lone receiver. The key difference is that now, the user could conduct the survey almost anywhere in the world, without regard to where the base station is. Another important advantage is that PPP’s position solutions are referred to a global reference frame, as compared to being relative to the local base station (Gao 2006).

If and when high-accuracy real-time single-frequency PPP is achieved, DGPS use could be reduced. This is because PPP does not require a base station and if the base station is a CORS network, the user need not pay for the base station data. However, as suggested by Bisnath and Gao (2008), since DGPS and PPP function independently, they could serve as mutual integrity checkers. Additionally, in urban areas where CORS networks and other base stations are already well-established, DGPS might still be used. In this way, DGPS would still serve a purpose, albeit maybe a reduced one.  In remote areas lacking CORS networks, PPP would probably be the preferred technique as no base station would be required.

Rizos et al. (2012) argues that since CORS networks provide a large role in providing the corrections required to achieve real-time centimetre-level PPP results, the need for CORS networks will not disappear. This could mean that in the future, within urban environments, PPP would be used to complement the DGPS technique instead of replacing it.

Once real-time single-frequency PPP is achieved, it seems that PPP would replace Point Positioning. This is due to PPP providing a more accurate position solution than Point Positioning with no obvious disadvantages.

Conclusion

Point Positioning uses C/A code to determine the receiver’s position. It includes receiver clock error, satellite clock error, satellite orbit error and atmospheric errors. DGPS accounts for these errors by having 2 separate receivers tracking the same satellites. PPP uses precise clock/orbit data along with a single receiver to obtain similar accuracy to DGPS. PPP would not completely replace DGPS as the CORS network is still important, especially in urban environments. PPP will provide a better position solution than Point Positioning in nearly all circumstances. With real-time high-accuracy PPP, asset mapping GPS surveys would no longer be restricted to always being within 15 kilometres of a base station.

References

Australian Maritime Safety Authority 2013, ‘Differential Global Positioning System’, viewed 02 Oct 2013, http://www.amsa.gov.au/navigation/services/dgps/

Bisnath, S & Gao, Y 2008 ‘Current State of Precise Point Positioning and Future Prospects and Limitations’ Observing our Changing Earth International Association of Geodesy Symposia, vol. 133, pp.615-623.

Chen, K 2004, ‘Real-Time Precise Point Positioning and Its Potential Applications’, Proceedings of ION GNSS-2004, pp. 1844-1854.

Chen, K and Gao, Y 2005, ‘Real-time precise point positioning using single frequency data.’ Proceedings of ION GNSS-2005, pp.1514-1523.

Gao, Y 2006 ‘Precise Point Positioning and Its Challenges, Aided-GNSS and Signal Tracking’ GNSS Solutions, November/December 2006, viewed 02 Oct 2013, http://www.insidegnss.com/auto/NovDec06GNSSSolutions.pdf

Hofmann-Wellenhof, B, Lichtenegger, H & Wasle, E 2008, ‘GNSS – Global Navigation Satellite Systems: GPS, GLONASS, Galileo, and more’ 1^st Edn, Springer-Verlag Wien, Austria.

Holden, L 2013a, ‘GNSS III’ Lecture Notes, GEOM 2066, RMIT, Australia.

Holden, L 2013b, ‘Postgraduate extra assessment’ Assignment Handout, GEOM 2066, RMIT, Australia.

Kouba, J and Héroux, P 2001, ,Precise point positioning using IGS orbit and clock products.’, GPS solutions, Vol 5, no. 2, pp. 12-28.

Rizos, C, Janssen, V, Roberts, C & Grinter, T 2012 ‘Precise Point Positioning: Is the era of differential GNSS positioning drawing to an end?’, Proceedings of FIG Working Week 2012, 6-10 May 2012, Rome, Italy

Witchayangkoon, B 2000 ‘Elements of GPS precise point positioning’, PhD dissertation, University of Maine, United States.