Developments in Precise Point Positioning for GPS and Galileo
van der Marel, H.; Le, A.Q.
Delft University of Technology
Precise Point Positioning (PPP) is a positioning technique aimed at processing undifferenced carrier phase and pseudorange measurements from a stand-alone GNSS receiver to compute positions with a high, decimeter or centimeter, accuracy everywhere on the globe. To facilitate the PPP technique precise GNSS satellite orbit and clock solutions are required. Precise GNSS satellite orbit and clock solutions are available from the International GNSS Service (IGS) for post-processing applications, but also more and more real-time orbit and clock solutions are becoming available.
The PPP technique has become very popular in the scientific and research communities for applications that require high accuracy and in which latency was not important, but for which a full network solution is too complicated. With the advent of real-time orbit and clock solutions the PPP technique is now also used for a wide range of applications such as offshore positioning, aircraft navigation, high-precision farming and meteorology. The main problem so-far for real-time applications is the relatively long convergence time (in the order of 10 minutes) of the algorithm to the desired accuracy.
In the paper we first show the equivalence between a PPP and the more traditional network solutions. It turns out that the PPP technique is actually a two step procedure. In the first step a global network is processed by a global analysis center, while in the second step the user receiver is processed using products from the first step. This means that the PPP technique is capable of delivering the same millimeter accuracies as network solutions do, and that integer ambiguity resolution should be possible. However, there are some practical implications to PPP which result in a slightly inferior performance compared to network solutions. These practical limitations, and possible solutions, are also discussed in the paper.
So far high-precision PPP was geared towards users carrying dual frequency receivers. More recently also high accuracy positioning using stand-alone single-frequency GPS receivers is investigated. Single frequency receivers, which are being used, for instance, in georeferencing applications and precise agriculture, are becoming popular thanks to a lower price relative to their dual-frequency counterparts. The performance of the single frequency PPP is mainly driven by the quality of existing ionosphere models. IGS provides, in addition to satellite orbit and clock information, the Global Ionosphere Model (GIM), a gridded global set of daily Total Electronic Content (TEC) estimates. This opens up the possibility for sub-meter precision for single-frequency positioning.
The performance of the single- or dual-frequency PPP approach is investigated using GPS data under realistic conditions and using different implementations, including Kalman filter design and measurements modeling.
Instead of the traditional PPP approach based on the ionosphere free linear combination we demonstrate a new PPP method using raw pseudorange and carrier phase observations that leads to improved accuracy and convergence times for kinematic applications. This latter approach can easily be extended to Galileo and multi-GNSS constellations.
In the paper we further discuss the future of PPP in view of new GNSS satellite signals and requirements for global products to facilitate multi-constellation PPP.