Using GAGAN to improve GPS Positioning Accuracy - Part 1

Professor Chinmaya S Rathore
Indian Institute of Forest Management Bhopal, India


This is a two-part article on how to use GAGAN ( or any other SBAS available in your region) to get better positional accuracy from your GPS receiver for free. Part 1 of this article (this article) provides a quick background on differential positioning and Satellite Based Augmentation System (SBAS) concept while part 2 shows how to activate and use the SBAS service on a typical GPS receiver.

In the previous posts, we have seen that many sources of error influence the accuracy [1] of the positioning solution determined by the GPS receiver. In short, a location being identified by the GPS receiver as a certain latitude and longitude could, in theory, be around 15 meters away from its true position. While the user might not be able to control the sources of error that contribute to this positional inaccuracy, there is a neat trick that can provide the user with much better positional accuracy using the same receiver. It's called differential correction. The basic idea behind differential correction, illustrated in figure 1, is rather simple (its implementation is not!). 

Referring to figure 1, a GPS receiver (called a reference or base station) is installed at a location whose latitude, longitude and altitude are precisely known (1). The GPS receiver after getting a fix from GPS satellites gets a latitude and longitude reading as determined via ranging GPS satellites (2). Because it already knows its position accurately, it can compare the position obtained via the GPS satellites with its known position and find out the quantum of positioning error (3). It can now pass on (4) the error correction parameters to any nearby GPS receiver (also called a rover) which can likewise adjust GPS positions with the error corrections received from the base station (5).

Figure 1: The DGPS concept with real-time correction

While this arrangement is really nice, there are three potential issues that come in way to make it work effectively:

1. You need to establish and operate a base station. This typically requires special equipment which is quite expensive. 
2. The base station can pass correction parameters to a rover in real time if it is equipped with a radio transmitter and the rover with an appropriate receiver. This additional capability make both the base station and rover much more expensive (also see supplementary note 4). 
3. This arrangement can operate within a limited distance (usually a few hundred kilometers) in the vicinity of the base station governed by the premise that both the base station and the rover being locationally proximate experience similar atmospheric conditions, and must therefore, be subject to the same errors.  

Satellite-based Augmentation System (SBAS) provide a really elegant solution to the above DGPS issues making available differential corrections over a large area (continents!) to GPS receivers for free! The DGPS and SBAS functioning is quite similar in concept. The SBAS implements the real time differential correction idea by gathering positioning errors from a network of permanent base stations, computing differential corrections and uploading them to geostationary satellites (also referred to as GEO satellites)  which in turn broadcast these corrections over large areas. A GPS receiver, which is SBAS capable, can receive these correction messages in real time and make the required positional corrections. It is for this reason that the SBAS is sometimes also referred to as Wide Area Differential GPS or WADGPS. Currently, four countries are operating SBAS services while some others have proposed to operationalize such services in the near future.  The USA operates the Wide Area Augmentation Service (WAAS) available over North America, European Geostationary Navigation Overlay Service (EGNOS) is operational over the Europe, the Indian GPS Aided Geo Augmented Navigation (GAGAN) available over the Indian sub-continent region and the Japanese Multi-functional Satellite Augmentation System (MSAS) covering Japan. See this exhibit for a summary of various operational SBAS and their coverage areas.

Figure 2 conceptually summarizes the SBAS concept using GAGAN as an example.

Figure 2: SBAS Concept

It is important to point out that these systems are interoperable which means that the same GPS receivers will be able to receive differential correction messages from all these systems. The primary beneficiary of the SBAS is the aviation sector but all GPS users having WAAS-capable GPS receivers can benefit from SBAS by getting a typical positioning accuracy of less than 3 meters, 95% of the time [2]. It will be also worthwhile to reiterate that the differential corrections from the SBAS mentioned in this article are applied to positional measurements from GPS constellation satellites and not GLONASS (or Biedou). A good overall summary of the SBAS/WAAS concept with an interesting animation is available at the US Federal Aviation Administration website.

WAAS/EGNOS/GAGAN/MSAS broadcast correction messages on same frequencies as GPS (L1 / L5) and as such GPS receivers can read the broadcast differential correction data without any additional equipment requirement as long as the SBAS satellite is visible (line of sight) to the receiver.   

In part 2 of this article, we will see how we can activate and use GAGAN to get better positional accuracy using a popular WAAS-capable GPS receiver.This should help get more accurate positional data from field surveys.

Supplementary Notes 

[1] While accuracy is one of the commonly used and understood navigational parameters in reference to GPS, other parameters also characterize the performance of the GPS system. These parameters (in addition to accuracy) are integrity, continuity and availability.  It is important to point out that in addition to providing better positional accuracy, SBAS also improves GPS integrity (by sending timely alerts when positioning cannot be relied upon) which is crucial for aviation applications particularly for flight safety while landing. The interested reader is referred to this article by Dr. Richard B Langley for a fuller explanation of these terms.  

[2] GAGAN, among other component units, comprises of 15 Indian Reference Stations (INRES) spread across India, two master control centers at Bangalore, three uplink stations and 3 Geostationary Satellites two (GSAT-8 and GSAT-10) transmitting correction messages and one (GSAT-15) an in-orbit spare. Technically, the combined footprint of GSAT-8 and GSAT-10 satellites extends from Africa to Australia filling in the airspace gap between EGNOS and MSAS Satellite-based augmentation systems. GAGAN has an operational accuracy performance requirement of 7.6 meters. For more technical information about GAGAN, the reader is referred to an excellent article titled GAGAN - Redefining Navigation over the Indian Region by Ganeshan et. al., InsideGNSS, January / February 2016, pp. 42-48.

[3] Russia is developing an SBAS called System for Differential Corrections and Monitoring (SDCM) and China has announced the Satellite Navigation Augmentation System (SNAS). South Korea has also announced to develop an SBAS by 2021. Some private operators like OmniSTAR also operate SBAS.

[4] DGPS corrections can also be applied after the GPS data has been collected (i.e. not in real time) using a technique called post-processing. Essentially, position data from the rover is corrected with data from the reference station using post-processing software. It has been reported that positions corrected via post-processing generally result in higher accuracy when compared to real time systems such as SBAS. For more details on DGPS and post-processing, the reader is referred to this white paper by Trimble.

[5] Figure 2 is a highly simplified conceptual description of the SBAS concept created to convey an overall general idea in layperson terms. It must be pointed out that while GPS uses signal travel time from 4 or more GPS satellites to the GPS receiver to compute a position solution, the SBAS concept works backwards by calculating a correction factor in signal travel time (ranging error) using the accurately known position to improve positional accuracy resulting in the kind of effect shown in figure 2.