GAGAN is India's system that corrects GPS errors for aircraft, broadcasting corrections from geostationary satellites so planes can navigate more accurately and safely.
GPS alone is fine for finding a restaurant. Aviation asks for more: a pilot needs to know not only where the plane is but also how far that position could be wrong.
GAGAN (GPS Aided GEO Augmented Navigation) solves this for Indian airspace. ISRO and the Airports Authority of India built it as a satellite-based augmentation system, or SBAS. It works alongside similar systems: WAAS in the US, EGNOS in Europe and MSAS in Japan.
GAGAN has four parts on the ground and in space.
A GAGAN-capable receiver on the aircraft applies the corrections to its GPS position.
India lies in the low-latitude, equatorial region, where the ionosphere is unusually variable. The ionosphere slows GPS signals by different amounts at different places and times. The standard model used by other SBAS systems did not work well here, so the GAGAN team developed a region-specific ionosphere model before the system could be certified.
India's aviation regulator, the DGCA, certified GAGAN for en-route navigation (RNP 0.1) on 30 December 2013 and for approach with vertical guidance (APV-1) on 21 April 2015.
Look for GSAT-8 and GSAT-10 among the geostationary satellites: they carry GAGAN's correction signals. Find GSAT-8 and GSAT-10
An SBAS sends three kinds of correction.
1. Satellite orbit and clock corrections. The reference stations know their own positions exactly. Any difference between the measured and true range to a GPS satellite, common to all stations, points to an error in that satellite's broadcast orbit or clock.
2. Ionospheric corrections. The ionosphere's delay depends on how many free electrons the signal passes through, and that varies across the sky. SBAS models the ionosphere as a thin shell and broadcasts vertical delays at grid points. Inside GNSS reports that GAGAN's model gives users a value at a 350 km shell height and was designed to capture the equatorial ionisation anomaly, which a single thin shell describes poorly.
3. Confidence bounds. With each correction comes a bound on how wrong it might still be; for example, users compute vertical ionosphere error bounds from the broadcast grid. This lets an aircraft's receiver judge whether its position is trustworthy enough for that phase of flight.
A simple numerical view: suppose the reference station at a surveyed point finds that a GPS satellite's range reads 6 m too long. A nearby aircraft seeing the same satellite through nearly the same patch of sky subtracts 6 m from its own range. Errors common to both cancel; only the local differences remain. GAGAN does this for every satellite and every grid point at once, and sends the result through GSAT-8 and GSAT-10.
GPS Aided GEO Augmented Navigation, built by ISRO and the Airports Authority of India.
Over India's equatorial region the ionosphere varies strongly, and the standard thin-shell model did not describe it well enough.
Geostationary satellites GSAT-8 and GSAT-10, with GSAT-15 as an in-orbit spare.