In this study we have proposed a method based on neural networks to retrieve refractivity, temperature,
pressure and humidity profiles by using FORMOSAT-3/COSMIC GPS radio occultation data. To overcome the constraint
of an independent knowledge of one atmospheric parameter at each GPS occultation, we trained three neural networks
with refractivity profiles as input computed from the geometrical occultation parameters relative to the FORMOSAT-
3/COSMIC satellites, while the targets were the dry and wet refractivity profiles and the dry pressure profiles obtained
from the contemporary European Centre for Medium-Range Weather Forecast data. We have considered 1041 available
satellite radio occultations covering the entire ocean area spanning within the Tropics during July-August 2006. We have
used 937 profiles for training the neural networks, the remaining 104 ones for the independent test.