The high level of signals transmissivity during severe weather as well as the low-cost, low-mass and lowpower GNSS-R receivers operating on low-Earth orbiting satellites, are some of the advantageous characteristics of this technique. Conventional GNSS-R retrieval algorithms rely on the parametric regression approaches inverting observables to the geophysical data products. These models are developed based on simplifying assumptions due to the complexity in the physics of signals scattering from ocean surface. Besides, these models are not yet fully validated for field conditions and might be subject to refinements. Similar to the conventional remote sensing techniques, the GNSS-R can benefit from AI methods. This project is dedicated to the study of AI algorithms for higher quality and novel geophysical GNSS-R data products. The research will also enhance the knowledge on the physics of signal scattering and ocean geophysics.
Exploiting signals of the Global Navigation Satellite System (GNSS) after reflection off the Earth’s surface has emerged as a novel remote sensing technique, which is called GNSS Reflectometry (GNSS-R)
