Neural Posterior Estimation of Terrain Parameters from Radar Sounder Data
For planetary scientists analyzing radar sounder data, this provides a principled Bayesian framework that handles noise and parameter correlations, improving over conventional point estimates.
This paper tackles terrain parameter inversion from radar sounder data, proposing a simulation-based inference approach using neural posterior estimation (NPE). The method achieves well-calibrated posteriors on simulated data and successfully transfers to real Mars radar profiles.
Radar sounders are electromagnetic instruments that can probe deep into the subsurface of Earth and other planetary bodies by processing the echo of transmitted radar waves. Conventional approaches for analyzing such data rely on approximate assumptions and often produce point estimates that ignore parameter correlations as well as galactic and measurement noise. We propose a simulation-based inference approach to terrain parameter inversion from radar sounder data, where synthetic observations from a GPU-based simulator are used to train a neural network-based density estimator for neural posterior estimation (NPE). By explicitly conditioning on reference surface assumptions, the proposed framework allows systematic evaluation of posterior robustness to reference surface variability. We demonstrate that our NPE model is well calibrated on simulated data and transferable to real Mars radar profiles, where we analyze terrain parameters using literature-informed reference values.