Neural Posterior Estimation of Terrain Parameters from Radar Sounder Data

TL;DR

This paper introduces a neural posterior estimation method for inferring terrain parameters from radar sounder data, improving robustness and calibration over traditional approaches.

Contribution

It presents a simulation-based inference framework using neural density estimators trained on synthetic data for terrain parameter inversion.

Findings

The NPE model is well calibrated on simulated data.

The approach is transferable to real Mars radar profiles.

Explicit conditioning on reference surface assumptions enhances robustness.

Abstract

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.