HydroDrone: A Circular External-Looking UAV-SAR Campaign for Short-Term Temporal Decorrelation Characterization
Abstract
This article presents the HydroDrone campaign, a dedicated uncrewed aerial vehicle (UAV)-based synthetic aperture radar (SAR) experiment designed to characterize short-term temporal decorrelation over vegetated scenes. The system employs a C-band frequency modulated continuous wave (FMCW) radar mounted on a multirotor UAV, performing repeated circular external-looking acquisitions over a chosen test site. This configuration enables the generation of dense interferometric stacks with temporal baselines ranging from tens of seconds to several hours. This work derives the specific mathematical framework for the computation of along-track wavenumbers and azimuth resolution in circular external-looking geometries, highlighting the fundamental differences with respect to standard linear trajectories. To compensate for residual trajectory errors arising from navigation uncertainties, a multiaperture interferometric trajectory compensation algorithm is introduced and applied across the stack. Preliminary results from a single campaign day are presented, including interferometric phase and coherence stacks before and after trajectory compensation. Coherence matrices computed over three different land-cover types (cornfield, medium-height vegetation, and meadow) are analyzed as a preliminary assessment of the coherence behavior and temporal sampling capability of the HydroDrone system. Thermal and spatial decorrelation contributions are quantified to confirm that the observed coherence loss is primarily driven by temporal decorrelation, providing a robust experimental basis for future vegetation clutter decorrelation modeling.