Interferometric Phase Unwrapping for W-Band GBSAR: Experiments on Simulated Signal

Abstract. Displacement monitoring of structures and the Earth’s surface have always been of great importance, aiming to minimize human casualties and financial losses. One of the prominent methods is radar interferometry using various types of radar sensors. Specifically, GBSAR (Ground-Based Synthetic Aperture Radar) sensors have gained more attention for this purpose. Recent GBSAR developments tend to using shorter wavelengths for displacement monitoring, such as mmWave W band. However, decreasing the wavelength increases the chance of phase wrapping and unwrapping error. This paper analyses the performance of prominent phase unwrapping techniques on a W-band GBSAR sensor. Various simulated scenarios have been conducted, and subsequently, to solve the unwrapping problem, temporal and spatial techniques, including Goldstein’s Branch-Cut and quality-guided algorithms have been investigated. The results show better performance of Goldstein’s Branch-Cut algorithm compared to the quality-guided method, with RMSE values for displacement extraction of 0.122 and 0.154 mm, respectively. Moreover, Goldstein’s Branch-Cut algorithm also demonstrated a faster processing speed.