Ray Tracing-Assisted Spotlight SAR for UAVs in Reflective Environments : Assessing the Influence of Position Stability on Image Formation

Sammanfattning

Synthetic Aperture Radar (SAR), traditionally deployed on satellite platforms with stable and predictable trajectories, is increasingly implemented on UAVs due to their flexibility in low-altitude applications. Ray Tracing (RT) is a modeling technique that can be used in combination with SAR to improve physical realism of the reconstructed images. However, UAV-based SAR systems are highly sensitive to motion disturbances, which introduce phase errors and degrade image quality, particularly in reflective environments with multipath propagation. Although previous work has investigated implementations such as Range-Doppler (RD) imaging geometry and phase correction algorithms as ways to mitigate motion errors, the explicit relationship between motion disturbances and their corresponding impact on reconstructed image quality has not been systematically characterized. This thesis evaluates the sensitivity of RT-assisted SAR imaging to UAV position instability by simulating an urban scene and introducing controlled range and altitude disturbances to the platform trajectory. Image quality is assessed using Entropy and RMS Contrast metrics. The results show that the system is relatively robust to small trajectory errors of approximately 4cm, with only minor degradation, but experiences significant image deterioration for errors exceeding approximately 10cm, with increased blurring and reduced feature separability. Range disturbances were found to produce slightly higher image disorder than altitude disturbances. Additionally, Entropy proved to be a more reliable indicator of image degradation than RMS Contrast under motion disturbances. These findings highlight the importance of motion stability and accurate trajectory estimation for maintaining image quality in UAV-based RT-assisted SAR systems.