RESEARCH ON AIRBORNE SAR IMAGING BASED ON ESC ALGORITHM

Xiaoting Dong, Xiaoying Yue, Yue Zhao, C. M. Han2017 · DOI

TL;DR

This paper presents a method for improving airborne Synthetic Aperture Radar (SAR) image quality by integrating motion compensation (MOCO) into the Extended Chirp Scaling (ECS) algorithm. By precisely correcting for deviations in the aircraft's flight path, the authors demonstrate significantly sharper, clearer SAR images with reduced artifacts.

What Problem Does This Paper Address?

Airborne SAR systems offer flexible and rapid Earth observation capabilities. However, unlike space-borne SAR, airborne platforms are more susceptible to deviations from their ideal straight-line flight paths. These deviations, even small ones, introduce motion errors that corrupt the phase of the radar signal. This corruption leads to a cascade of image quality issues, including reduced signal-to-noise ratio (SNR), increased azimuth ambiguity, and defocusing, ultimately degrading the resolution and interpretability of the SAR imagery. Accurately measuring and compensating for these nonlinear motion errors is crucial for obtaining high-quality airborne SAR data.

Methods

The core of this work is the integration of a Motion Compensation (MOCO) technique into the Extended Chirp Scaling (ECS) algorithm. The ECS algorithm is a well-established method for processing SAR data, known for its efficiency in handling range migration and Doppler centroid variations. The authors incorporate a two-step MOCO approach within the ECS framework. This MOCO process utilizes data from a Position and Orientation System (POS) to accurately track the aircraft's trajectory and the antenna phase center (APC). The MOCO algorithm corrects for both linear and quadratic phase errors introduced by these trajectory deviations. The paper outlines the geometric model of SAR imaging, analyzes the nature of motion errors, and derives the necessary phase compensation formulas for both first and second-order motion errors. The ECS algorithm, with this integrated MOCO, is then applied to process the raw SAR echo data.

Datasets

The experiment described in the paper uses VV-Polarization raw data from a C-band airborne SAR system. The data was acquired over Sanya, Hainan province, in November 2014. No new datasets were introduced; existing raw SAR data was used for evaluation.

Key Insights and Ideas

  • Motion Errors are Critical for Airborne SAR: The paper strongly emphasizes that even minor deviations from an ideal flight path have a significant negative impact on airborne SAR image quality, a problem less pronounced in stable space-borne platforms.
  • Integrated MOCO within ECS is Effective: Combining motion compensation directly into the ECS processing chain is shown to be a powerful approach, offering a more streamlined and efficient way to correct for motion-induced artifacts.
  • POS Data is Essential for Precision: The accuracy of the motion compensation is directly tied to the quality of the Position and Orientation System (POS) data, highlighting the importance of high-precision sensors for airborne SAR.
  • Tangible Improvements: The integrated MOCO approach leads to demonstrable improvements in key image quality metrics, not just theoretical gains.

Key Results

The experimental results demonstrate that the airborne SAR images processed with the ECS algorithm integrated with MOCO exhibit significantly better quality compared to those processed without MOCO. Specifically, the compensated images show:

  • Improved Object Recognition: Ground objects are more clearly delineated and recognizable.
  • Enhanced Resolution: The sharpness and detail in the images are noticeably improved.
  • Reduced Azimuth Ambiguity: Artifacts that appear as repeating patterns in the azimuth direction are markedly reduced.
  • Better Peak Side Lobe Ratio (PSLR): This indicates improved spatial resolution and reduced spurious signals from strong targets.

Limitations and Future Work

  • Limitations: The paper mentions that the imaging performance is sensitive to the "envelope and phase of motion errors." While the ECS with integrated MOCO addresses this, the inherent accuracy of the POS data will ultimately limit the achievable compensation quality. The paper also assumes "narrow-beam antennas and horizontal imageable area" for the ECS algorithm.
  • Future Work: The authors conclude that their method is "valid, useful and operative," suggesting its readiness for practical application. Further work could explore the performance of this integrated approach with different SAR systems, polarizations, or in more challenging terrain conditions. The paper doesn't explicitly detail future research directions beyond the successful demonstration of their current method.

Summary updated

Abstract

Due to the ability of flexible, accurate, and fast obtaining abundant information, airborne SAR is significant in the field of Earth Observation and many other applications. Optimally the flight paths are straight lines, but in reality it is not the case since some portion of deviation from the ideal path is impossible to avoid. A small disturbance from the ideal line will have a major effect on the signal phase, dramatically deteriorating the quality of SAR images and data. Therefore, to get accurate echo information and radar images, it is essential to measure and compensate for nonlinear motion of antenna trajectories. By means of compensating each flying trajectory to its reference track, MOCO method corrects linear phase error and quadratic phase error caused by nonlinear antenna trajectories. Position and Orientation System (POS) data is applied to acquiring accuracy motion attitudes and spatial positions of antenna phase centre (APC). In this paper, extend chirp scaling algorithm (ECS) is used to deal with echo data of airborne SAR. An experiment is done using VV-Polarization raw data of C-band airborne SAR. The quality evaluations of compensated SAR images and uncompensated SAR images are done in the experiment. The former always performs better than the latter. After MOCO processing, azimuth ambiguity is declined, peak side lobe ratio (PSLR) effectively improves and the resolution of images is improved obviously. The result shows the validity and operability of the imaging process for airborne SAR.
BibTeX
@inproceedings{dong2017research,
  author = {Xiaoting Dong and Xiaoying Yue and Yue Zhao and C. M. Han},
  title = {{RESEARCH ON AIRBORNE SAR IMAGING BASED ON ESC ALGORITHM}},
  booktitle = {˜The},
  year = {2017},
  doi = {10.5194/isprs-archives-xlii-2-w7-601-2017},
}
Figure