Back-projection (BP) Imaging Algorithm

BP algorithm is a typical time-domain synthetic aperture radar (SAR) imaging method which conducts coherent superposition in the azimuth direction after the range compression. Theoretically, it ignores any approximation, leading to high-quality imaging results. However, the huge computational complexity exists in the value calculation of every pixel one by one in the processing. To improve the execution time, I take advantage of Matrix Operation in MATLAB to use 2D matrix to perform the BP processing instead of reconstructing 2D image point by point. Undergo this change, the time can be cut down by 10%.

• BP 2D imaging formula

• BP imaging simulation result

Motion Error Compensation

The trajectory estimation has a great impact on the focusing of both range direction and azimuth direction in the final 2D SAR image. In practice, the platform movement is not the ideal linear track, but rather irregular which is usually caused by atmospheric turbulence and mechanical vibration made it more difficult to obtain the precise track. The fail in track estimation will create unexpected phase error (or motion error) in the imaging results. Since the accurate distance estimation between the platform and the imaging scene is crucial in high-quality BP SAR imaging, the motion error compensation is required in the BP SAR imaging.

• phase error occurred in imaging results

• BP imaging simulation result with phase error

Autofocus Algorithm

Autofocus algorithm compensates the motion error based on the radar echo signals which works on the signal processing layer, rather than make any effort to acquire the high-accurate trajectory data with high-quality motion sensors. At first, the existing autofocus algorithm based on maximum image intensity is simulated, and the imaging result is not as much as expected.

• imaging results of existing autofocus algorithm based on maximum image intensity

To improve the imaging results, the thesis proposes an autofocus algorithm based on maximum image contrast to figure out the phase error for BP SAR imaging. The theory behind is the positive relationship between the SAR images focusing level and its image contrast. The problem of focusing can be converted to the determination of maximum contrast.

• phase error compensation of autofocus algorithm based on maximum image contrast

To acquire the high-quality focusing SAR 2D imaging results, the work combines the two autofocus algorithm which applies image intensity to estimate the proper antenna phase centre and introduces image contrast to improve the image quality.

• imaging results of combined autofocus algorithm