Processing of Synthetic Aperture Radar Data Using Frequency Modulated Signals

摘要

The main objective of Synthetic Aperture Radar (SAR) processing is converting the raw acquired data into pixels in the final processed image. The signal is spread in the range and azimuth directions, and the signal migrates to other range cells. Range Doppler Algorithm (RDA) is the most widely applied algorithm in SAR data processing. It performs range compression followed by azimuth compression to form the final reconstructed image. Unfortunately, SAR images suffer from many imperfections that will be studied and analyzed in this paper. Firstly, range compression introduces sidelobes that reduce the Signal-to-Noise Ratio (SNR). For this case, a comparison between three configurations of frequency modulated signals is presented. Moreover, the channel effect that causes degradations in the received signals is investigated and regularized de-convolution is used to reduce this effect. Finally, the random noise that causes a pixel-to-pixel variation and appears as a granular noise in the image will be reduced by applying a median filter to the reconstructed image.