Evaluation of weighting functions for Sidelobe control in UWB SAR image processing

Abstract: Ultra wideband (UWB) Synthetic Aperture Radar (SAR) holds huge possibilities for both terrestrial and celestial object sensing with excellent details which assists in science and technology. SAR systems associated with large antenna beamwidth, large signal bandwidth and low frequency operating in the VHF/UHF region is becoming gradually more popular because of their rising number of application in the areas of ground-penetrating radar (GPR) and foliage penetration radar (FOPEN). Apodization techniques in UWB SAR imaging have attracted significant interest in recent years for sidelobe suppression in SAR images. This technique is split into two groups: linear apodization and non-linear apodization. Linear apodization technique means to apply amplitude weighting functions in frequency domain prior to the final inverse Fourier transform requisite to appropriately focus on the SAR images. Both linear and non linear techniques can be used to suppress sidelobes level. Frequently used linear weighting functions are Hanning, Hamming and Blackman. Linear techniques can control the sidelobes level but image resolution reduces simultaneously. But non-linear techniques like Spatially Variant Apodization (SVA), Complex Duel Apodization and Dual-Apodization can suppress sidelobes and preserve the spatial resolution concurrently. However for these methods, it can be hard to understand how the output signal relates to input signal and also the phase information of image is lost. In this thesis paper, the main focus is, on apodization techniques to propose a new weighting function for sidelobe apodization and investigate it on real SAR images. In this thesis, we also study Impulse response (IPR) function for UWB SAR image processing. A two dimensional sinc function is used as an impulse response function for narrow band (NB) SAR system. This function can be obtained from a two dimensional Fourier Transform of a SAR image. This rectangular estimation is reasonable for narrow band and narrowbeam SAR. But for large bandwidth and wide integration angles, this approximation for the UWB SAR spectrum is not valid. It can provide erroneous SAR image quality measurements. To obtain precise image quality measurement, SAR image need to be generated for a range of different integration angle as UWB SAR systems are related with large integration angle to maintain azimuth focusing. So, in this work the choices of optimum windows have been investigated at different integration angles in order to see if there are large differences between NB SAR Apodization and UWB SAR Apodization.