CAN SPECAN BE MADE PHASE PRESERVING? FURTHER OPTIONS TO INTERFEROMETRIC BURST-MODE AND SCANSAR PROCESSING

Kurzfassung

SPECAN has a long history. It was applied foremost in SIR-C and low-resolution strip-map and ScanSAR processing and has shortly been made available also to Spotlight processing. It was introduced to SAR data compression in [1] as an option for the efficient processing of SAR data. It is computationally efficient, since only one FFT is required. Then it was mostly used for low-resolution SAR image generation at low computational cost. Meanwhile, SPECAN was employed for processing SAR data within DLR‘s ECS processor. It seems that apart from the range variant sampling problem, that has been solved by azimuth scaling, it could be also made phase preserving by simply applying a quadratic phase to the data. ScanSAR and burst-mode interferometry suffer from the complexity of burst-mode and ScanSAR data in its signal properties and data/ product handling. However, it can be shown that burst-mode data can be processed by a standard chirp scaling time-domain or range-Doppler domain processor without loss of computational efficiency when the resampling and range spectral filtering steps in Interferometry are integrated into a interferogram formation step based on a bank of azimuth and range resampling filters. Cross-correlation for fine-registration of the data sets can be optimized by a in-advance raw data line nulling and alignment. After application of a quadratic phase matching the SAR’s FM rate burst images may be treated like low resolution strip-map images. This may be exploited to efficiently process burst-mode imagery with standard interferometric strip-map SAR processors. In this paper the question of phase preserving SAR processing with SPECAN is discussed. Time processing serves as a reference for discrimination. Moreover, the pack-and-go algorithm is verified. In this respect, a novel combination for ScanSAR processing and two new processing strategies for burst-mode interferometry are presented.