Level-1 SLC Processing Algorithms
Single Look Complex (SLC) processing focuses the data in azimuth and range to form an image. The SLC processing takes as input, the signal data and the pre-processing output including orbit information and Doppler centroid estimation polynomials. The processing then applies range processing, azimuth pre-processing, azimuth processing and azimuth post-processing, as shown in the figure below.
Each burst in each sub-swath for TOPSAR products is processed independently. The independently focused burst images are then included in a single sub-swath image ordered by azimuth time and separated by black-fill.
Range processing consists of:
The SWST bias is accounted for by adding it to the range start time.
Azimuth pre-processing applies azimuth zero-padding, azimuth forward FFT and, for TOPSAR products, azimuth frequency unfolding and resampling.
Azimuth pre-processing converts the data into the range-Doppler domain needed for the range-Doppler algorithm in the azimuth processing.
The azimuth processing utilises the range-Doppler algorithm characterised by a hyperbolic range equation.
The range-Doppler algorithm is computationally efficient and, for typical space-borne imaging geometries, the range-Doppler algorithm is an accurate approximation of the exact SAR transfer function. The algorithm is phase-preserving and SLC images formed with it can be used for applications such as interferometry.
The SENTINEL-1 IPF enhances the range-Doppler algorithm previously used for ENVISAT ASAR by using a higher accuracy secondary range compression method to satisfy phase preservation. Also, the phase of the azimuth matched filter used for SENTINEL-1 is more accurate than the quadratic phase of the matched filter used for ASAR.
The azimuth processing can be summarised as shown in the figure below.
Secondary Range Compression
For range compression, raw data is multiplied in the frequency domain by the range reference function. This multiplication is carried out with complex values, i.e. the phase information in the data is preserved.
Range Cell Migration Correction
The range-Doppler algorithm uses the large difference in timescale of range and azimuth data and approximately separates processing in these two directions using Range Cell Migration Correction (RCMC). Target trajectories need to be corrected to account for the fact that the instantaneous slant range changes with azimuth time by moving all responses of a target from the trajectory into a straight line. RCMC is performed in range frequency and azimuth frequency domain.
For TOPSAR data, a range resampling step is combined with RCMC.
Azimuth compression applies an azimuth matched filter to each azimuth line and applies the inverse FFT to the azimuth line. The result is a focused image.
Azimuth post-processing is for TOPSAR products only, to resample to a common grid. This resolves issues with time aliasing due to the limited support during azimuth focusing. The algorithm includes the following steps: