Day 4

Detailed paper information

Back to list

Paper title De-ramping of SLC-IW TOPS data and Ocean Circulation Parameter Estimation
Authors
  1. MUHAMMAD AMJAD IQBAL University POLITEHNICA of Bucharest (UPB) Romania Speaker
  2. Mihai Datcu DLR - German Aerospace Center
  3. Andrei Anghel CEOSpaceTech
Form of presentation Poster
Topics
  • A8. Ocean
    • A8.14 Remote-sensing of Ocean Waves and their Applications
Abstract text Abstract:

The spectral characteristics of SLC-IW TOPS are significantly different from Strip-map (SM). Due to the burst mode and series of sub-swaths, the target area is scanned for a short period of time, consequently, swath width comes at the expense of azimuth resolution. Significant focusing is required to remove quadratic phase drift and achieve an SLC base-band. The de-ramping effectively eliminates the quadratic drift and restores the baseband data. The ocean circulation parameters are extracted from the echo signal based on data-driven Doppler centroid (DC). The ocean circulation parameters include surface velocity, wave height, and direction swell, while compared with the synergy of benchmark data.

Background:

Due to burst-mode, SLC-IW TOPS differs from SM in terms of schematics, and the system observes in the form of sub-swaths periodically. As a result, the target region is scanned only for a fraction of the burst duration, and thereby the illumination is reduced, and the wide swath comes at the cost of azimuth resolution [1]. Sentinel-1 IW TOPS data preserves quadratic phase term in the azimuth direction which
leads to phase ramps, this needs to be eliminated from the SLC data for the subsequent applications.
In literature, the ocean circulation parameters for SLC-IW data are estimated based on the information provided in the OCN level-2 product, or geophysical interpretation is calculated from satellite orbit parameters [2]. The orbit parameters velocity V and incident angle θ in practical is usually not accurate enough to obtain the DC which fulfills the need for SAR imaging. Therefore, this work estimates the DC and all associated parameters from echo data [3], and all the ocean circulation parameters are data-driven.

Methodology:

To remove the quadratic drift, it is essential to move the spectral component of SLC-IW to the baseband by deramping. The phase term for deramping is defined as:
ϕ(η, τ ) = exp{(−j.π.k_t(τ )) × (η − η_ref (τ ))^2} (1)
whereas, reference time η_ref (τ ), and Doppler centroid rate k_t(τ ) are functions of range samples, while η is zero-Doppler azimuth time. The phase term needs to be multiplied in time
domain with SLC signal S_slc.
S_d(η, τ ) = S_slc × ϕ(η, τ ) (2)
Alternatively, deramping can be done in the SNAP tool using the Sentinel-1 TOPS operator. The flow of the process is given in the figure.

On that account, the Doppler centroid is the essence of this topic. In the literature, the DC has been predicted by conventional OCN product information using DC polynomial information provided in the metadata. We use correlation doppler estimation (CDE) which takes an advantage of azimuth shift and the PRF [4]. This DC history is utilized to retrieve radial surface velocity (RSV) with incident angle and radar frequency information. And with the empirical relationship of RSV, we estimated significant wave height (SWH) [5]. The SWH is the average wave height (from trough to crest) of the highest third of the wave height during the sample period. The comparisons are made with the synergy of benchmark data (measured by OCN product) for the same location, date, and time while using the SLC-IW TOPS product [6].

Results and discussion:

The quadratic drift is removed when phase term ϕ(η, τ) is multiplied with the original SLC image the data moves to the baseband domain. Deramping is done so far to eliminate the quadratic effect of phase term by chirp signal. , we extract ocean circulation parameters for the post-processing. For this, we measure the RSV based on DC information estimated by the CDE method, which perfectly matches with benchmark data. The RSV is in a good match and within the limit of error bounds, while in the core of the stream it reaches up to 2.5 m/s.
The RSV is an associated term for retrieving significant wave heights (Hs), which varies by a few meters. we use dual-polarization VH, which provides a better estimate of Hs than single polarization.

Conclusion:

The designed chirp function de-ramps the data and the result is theoretically correct, whereas the data moves to the baseband. The ocean
circulation parameters are measured and numerical values are compared with benchmark data and found perfectly matched. The numerical merit of comparisons is in good spatial correlation with minimum root mean square error (RMSE) and negligible mean absolute error (MAE).

References:

[1]. De Zan, Francesco, and A. Monti Guarnieri. "TOPSAR: Terrain observation by progressive scans."IEEE Transactions on Geoscience and Remote Sensing, 44.9 (2006): 2352-2360.
[2]. Hansen, Morten Wergeland, et al. "Retrieval of sea surface range velocities from Envisat ASAR Doppler centroid measurements."IEEE Transactions on Geoscience and Remote Sensing, 49.10 (2011): 3582-3592.
[3]. Zou, Xiufang, and Qunying Zhang. "Estimation of Doppler centroid frequency in spaceborne ScanSAR."Journal of Electronics (China),25.6
(2008): 822-826.
[4]. M. Amjad Iqbal, Andrei Anghel, and Mihai Datcu, ”Doppler Centroid Estimation for Ocean Surface Current Retrieval from Sentinel-1 SAR
Data”, IEEE Eu-RAD conference European Microwave week, 2022.
[5]. Pramudya, Fabian Surya, et al. "Enhanced Estimation of Significant Wave Height with DualPolarization Sentinel-1 SAR Imagery." Remote Sensing, 13.1 (2021): 124.
[6]. AElyouncha, Anis, Leif EB Eriksson, and Harald Johnsen. "Comparison of the Sea Surface Velocity Derived from Sentinel-1 and Tandem-X." 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS. IEEE, 2021.