IEEE Geoscience and Remote Sensing Magazine - March 2020 - 49

obtain the slant-range displacement. The deformation-inversion
process generally uses the GB-DIn
technique, which includes the following six steps:
1) DIn
2) phase filtering
3) pixel selection
4) phase unwrapping, which is used
in the discontinuous operation
mode mentioned in the "Interferometry Phase Filtering" section
5) atmospheric phase compensation
6) deformation inversion.
The DIn processing flow is displayed
in Figure 9.

GB-SAR Images
DIn

Pixel
Selection

Differential
Phase Image

Phase Filtering

PS Points
Image

Filtered Phase
Image
APS Phase

DIFFERENTIAL INTERFEROMETRY
Estimating
The GB-DInSAR principle is based
on the DIn processing that utilizes
Atmospheric Phase
the time-adjacent images of the
Compensation
APS Phase
same region to generate the interferogram. From the phase of this
i nt e r fe ro g r a m , the deformation
fferential Phase After
Differential
information can be obtained [52],
Final Results
APS Compensation
Deformation
[53]. The technical principle is
shown in Figure 10. The SAR echo
Deformation
Geocoding
Inversion
signal of the ground target includes
the backscattering amplitude information A and the phase information z. The scattering information of each pixel in the image can FIGURE 9. A DIn processing flowchart. PS: permanent scatterer; APS: atmospheric
be expressed as c = Ae jz. The phase phase screen.
information includes the distance
information R between the phase center of the radar anTx
tenna and the target as well as the phase of the targetscattering characteristic z obj . The phase of the SAR imRx
age is expressed as

z=-

4rR
+ z obj . (15)
m

The phases of images 1 and 2 in the time series can be
represented by

z1 = -

4rR
4rRl
+ z obj1, z 2 = + z obj2, (16)
m
m

where m is the radar wavelength, and z obj1 and z obj2 are
the scattering characteristic phases in images 1 and 2, respectively. When the GB-DInSAR system is used for deformation measurement, the radar position is fixed, resulting
in a spatial baseline of zero. DIn processing is achieved by
the conjugate multiplication of two complex image pairs,
as expressed by

I = c 1 c )2 = ; A 1 A 2 ; exp" j (z 1 - z 2) ,, (17)

MARCH 2020

IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

Tx
Rx
∆R
FIGURE 10. A schematic diagram of DIn. Tx: transmitter; Rx: receiver.

where ) represents conjugate multiplication and z 1-z 2
is the DIn phase, which indicates the phase difference
between the image pair. Depending on the number of
acquisitions, time interval between the acquisitions, coherence development, atmospheric phase components,
and displacement velocities, different approaches for interferogram generation can be used to separate the phase
component caused by the surface displacement from the
unwanted effects related to noise and atmospheric perturbations, as shown in Figure 11 [1].
49

IEEE Geoscience and Remote Sensing Magazine - March 2020

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - March 2020