IEEE Geoscience and Remote Sensing Magazine - June 2020 - 113

period, the time resolution of the single interference pair is
relatively low, and the pair inevitably contains more of the
atmospheric phase. Therefore, for the differential interferogram of InSAR technology, the stability of the interference
phase is affected by the time base of the interference image pair and the spatial baseline, which leads to a problem
of low accuracy in the monitoring and results for D-InSAR
technology [49].
On the basis of D-InSAR technology, time series InSAR
technology has also been rapidly developed. The time series InSAR technology overcomes many limitations in traditional techniques and obtains a surface deformation time
series with high precision and reliability. The primary time
series InSAR methods are persistent scatterer InSAR (PSInSAR), least squares, SBAS, and SqueeSAR. Ferretti et al. [50]
proposed the permanent scatterer technique in 2001. The
basic idea is to extract target points with stable phase and
high backscattering characteristics by processing long-term
sequences of InSAR image data. These feature points are
permanent scatterer points. PSInSAR technology is a time
series analysis of these permanent scatterer points. PSInSAR can invert the surface-shaped variables by eliminating
the effects of the atmosphere, noise, and DEM. This method yields results on the order of millimeters.
The interferometric point target analysis (IPTA) technology [51], [52] and coherent point target (CPT) technology
[53] have the same idea as PSInSAR. PSInSAR can effectively improve the utilization of data by processing targets with
high coherence, such as artificial buildings and exposed
rocks, in the interference data set. With the development
of the permanent scatterer screening algorithm and the
permanent scatterer inversion network construction algorithm, more permanent scatterer points in the interference
data set were searched to improve the spatial resolution and
reliability of the results. Since the PSInSAR method does
not perform PU before the deformation rate inversion, the
differential interference phase is the wrapped phase. To obtain the real phase information, it is necessary to perform
PU to get the phase ambiguity of the entire cycle.
However, in the PSInSAR technique, the PU method
requires sampling data with arbitrary spacing because
the spacing among permanent scatterer points is usually
large and the sampling deformation is random. Therefore,
sparse PU has an important application to PSInSAR [54].
The primary data processing flow of PSInSAR technology
is to select one of the SAR images of K + 1 different phases
in the study area as the main image, and the other slave
images are, respectively, registered with the main image
and interfered with, obtaining K interferograms. Then, differential interference processing is performed according
to the owned DEM, and K differential interferograms are
obtained. The permanent scatterer point is detected from
the SAR image after registration and calibration, and the
differential interference phase set of the permanent scatterer point is obtained by combining the differential interferogram. Then, we can use an appropriate algorithm and
JUNE 2020

IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

K +1 SAR Images

Selection of the Common
Master Image
and Registration

DEM Data

Selection of
Permanent Scatterer
Candidates

Generate Differential
Interferograms

Removal of Atmospheric
Phase Screen
Time Series Analysis of
Pixels to Select
Permanent Scatters

Estimation of
Deformation Information

FIGURE 4. The PSInSAR data processing flowchart.

function model to solve the surface deformation value of
the permanent scatterer point and get the deformation value of the whole monitoring surface by interpolation. The
processing flow is shown in Figure 4.
PSInSAR technology is widely used in monitoring land
deformation, especially in cities, oil fields, mines, and other
areas with a large number of coherent targets. For example, in
2006, Meisina et al. [55] tested the applicability of PSInSAR
to detect and monitor ground displacement in the Oltrepo
Pavese territory. Kim et al. [56]
used PSInSAR technology to
monitor and measure the surTRADITIONAL GROUND
face subsidence near Incheon
DEFORMATION
Port and selected locations in
MONITORING METHODS
Busan, Korea. Meisina et al.
MAINLY INCLUDE LEVELING
[57] proposed a regional-scale
AND THE USE OF GPS.
PSInSAR data interpretation
method for public administration. In 2009, Perissin et al.
[58] used classical InSAR, PSInSAR, and the combination
of coherent-uncoherent analysis to obtain the main results
of deformation in Shanghai, Tianjin, and the Three Gorges
Dam test sites.
Mateos et al. [59] combined detailed geological and hydrogeological data with PSInSAR data to better understand
113

IEEE Geoscience and Remote Sensing Magazine - June 2020

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