IEEE Geoscience and Remote Sensing Magazine - March 2020 - 44

motion of the radar antenna to the target is synthesized for a
larger equivalent antenna aperture by data processing methods. During the motion, the radar transmits coherent signals
within the synthetic aperture and receives the echo that is
used for coherent processing for imaging. By repeating the
scanning of the same scene, the DIn method with complex
image pairs is further used to extract the radar phase of the
target region for deformation inversion [1]-[4].
Conventional deformation measurement techniques,
such as strainometer or total station, usually require the
sensor to directly contact or be buried in the target. In
some cases, if the sensor cannot approach the target, the
noncontact measurement method of GB-DInSAR proves
advantageous. Even if single-point data are accurate and
taken particularly in significant areas, they cannot be assumed capable of deducing the whole landslide area. This
is mainly important in large-size landslides or complex
slope movements, which are
characterized by different
movement patterns [2]. BeGB-DInSAR IS A GOOD
cause GB-DInSAR can obtain
COMPLEMENT TO
the deformation at each pixel
CONVENTIONAL
point in the slant-range plane,
MEASUREMENT
its measurement efficiency is
TECHNIQUES, BUT IT
very high compared to that
of conventional techniques.
CANNOT COMPLETELY
However, GB-DInSAR obREPLACE CONVENTIONAL
tains the deformation of the
METHODS.
radar line-of-sight (LOS) direction, which is the projection component of the actual
deformation of the target in the LOS direction. Therefore,
GB-DInSAR is a good complement to conventional measurement techniques, but it cannot completely replace conventional methods.
Usually, the platform of GB-DInSAR is completely stationary on the ground and can achieve higher monitoring
accuracy and a much shorter revisit period than traditional
spaceborne or airborne DInSAR. In particular, it is more accurate when monitoring minute- or hour-level changes in
deformation areas. It is known from the literature that this
technology has been successfully applied to stability monitoring [5], [6], dam safety monitoring [7], [8], and landslide
rescue [9], [10] of open-pit slopes.
According to the formation method of synthetic aperture, GB-DInSAR is mainly divided into two modes: linear
scanning mode and arc scanning mode. The main difference between these modes lies in their formations of synthetic aperture.
In linear scanning mode, the radar antenna reciprocates
along a high-precision linear rail to form a linear synthetic aperture for azimuth resolution. Representative systems include
Ingegneria Dei Sistemi's (IDS's) IBIS-L system at Ku band, an
acquisition time of 8 min, and an azimuth resolution of
4.4 mrad [11]; the European Union Joint Research Centre's (JRC's) linear SAR (LiSA) system at C and Ku band, an
44

acquisition time of 12 min, and an azimuth resolution of 3 mrad
[12]-[14]; Dutch MetaSensing's FastGB-SAR system at Ku
band, an acquisition time of 5 s, and an azimuth resolution
of 4.5 mrad [15], [16]; Polytechnic University of Catalonia's
(UPC's) RiskSAR system at X band, an acquisition time of
1 min, and an azimuth resolution of 4 mrad [16]; and North
China University of Technology's (NCUT's) GB-SAR system
at Ku band, an acquisition time of less than 1 min, and an
azimuth resolution of 3 mrad, and so on.
In the arc scanning mode, the radar antenna performs
an arc scan in the horizontal plane to form an angular synthetic aperture. Representative systems include ArcSAR from
Kangwon National University (KNU) at X band, an acquisition time of 14 min for 180°, and an azimuth resolution of
1 mrad [18]; IBIS-ArcSAR's system from IDS in Italy with an
acquisition time of 20 s for 180° [19]; an arc frequency-modulated (FMCW)-SAR system from the Institute of Electronics,
Chinese Academy of Sciences (IECAS) at X band [20]; and an
NCUT-ArcSAR at Ku band, an acquisition time of 3 s for 180°,
and an azimuth resolution of 10 mrad, and so on. A more detailed discussion is given in the "Typical GB-DInSAR System"
section. The arc scanning mode has a more significant advantage in 3D imaging and 360° panoramic monitoring.
Early GB-DInSAR systems generally used the stepped-frequency continuous wave (SFCW), which consists of a series of
continuous narrowband signals. Due to the small pulsewidth
of the narrowband signal, the distance between the target and
the radar platform during transmission is hardly changed.
The narrow bandwidth of the SFCW signal induces high peak
power, resulting in large size and a mass of sensors as well
as a complicated structure. Unlike the stepped-frequency
mode of operation, the FMCW-SAR transmits a continuous
chirp signal and receives continuous echo signals during the
motion of the radar antenna. The longer sweep time suggests
that the distance between the target and radar could change
in the process of receiving the signal. For a farther operating
distance, the peak power of emission in FMCW radar is lower,
which leads to the advantages of small size, light weight, and
low cost [21].
GB-DInSAR technology largely draws upon the airborne
SAR interferometry method, but there are certain technical
differences. For example, the GB-DInSAR observation platform is absolutely fixed, and generally no image registration
is required. The specific processing steps include DIn, phase
filtering, pixel selection, atmospheric effect removal, and deformation inversion.
The GB-DInSAR system has been successfully put into
practical use in many ways. In [5] and [6] and [22]-[24], the
authors applied the GB-DInSAR system to the monitoring of
open-pit mines. Unstable slope monitoring is another important application of the GB-DInSAR system, but the system's
high resolution and high sensitivity make it very responsive to
small changes that are common in the natural environment,
which may lead to a loss of coherence [21]. Other important
applications include urban [25], structural [26], and dam
monitoring [27]. In addition, although GB-DInSAR is used
IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

MARCH 2020

IEEE Geoscience and Remote Sensing Magazine - March 2020

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