Geology Reference
In-Depth Information
Aperture azimuth
beamwidth ( θ a )
h
θ a
Ground target
resolution element
Figure 7.30 Concept of SAR is demonstrated by nine positions of antenna transmitting beams (the dots along the
satellite track). The distance over which the beams illuminate the ground target is the length of the synthetic aperture.
width θ a from equation (7.74) into the definition of
X a = a /2, which produces
2 V
L
DB
(7.78)
XL
a
SA /2
(7.76)
When the inverse of this expression is multiplied by the
velocity of the platform, it produces the azimuth
resolution. Therefore, this approach yields the same
expression for the SAR azimuth resolution as equation
(7.76). SAR resolution in the azimuth direction is
equal to half of the length of the synthetic antenna.
Thus, SAR has the remarkable property of its finer
azimuth resolution as the physical length of the
antenna decreases. Moreover, it is independent of the
distance between the platform and the ground target.
More rationale for this exceptional result is provided
in Elachi [1988]. It should be noted that the backscat-
ter returned from numerous echoes is accumulated
into the relevant cell's location through the signal
processing at the receiving station.
It is worth mentioning that while backscatter is usually
recorded in terms of the power ratio of the received to the
transmitted signal (i.e., σ 0 ) in any SAR acquisition mode,
the recording of the amplitude and the phase shift is
achieved through a particular mode called single look
complex (SLC). Here, the pixel value is no longer a real
number but a complex number that includes both the
amplitude and the phase of the received signal. The phase
can be recorded because the transmitted radar pulses are
coherent as explained below. SLC data are resolved to
construct images with pixels positioned in slant (not
ground) range (Figure  7.28). In the case of Radarsat‐1
and Radarsat‐2, SLC mode has been available from all
beam modes except ScanSAR (Figure  7.8). Therefore,
this mode may not be useful for operational monitoring
of sea ice, which depends mainly on visual interpretation
of images from ScarSAR mode.
This is an approximate expression for the finest possible
resolution of SAR in the azimuth direction. It assumes
that the synthetic aperture has uniform illumination. A
more appropriate derivation that accounts for nonuni-
form illumination is given in Ulaby et al. [1982].
The second approach involves the use of the Doppler
shift of the returned signal from each antenna position
within the synthetic aperture with respect to the phase of
the transmitted pulse. The Doppler shift can be used
because the successive transmitted pulses of SAR are
coherent (i.e. in‐phase). This allows measuring the phase
of the received signal with respect to the phase of the
transmitted signals.
When the ground cell is ahead of the satellite position,
it produces a positive Doppler shift. When it is behind the
satellite, it produces a negative shift. As the satellite
platform moves, several echoes are received successively
but processed simultaneously to determine the location
of the target.
The Doppler shift of the returned signal of the last
pulse that “sees” the ground target with respect to that of
the first pulse is denoted ϕ , and the rate of this shift with
respect to the exposure (azimuth) time is
2
d
dt
2
V
R T e
(7.77)
The total Doppler bandwidth (DB) between the first and
last pulses results from substituting the expression in
equation (7.75) into equation (7.77):
Search WWH ::




Custom Search