US 8,049,657 B2 · Method for processing TOPS (terrain observation by progressive scan)-SAR (synthetic aperture radar)-raw data

Abstract

Sub-aperture processing is carried out. Within each sub-aperture, range compression and a correction for the target range variation are carried out. Baseband azimuth scaling is used for processing the azimuth signal, wherein a long azimuth reference function and thus a wide azimuth dimension are prevented. The scaling range is not constant and depends on the range, which is not equal to the original range vector. It is calculated such that, in combination with a subsequent derotation step, constant azimuth scanning is achieved for all ranges. The selected derotation function, which is applied in the azimuth time domain, makes it possible for all the targets to be in base band, in this way varying the effective chirp rate. Since the phase is purely quadratic because of the azimuth scaling step, it is thus possible to use an optimal filter which takes account of the effective chirp rate. IFFT results in a focused image, and a final phase function in the time domain allows phase maintenance. Application for SAR, SONAR and seismic raw data processing in the TOPS mode, as well as other modes which make use of the antenna polar diagram being scanned in the azimuth and/or elevation direction.

A method for processing TOPS (Terrain Observation by Progressive Scan)-SAR(Synthetic Aperture Radar)—raw data into highly accurate image data, characterized by two-dimensional processing of the TOPS-SAR raw data through a chirp scaling method and a baseband azimuth scaling method using a processor to carry out the following successive steps:

a) subdividing the TOPS-SAR raw data into azimuth subaperture data (3.1);
b) performing a short-term azimuth FFT (3.2);
c) multiplying the signal by a chirp scaling function Hcs which is given by
Hcs(fa,τ;r0)=exp└−j·π·Keff(fa;rref)·a(fa)·(τ−τref(fa))┘
wherein fa is the azimuth frequency, ro is the shortest range from a target, rref is a reference range, is the time in the distance direction which is also referred to as the range direction, i.e. the echo propagation time, the factor a(fa) is the frequency dependency of the range variation according to
a
⁡
(
f
a
)
=
1
1
-
(
λ
·
f
a
2
·
v
)
2
-
1
,
λ is a wavelength, ν is the speed on the ground, Keff(fa;rref) is the modulation rate in the range direction depending on the azimuth frequency fa and the reference range rref, and τref(fa) is a reference echo propagation time (3.3, 3.4) depending the azimuth frequency fa;
d) performing a range FFT for obtaining the signal in the two-dimensional frequency range (3.5);
e) multiplying the signal by the phase function Hrcmc which is given by
H
rcmc
⁡
(
f
a
,
f
r
;
r
ref
)
=
exp
[
-
j
⁢
π
K
eff
⁡
(
f
a
;
r
ref
)
·
(
1
+
a
⁢
(
f
a
)
)
⁢
f
r
2
]
·
exp
⁡
[
j
⁢
4
·
π
c
·
r
ref
·
a
⁢
(
f
a
)
·
f
]
for performing the correction of the range variation, of the range compression and of the secondary range compression, fr being the range frequency and c being the light velocity (3.6, 3.7);
f) performing a range IFFT so as to include the signal into the range Doppler region (3.8);
g) multiplying the signal by a phase correction function Hcorr which is given by
H
corr
⁡
(
f
a
;
r
0
)
=
exp
⁡
[
j
·
π
·
K
eff
⁡
(
f
a
;
r
ref
)
·
(
1
+
a
⁡
(
f
a
)
)
·
a
⁢
(
f
a
)
·
(
2
c
·
(
r
0
-
r
ref
)
)
2
]
wherein r0 is the shortest range to a point target (3.9, 3.10);
h) multiplying the two-dimensional data which are taken over in the range Doppler region and after performing the main steps of range compression and range variation correction of a target, by a phase function Ha(fa;r0) defined by
H
a
⁡
(
f
a
;
r
0
)
=
exp
⁡
[
j
·
4
·
π
λ
·
r
0
·
(
β
⁡
(
f
a
)
-
1
)
]
·
exp
⁡
[
-
j
·
π
K
scl
⁡
(
r
)
·
f
a
2
]
wherein
β
⁡
(
f
a
)
=
1
-
(
λ
·
f
a
2
·
v
)
2
⁢
⁢
K
scl
⁡
(
r
)
=
-
2
·
v
2
λ
·
r
scl
⁡
(
r
)
⁢
⁢
r
scl
⁡
(
r
)
=
r
scl
⁢
⁢
0
r
rot
⁢
⁢
0
⁢
r
rot
⁢
⁡
(
r
)
⁢
⁢
r
rot
⁢
⁡
(
r
)
=
r
rot
⁢
⁢
0
-
r
1
-
r
scl
⁢
⁢
0
/
r
rot
⁢
⁢
0
and r is the range vector, Kscl(r) is a range-dependent Doppler rate, rrot0 is a rotation range due to the TOPS geometry, rscl0 is a scaling range selected according to the final desired image range, rscl(r) a scaling range vector, whereby the scaling range values are within a range dimension of an image and whereby rrot(r) is a range-dependent rotation range vector (3.11, 3.12);
i) transforming the subaperture data through short azimuth IFFTs back into the azimuth and range time domain (3.13);
j) joining the azimuth subapertures (3.14);
k) multiplying the two-dimensional data by a derotation function Hderot(t,r) given by
H
derot
⁡
(
t
,
r
)
=
exp
⁢
⌊
-
j
·
π
·
K
rot
⁡
(
r
)
·
(
t
-
t
mid
)
2
⌋
K
rot
⁡
(
r
)
=
-
2
·
v
2
λ
·
r
rot
⁡
(
r
)
wherein t is the time vector in the azimuth direction and tmid is the middle time of the burst (3.15, 3.16);
l) transforming the derotated data through long azimuth FFTs into the range Doppler region (3.17);
m) subsequently, multiplying the two-dimensional data through an azimuth compression function Hcomp(fa,r) which is given by
H
comp
⁡
(
f
a
,
r
)
=
W
⁡
(
f
a
)
·
exp
⁡
[
j
·
π
K
eff
⁡
(
r
)
·
f
a
2
]
,
⁢
-
PRF
2
+
f
DC
<
f
a
<
PRF
2
+
f
DC
wherein
K
eff
⁡
(
r
)
=
K
scl
⁡
(
r
)
-
K
rot
⁡
(
r
)
is the effective chirp rate after the scaling and derotation operations, W(fa) is the weighting function for sidelobe suppression and fDC is the average Doppler centroid of the data capturing, wherein, in the TOPS mode, this is the only optimal point at which the application of the weighting function for sidelobe suppression leads to correct results because all targets are in baseband (3.18, 3.19);
n) transforming the data through a long azimuth IFFT into the image data (3.20);
and
o) multiplying the two-dimensional image data through a phase function Hphase(t,r) which is given by
H
phase
⁡
(
t
,
r
)
=
exp
⁡
[
j
·
π
·
K
t
⁡
(
r
)
·
(
1
-
r
scl
⁢
⁢
0
r
rot
⁢
⁢
0
)
2
·
(
t
-
t
mid
)
2
]
wherein
K
t
⁡
(
r
)
=
-
2
·
v
2
λ
·
(
r
rot
⁡
(
r
)
-
r
scl
⁡
(
r
)
)
,
to achieve a phase maintenance (3.21, 3.22).

530 more elements

Method for processing TOPS (terrain observation by progressive scan)-SAR (synthetic aperture radar)-raw data

Life of this patent

Method for processing TOPS (terrain observation by progressive scan)-SAR (synthetic aperture radar)-raw data

Life of this patent

Abstract

Claims2