The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part Bl. Beijing 2008 
460 
complex images of the proposed tri-baseline configuration can 
be modelled as: 
S', =Z C1 +N t +Z ri 
S 2 =Z C1 +N 2 + Z T2 (!) 
S 3 = Z C3 + N 3 +Z r3 
where Z C1 , Z C2 and Z c3 are the clutter processes representing 
the SAR complex images acquired from the three antennas, N u 
N 2 and N 3 are due to the presence of thermal noise at the 
receivers and Z n , Z n and Z73 denote the SAR images of the 
moving target relative to the three interferometric antennas. 
ATI analysis exploits the correlation between the complex 
sample sets of the first aperture image, SI, and the second 
aperture image, S2, to produce the interferogram matrix 
/ = S t S’ = |S, \\S 2 1 exp[/0, - (p 2 )] (2) 
Where (p x and (p 2 are the phase angle of the first and second 
aperture images, respectively. 
In presence of a moving target (in absence of noise and 
decorrelation effects ), the interferogram phase angles 
y/ = (p x - cp 2 can be related to the line-of-sight component of 
the target velocity, V r by 
This novel proposed configuration can be adapted to the 
velocity range of the objects of interest. Short baseline and long 
baseline enables highly accurate velocity estimates for slow and 
fast object MTI (as shown in table 1). Hence, an acquisition 
with multiple long & short along-track baselines can enable 
improved and more accurate measurements over a wide 
spectrum of potential scatterer velocities. 
2.3 INSAR 
The goal of INSAR is to measure the difference in range 
between two observations of a given ground point with 
sufficient accuracy to allow accurate topographic reconstruction. 
Operational DEM generation will be performed through the 
interferometric phase and knowledge of the interferometer 
geometry (see Figure 2). Accuracy is obtained by careful 
measurement of the baseline length and orientation and the 
location of the platform relative to the reference coordinate 
system. 
Figure 2. Geometry of the interferometer 
% in An 
w = —AR = v -T 
X X 
(3) 
A simplified expression for the target height h is 
Where A is the radar wavelength, t is the time lag. B is the 
distance between the two SAR antenna. For DSS, while A1 is 
transmitting, A1 and A2 are receiving from pulse to pulse 
resulting in an effective baseline of B/2. A moving target can 
then be detected by comparing the interferometric phase with a 
threshold. 
r 2nB 
(4) 
h = H -rcos[sin _l (——) + a] 
2 kB 
(5) 
where h is the platform height (antenna altitude with respect to 
the WGS84 reference ellipsoid), r is the range, <D is the 
measured interferometric phase, a is the baseline roll angle, 
X is the observing wavelength, and B is the baseline length. 
For Distributed SAR Satellite the radar instrument provided 
data necessary to determine r and <X>, while a dual frequency 
GPS, measured the detailed shape of the interferometer, in 
essence (B, a )or (Ax, Ay, Az). 
DRA ATI 
DSS ATI 
Along-Track Baseline 
2.4m 
150m 
Signal to Noise Ratio 
15dB 
lOdB 
MDV 
2.5m/s 
0.15m/s 
Vmax 
59.37m/s 
0.7919m/s 
Table 1. short and long baseline performance 
Error source 
Error 
Allocation 
Ah(m) 
a ( arcsec) 
2 arcsec 
4.5m 
B(mm) 
8mm 
4.56m 
Platform position(m) 
lm 
lm 
Table 2. Baseline estimation accuracy requirements