201 
¡.e. the GCP coordinates and the phase are assumed to 
be not correlated, 
al is the variance associated to the L th 
u ParamL 
pseudo-observation, 
1 
Parang 
is the relative weight. 
The weight associated to each pseudo-observation is 
employed to select the appropriate set of model 
parameters. 
From equation (9) the LS solution can be computed 
obtaining the adjusted values of unknown parameters and 
observations (Sansò, 1991): 
ii=N' , 'A T K‘ , '(D-Ç 0 -b) 
C=5o-Q-d t -k- 1 .u 0 
where: 
Gens R., Van Genderen J.L., (1996). "SAR Interferometry 
- Issues, Techniques, Applications", Int. J. Remote 
Sensing, voi. 17, No. 10, pp. 1803-1835. 
Goldstein R.M., Zebker H.A., Werner C.L., (1988). 
"Satellite Radar Interferometry: Two Dimensional 
Phase Unwrapping". Radio Science, voi. 23, No. 4, 
pp.713-720. 
Hanssen R., Feijt A., (1996). "A First Quantitative 
Evaluation of Atmospheric Effects on SAR 
Interferometry". Proceedings of ESA Fringe ‘96, Zurich 
(Switzerland).Http://www.geo.unizh.ch/rsl/fringe96/pa- 
pers/hanssen. 
Hanssen R., (1998). "Atmospheric Heterogeneities in ERS 
Tandem SAR Interferometry". DEOS Report, No. 98.1, 
Delft University Press, Holland. 
Koskinen J., (1995). "The ISAR-Interferogram Generator 
Manual". ESA/ESRIN, Frascati, Italy. 
Sansò F., (1991). "Il Trattamento Statistico dei Dati", (in 
italian) Città Studi Edizioni, Torino, Italy. 
K = DQD T 
N = A T • K' 1 • A is the normal matrix, 
U 0 =D ^o-A iî-b 
Tannous I., Pikeroen B., (1994). "Parametric Modelling of 
Spaceborne SAR Image Geometry. Application: 
SAESAT/SPOT Image Registration". Photogr. 
Engineering & Remote Sensing, vol. 60, No. 6, pp. 
755-766. 
So=yo -9 
yo 
y 
- C yy 
is the observation vector, 
is the vector of the approximate 
values of the observations, 
is the cofactor matrix, 
is the variance-covariance matrix, 
Gq is sigma nought (variance of unit 
weight). 
The solution of the non-linear system is obtained 
iteratively updating, at every iteration, the matrices A, D 
and b. 
The variance-covariance matrix of the solution, obtained 
by variance propagation, is: 
C n - n -=<îo 2 N- 1 
where: 
2 Uj-K-’Uo 
°o =—r— 
I is the number of equations, 
m is the number of unknown parameters. 
REFERENCES 
Crippa B., Crosetto M., Mussio L., (1998). “The Use of 
Interferometric SAR for Surface Reconstruction”. Proc. 
of the ISPRS-Comm. I Symposium, Bangalore (India), 
Int. Arch, of Phot, and RS, vol. XXXII, Part 1, pp. 172- 
177.