The nearly exact repeat orbit allows formation of an interferometric baseline as shown in the figure. The ERS-1 sensor provides a platform to record SAR data which is useful for interferometry.
This exploitation of the orbit repeat feature is known as multi-pass INSAR.
For a point target that is at (x, y, z), the phase difference f between the signal s1 and s2 received at r1 and r2 is
Where λ is the radar signal wavelength. The interferometric phase is obtained by complex correlation of the first complex SAR image, obtained during the first overpass, relative to the second image, belonging to the second pass, after precise co-registration.
The interferometric phase can be used to determine the precise look angle θ by first solving the cosine of the angle between the baseline vector and the look vector:
The height z at the location r1 is determined from:
Where h is the altitude of the platform above the reference plane.
The phase in the interferogramme is known only modulo 2p, therefore it is necessary to determine the correct multiple of 2p to add to the phase to obtain consistent height estimates.
The correlation coefficient g of the complex backscatter intensities s1 and s2 at r1 and r2 is defined by:
In the following we will call g the interferometric correlation. The correlation coefficient g in terms of the baseline correlation factor a is equal to
where the geometrical baseline correlation factor a is given by:
where SNR is the signal to noise ratio, dr the slant range resolution, and alpha is the radar incidence angle.
The ERS-1 radar sensor has a slant range pixel spacing of 7.9m, nominal incidence angle of 23 degrees, and nominal range of 853 Km to the center of the image swath.
For an 8 look image with a SNR of 20, the theoretical height resolution is better than 2.5 m for a wide range of correlation values y.