148 
■ Most of the effect (shift of the fringe pattern) 
produced by the movements of the optical 
interferometer mirrors is common to both FOVs. 
Therefore, since the basic measurements are 
angular separations of the stars belonging to 
different FOVs, which are routed into the same 
instrument by the beam combiner, it is possible to 
relax the measurement/control requirements of the 
relative position/orientation of the telescope 
mirrors. 
2.2 Astrometric Precision 
The focal plane (FP) image produced by an 
interferometer is an interference pattern, i.e. fringes, 
which modulates the overall light distribution described 
by the Airy disk corresponding to each aperture. The 
principal Airy disk diameter is T A = 2AW ID (D is the 
apertures diameter), while the fringe period is 
T y = X / B where B is the baseline of the 
interferometer. In the case of GAIA, at an operating 
wavelength of Kff = 150nm , the previous quantities 
are 581 mas and 63.1 mas respectively. 
A fringe pattern has an intrinsic higher resolution with 
respect to images produced with a single aperture 
telescope. However, the resolution of the optical 
interferometer is not sufficient for achieving accuracy at 
10 pas level, which requires also a proper intensity 
(high number of photons). 
As we have seen, the feasibility studies suggest two 
different options for the optical configuration. In both 
cases, the lower limit for the location error in the along 
scan direction, for the single measure, depends on the 
signal to noise ratio (SNR) and the geometric 
characteristics of the telescope, precisely: 
Figure 3: Beam Combiner 
However the quantity \ D 2 + B 2 is greater than L, so 
that the product L-SNR is of the same order in both 
cases. 
In other words, both configurations achieve in principle 
the precision requested for the GAIA mission. The 
difference is on the strategy for controlling systematic 
errors. Laboratory tests show that the interferometric 
design can achieve the accuracy value allocated in the 
error budget (see section 3). Experiments are underway 
for analogous tests on the monolithic configuration. 
2.3 Focal plane and detection system 
The detection system consists of a CCD (Charge 
Coupled Device) mosaic placed on the focal plane of 
the telescope (Cesare, 1998). 
The detection area is functionally subdivided in three 
parts (figure 4): 
a > F 
L-SNR 
SNR < -J~N 
(D 
where for the single aperture option L = D (aperture 
diameter), while for the interferometric option L = 
I 2 2 
+B~ , N = number of photons, and F is a 
parameter related to the geometry of the telescope and 
the detection system. 
The product L-SNR should in general be maximized 
in order to obtain the precision requested for GAIA 
taking into account the various constraints on size, 
mass and complexity of the satellite payload. 
For example, the monolithic option has a greater 
aperture (L = 1.7 m) compared to the interferometer (D 
= 0.65 m), therefore the signal to noise ratio is higher 
for the former. 
the Astrometry area, placed in the central part of 
the focal plane where the fringe visibility is higher, 
is dedicated to the very accurate measurement of 
the along-scan coordinate of stars up to the 18 th 
magnitude (single-exposure S/N ratio > 4). 
the Photometry area, constituted by two zones 
(‘preceding’ and ‘following’ Photometry areas) 
placed at the two sides of the Astrometry area and 
covered by a set of pass-band filters, is dedicated 
to the measurement of the star light flux in different 
spectral regions (U, B, V, Rc, Ic, etc..) 
the Star Mapper, constituted by two “strips” placed 
at the outer edges of the FP in the scan direction, 
dedicated to the attitude determination as well as 
to photometric measurements.