We explore the possibility of detecting small Kuiper Belt Objects (KBO) by
serendipitous observation of stellar occultations:
We show that such unpredictable
occultations may allow us to detect a population of very small objects
(typically of

100m radius at 40AU), invisible by any other
observational
method, as long as (i)the assumed population fills up a sufficient area on
the sky plane, (ii)the instrumental sensitivity and acquisition frequency are
high enough and (iii)the observed star has a small angular radius.
This result is basically due to the diffractive broadening of the geometric
shadow of small (assumed numerous) occulting objects. This diffractive
broadening is more pronounced for smaller stellar disks and better
photometric precision.
Assuming there exist about

objects of radius

km,
located between 30 and 50 AU near the Ecliptic, and that the differential size
distribution varies as

with the index

extending down to decameter-sized
objects, we expect a number of valid occultations (i.e. a

event)
between a few to several tens per night, if we may obtain an
r.m.s. signal fluctuation

and observe a star in
the ecliptic with an angular radius

0.01 mas.
Since this occultation rate is very sensitive to the index slope
q and
plummets when

, a KBO occultation observation campaign could
provide a decisive constraint on the actual slope of the KBO size distribution
for sub-kilometer-sized objects.
Blue O class stars are the best candidates for detecting KBOs since
they have the smallest angular radius for a given visual magnitude.
The occultation events are typically very brief (

1s) and they are
shorter but more numerous when observed in the antisolar direction, so rapid
photometry(>1Hz) is required and high speed photometry(

Hz) is
preferred.
The French space mission Corot will provide an excellent
opportunity to observe occultations by KBOs using high precision photometry.