We first present an overview of background processes which could simulate the presence
of a CC 
interaction.
We then explore the strategy to distinguish background from the signal.
This approach leads us to a set of requirements for a detector.
A more detailed quantitative study of the number of background events will be
described in a following section.
The most abundant events in the detector are inclusive CC interactions
of 
.
They have a negatively charged
muon in the final state, which does not produce a decay topology, nor a sizable missing energy.
Only processes with a special topology, such as charm production can be confused
with events with a 
in the final state.
Decays of the primary muon or of pions and kaons are easily rejected by the requirement of
a minimum decay transverse momentum.
Charmed particles produced in and 
CC interactions decay with a lifetime
similar to the 
, but can be distinguished from by their positive charge.
Negatively charged charmed particles can be produced in the charged-current interactions
of 
and 
.
The additional presence of a 
or 
allows to reject these events.
In the cases where the primary leptons are not identified, kinematical
criteria can be used as further reduction.
In particular, the presence of a 
in their decay products can be used to
discriminate these events from -decays.
Another source of negatively charged charmed particles is the associated charm production in neutral-current (NC) reactions. These events are rare and can be suppressed by the detection of a second decay topology in the same event. A powerful rejection can be obtained by comparing the direction of the missing energy as predicted from the decay angle with the total transverse momentum vector of the event. The latter is dominated by the unmeasured primary neutrino momentum, and is uncorrelated with the decay topology.
Neutral-current interactions without charmed particle production can only simulate a
decay topology, when one of the negatively charged
hadrons interacts close to the primary interaction vertex.
Only a very small part of these secondary interactions have no additional visible activity.
These are called ``white kinks'' in emulsion terminology.
These events do not exhibit the characteristic decay kinematics.
CC interactions with white kinks are rejected making use of the presence of the primary lepton.
Although the kinematics of the remaining events is more similar to that of
CC events than the NC events with white kinks, it is expected that their
final contribution to the background is negligible.