Searches for the Theta_s(1540)+ Strange-Pentaquark Candidate in e+e- Annihilation, Hadroproduction and Electroproduction with the BaBar Detector Page: 44 of 130

                                
3.3 Event Preselection

3.3.1   Reconstruction and Selection of Inclusive K0 Candidates
K0 mesons are selected by attempting to isolate candidates that originate from the in-
teraction region (IR). The method and procedures are outlined below. Initially opposite
sign combinations of tracks with no kinematic constraints and the pion mass hypothesis
assigned, are vertexed. A mass window of  50 MeV/c2 around the PDG value for the
K0 mass is used to select candidates. The distance between the point of closest approach
(POCA) of the daughter tracks is required to be less than 6 mm. This retains all visible
K0 candidates in the mass spectrum with a substantial reduction in background.
In this context, the IR is reconstructed as a straight line through the collision region
("Collision Axis"). This is achieved by using an averaged run-by-run coordinate position
in the center of the IR, referred to as the Interaction Point (IP). The IP has an associated
covariance matrix describing the ellipsoidal shape of the IR within the BABAR coordinate
system. A direction1 can be derived from this matrix and a straight line following this
through the IP is used to define the Collision Axis.
The reconstructed K0 trajectory (the momentum vector through the r+7r- vertex)
should extrapolate close to the Collision Axis for real K0 candidates for the majority
of physics processes that are of interest. Therefore, to ensure the required consistency
with an origin in the IR, selection criteria are applied to the distance of closest approach
(DOCA) of the K0 trajectory to the Collision Axis. Initially this is determined in the
transverse plane, by calculating Roy - the radial separation distance.
2hy =    [XpO - XBeam(Za$)] +      Yat -   Beam(Za'$)             (3.1)
yR ,70PK                                 -YemZ                     31
where (X" a, Yp, Za o) is the point of closest approach between the fitted momentum
KS   KS   KS
vector of the K0 candidate and the collision axis, and:
XBeam(Z o) = XIp +       z   x (Za  - ZIp)                  (3.2)
YBeam(Zp ) = YIP +         x (dZy   -ZIP)                 (3.3)
Here d and z define the slope of the Collision Axis. Further to this, a requirement is
placed on the separation between ZO and the z-coordinate of the IP (ZIp).
S
The spread in z ,JZ0' - ZIpl, is determined by taking into account the known mea-
surements of the Beam Spot2 and degradation in resolution by reconstruction effects. The
'See http://www.slac.stanford.edu/~wmd/bbkinematics/CollisionAxisTilt.note
2The size of the luminous region is obtained from vertex reconstruction:
ox~110 - 120 pm
o  ~25 - 35 pm
a- ~ 7.5 - 8.0 mm : in the early days, this was 8.3 mm, but it has decreased with time because of
additional r.f., which yields a smaller bunch length

38