| 1 |
\section{Event Selection}
|
| 2 |
\label{sec:eventSel}
|
| 3 |
|
| 4 |
Samples of MC events are used to guide the design of the analysis.
|
| 5 |
These events are generated using either the
|
| 6 |
\PYTHIA6.4.22~\cite{Pythia} or \MADGRAPH4.4.12~\cite{Madgraph} event
|
| 7 |
generators. They are then simulated using a GEANT4-based
|
| 8 |
model~\cite{Geant} of the CMS detector, and finally reconstructed and
|
| 9 |
analyzed using the same software as is used to process collision data.
|
| 10 |
|
| 11 |
We apply a preselection based on that of the $t\bar{t}$ cross section
|
| 12 |
measurement in the dilepton channel~\cite{ref:top}. Events
|
| 13 |
with two opposite-sign, isolated leptons ($e^+e^-$,
|
| 14 |
$e^{\pm}\mu^{\mp}$, or $\mu^+\mu^-$) are selected. At least one of the leptons must
|
| 15 |
have $\pt > 20\GeVc$ and both must have $\pt > 10\GeVc$, and the
|
| 16 |
electrons (muons) must have $|\eta| < 2.5$ ($|\eta| < 2.4$). In events
|
| 17 |
with more than two such leptons, the two leptons with the
|
| 18 |
highest \pt are selected. Events with an $e^+e^-$ or $\mu^+\mu^-$ pair
|
| 19 |
with invariant mass between 76\GeVcc and 106\GeVcc or below
|
| 20 |
12\GeVcc are removed, in order to suppress Drell--Yan (DY)
|
| 21 |
$Z/\gamma^{*}\to\ell\ell$ events, as well as low mass dilepton
|
| 22 |
resonances.
|
| 23 |
|
| 24 |
Events are required to pass at least one of a set of $ee$, $e\mu$ or $\mu\mu$
|
| 25 |
double-lepton triggers. The efficiency for events containing two
|
| 26 |
leptons passing the analysis selection to pass at least one of these
|
| 27 |
triggers is measured to be approximately 100\%, 95\%, and 90\%
|
| 28 |
for $ee$, $e\mu$ or $\mu\mu$ double-lepton triggers, respectively.
|
| 29 |
In the following, the MC yields are weighted by these trigger efficiencies.
|
| 30 |
|
| 31 |
Because leptons produced in the decays of low-mass particles, such as
|
| 32 |
hadrons containing $b$ and $c$ quarks, are nearly always inside jets, they can be
|
| 33 |
suppressed by requiring the leptons to be isolated in space from other
|
| 34 |
particles that carry a substantial amount of transverse momentum. The
|
| 35 |
details of the lepton isolation measurement are given in
|
| 36 |
Ref.~\cite{ref:top}. In brief, a cone is constructed of size
|
| 37 |
$\Delta{}R\equiv\sqrt{(\Delta\eta)^2+(\Delta\phi)^2}=0.3$ around the
|
| 38 |
lepton momentum direction. The lepton relative isolation is then
|
| 39 |
quantified by summing the transverse energy (as measured in the
|
| 40 |
calorimeters) and the transverse momentum (as measured in the silicon
|
| 41 |
tracker) of all objects within this cone, excluding the lepton, and
|
| 42 |
dividing by the lepton transverse momentum. The resulting quantity
|
| 43 |
is required to be less than 0.15, rejecting
|
| 44 |
the large background arising from QCD production of jets.
|
| 45 |
|
| 46 |
We require the presence of at least two jets with $\pt > 30\GeVc$ and $|\eta| < 3.0$,
|
| 47 |
separated by $\Delta R >$ 0.4 from leptons passing the analysis
|
| 48 |
selection with $\pt > 10\GeVc$. The anti-$k_T$ clustering
|
| 49 |
algorithm~\cite{antikt} with $\Delta{}R = 0.5$ is used for jet
|
| 50 |
clustering. The jets and \MET are reconstructed with the Particle Flow
|
| 51 |
technique~\cite{CMS-PAS-PFT-10-002}.
|
| 52 |
The event is required to satisfy $\HT > 100\GeV$, where \HT\ is defined as
|
| 53 |
the scalar sum of the transverse energies of the selected jets. In
|
| 54 |
addition, the \MET\ in the event is required to exceed 50\GeV.
|
