Skip to content

Signal and Background Processes

The Signal Process

As established in the previous section, this analysis targets the fully leptonic different-flavor final state of the Higgs boson produced via gluon-gluon fusion:

\[gg \to H \to WW^* \to e^{\pm}\mu^{\pm}\nu\bar{\nu}\]

This signal process is characterized by:

  • Two leptons of different flavor: One electron and one muon, each carrying high transverse momentum
  • Two neutrinos: Manifested as missing transverse energy (MET) in the detector
  • No jets: Ideally, no hadronic jets

Several Standard Model processes can mimic this signature, either exactly or through partial overlap with the final state. Understanding and suppressing these backgrounds is crucial for isolating the signal.

Background Processes

The backgrounds relevant to this analysis are several Standard Model processes that can mimic or partially overlap with the signal signature.

Drell-Yan to Tau-Tau

Final State: \(q\bar{q} \to Z^*/\gamma^* \to \tau^{+}\tau^{-}\) where, \(\tau \to \ell\nu\)

Why It Mimics the Signal:

  • When \(\tau\) leptons decay leptoically to different flavors (\(\tau \to e^\pm \nu\), \(\tau \to \mu^\pm \nu\)), the visible dilepton state (\(e^{\pm}\mu^{\pm}\)) is identical to the signal topology.

  • Both processes produce leptons with missing transverse energy.

Suppression Strategy:

  • Azimuthal Separation: In Drell–Yan processes, the final-state particles are typically produced back-to-back, whereas in Higgs decays they are more collinearly produced.

  • Transverse Mass Selection: The transverse mass \(m_T^H\) distribution for signal peaks at the Higgs mass, while DY events have a broader, lower distribution. This provides additional separation

Top Quark Pair Production

Final State: \(q\bar{q} \to t\bar{t} \to (b\ell\nu)(b\bar{\ell}'\bar{\nu})\)

Why It Mimics the Signal:

  • Both the signal and top pair production result in two leptons of opposite charge with missing transverse energy
  • Top events genuinely produce two neutrinos (one per top decay), resulting in significant MET similar to the signal
  • The kinematic distributions can partially overlap with the signal

Suppression Strategy:

  • Jet Veto / \(b\)-Jet Veto: Requiring the absence of jets (especially \(b\)-tagged jets) suppresses top events while retaining signal events. Signal production via gluon fusion naturally has minimal jet activity

  • MET and Transverse Mass Cuts: While top events do have significant MET, their transverse mass and dilepton momentum distributions differ from the signal, allowing additional discrimination

Diboson Production (VZ: WZ, ZZ)

Final State: \(q\bar{q} \to VV \to (W \to \ell\nu)(Z \to \ell^+\ell^-)\)

Why It Mimics the Signal:

  • WZ production with \(W \to \ell\nu\) and \(Z \to \ell^+\ell^-\) produces exactly two leptons and MET

  • MET from the \(W\) decay neutrino can be substantial

Suppression Strategy:

  • Dilepton Invariant Mass Windows: Selecting events outside the \(Z\) mass region suppresses this background

  • Additional Lepton/Jet Veto: Rejecting events with third leptons or jets reduces contamination

WW Production (WW, ggWW)

Final State: \(q\bar{q}/gg \to WW \to (W \to e\nu)(W \to \mu\nu)\)

Why It Mimics the Signal:

  • Both WW and signal processes produce exactly two leptons of opposite charge and missing transverse energy
  • The distributions are kinematically similar: two leptons and two neutrinos

Suppression Strategy:

  • Kinematic Distributions: The transverse mass \(m_T^H\), dilepton invariant mass \(m_{\ell\ell}\), and angular separation \(\Delta\phi_{\ell\ell}\) all differ between signal and WW

Diboson with Photons (VG)

Final State: \(q\bar{q} \to VG \to (\ell\nu)(\gamma)\)

Why It Can Mimic the Signal:

  • In rare cases, photons may be misidentified as leptons

  • The MET from the \(W\) or \(Z\) decay neutrino can appear similar to the signal

Suppression Strategy:

  • Lepton Identification: Strict electron and muon identification criteria suppress photon misidentification

Fakes

Final State: Jets misidentified as leptons, producing a false dilepton + MET signature

Why It Mimics the Signal:

  • Jets from b-quark or charm-quark decays can sometimes pass loose lepton selection criteria

  • When multiple jets are selected as "leptons", the resulting four-vectors and MET can resemble the signal

Suppression Strategy:

  • Tight Lepton Identification: Applying strict criteria on lepton isolation reduces this background

  • Lepton Multiplicity Requirements: Selecting events with exactly two leptons, with well-defined kinematic properties, suppresses multi-jet contamination