Tests for twikis

2012 Trigger matching

For electron and muon trigger matching, several methods such as matchDimuon, matchDimuon, and match are provided in TrigMuonEfficiency (the tag to be used for this package is given below. In 2012, we use both symmetric and asymmetric pT threshold dimuon triggers. In order to see whether a given di-muon object (muon1 and muon2) matches to a dimuon trigger item, one should first call the function

bool matchDimuon(const TLorentzVector& muon1,
                        const TLorentzVector& muon2,
                        const std::string& chain,
                        std::pair<bool, bool>& result1,
                        std::pair<bool, bool>& result2);

and then require

// the first condition is for muon1 as seeded, muon2 as full scan
// the second condition is for muon2 as seeded, muon1 as full scan
bool ismatched = ((result1.first && result2.second) || (result2.first && result1.second));

The higher pT threshold should be applied on the muon that acts the seed. For example adding the pT threshold requirement on EF_mu18_tight_mu8_EFFS trigger item should be done

bool ismatched_and_passthreshold =  (  ( result1.first && result2.second && ( muons1.Pt() > 19000.0 ) && ( muons2.Pt() > 9000.0 ) )
                                                              || ( result2.first && result1.second && ( muons1.Pt() > 9000.0 ) && ( muons2.Pt() > 19000.0 ) ) )

Single electron, single muon or di-electron matching should be done as in 2011.


In release 15.3.0 and above, the four momentum parameters (E, η, φ ) and its error matrix are determined by combining information coming from EM and inner detector. For a track, a 5x1 vector with (d0, z0, φtrack, ηtrack, Ptrack) parameter is created. A Jacobian matrix is needed to go from a track default error matrix (d0, z0, φtrack, θtrack, q/Ptrack) to (d0, z0, φtrack, ηtrack, Ptrack) representation. The same vector and matrix is also built for the associated cluster (10e10, 10e10, φclus, ηclus, Eclus). The d0 and z0 parameters in the cluster vector is chosen big enough so that no effect can be seen in the combination.

The method to combine the information is based on the sigma that is computed from the energy in EM (Eclus) and the track momentum (Ptrack). Depending on the sigma value (which is a job option parameter and can be set to any value, by default it is 3), the E is taken from either from Eclus ( for σ > 3) or is set by combining (Eclus and Ptrack). The η is set depending on the number of hits in the inner detector. In case less than 3 hits are found, ηclus and ηtrack are combined, otherwise ηtrack is set. The φ angle always comes from the φtrack, no combination is done for φ angle. The method is shown in a diagram below:


The same procedure is applied for double and single track conversion. From release 15.5.0 and beyond, the E/P combining method is not applied to the single track conversion.

The 4-momentum (E, η, φ, M) and its error matrix of an egamma object are set in EMFourMomBuilder. The 4x4 error matrix is constructed in primary representation (E, η, φ, M) from the 5x5 combined error matrix The 4x4 matrix is then persistified in release 15.5.0 from which the error on the 4-momentum variables can be retrieved. For example, to get the pError

const I4MomentumError *mom_err =  eCand[i]->errors();
float pErr = mom_err->pError();

The I4Momentum interface, which is part of the interface of egamma, is extended with an I4MomentumError interface in order to access individual errors and error matrices in different parameterizations.The 4-momentum functionality is provided by the package FourMom, from which egamma inherits it. The primary representation of the 4-momentum is EEtaPhiM which can then be converted to any other three representations, PxPyPzE, EEtaPhiM and PtCotThPhiM, using the I4MomentumError interface.

Combined Covariance matrix items availbale 15.5.0:

  • egammaParameters::EMTrack_Combined_Covd0d0 : Covariance matrix item (d0,d0)
  • egammaParameters::EMTrack_Combined_Covd0z0 : Covariance matrix item (d0,z0)
  • egammaParameters::EMTrack_Combined_Covd0phi : Covariance matrix item (d0,phi)
  • egammaParameters::EMTrack_Combined_Covd0eta : Covariance matrix item (d0,eta)
  • egammaParameters::EMTrack_Combined_Covd0P : Covariance matrix item (d0,P)
  • egammaParameters::EMTrack_Combined_Covz0z0 : Covariance matrix item (z0,z0)
  • egammaParameters::EMTrack_Combined_Covz0phi : Covariance matrix item (z0,phi)
  • egammaParameters::EMTrack_Combined_Covz0eta : Covariance matrix item (z0,eta)
  • egammaParameters::EMTrack_Combined_Covz0P : Covariance matrix item (z0,P)
  • egammaParameters::EMTrack_Combined_Covphiphi : Covariance matrix item (phi,phi)
  • egammaParameters::EMTrack_Combined_Covphieta : Covariance matrix item (phi,eta)
  • egammaParameters::EMTrack_Combined_CovphiP : Covariance matrix item (phi,P)
  • egammaParameters::EMTrack_Combined_Covetaeta : Covariance matrix item (eta,eta)
  • egammaParameters::EMTrack_Combined_CovetaP : Covariance matrix item (eta,P)
  • egammaParameters::EMTrack_Combined_CovPP : Covariance matrix item (P,P)

Low pT Photon Reconstruction ( release >= 17.0.0)

Low pT photons covering energy range between 1 GeV to 10 GeV, are built from a seed provided by EM topological cell clusters. The details of the topo seeded clusters algorithm can be found in the note ATL-LARG-PUB-2008-002 (W.Lampl et al., Calorimeter Clustering Algorithms: Description and Performance) and in twiki.

Low pT photon reconstruction and identification algorithm can be found in:

The egammaTopoSeededBuilder (LXR, doxy)
The corresponding python script is =egammaTopoSeededGetter.py (LXR, doxy)

These photons can be accessed from ESD, AOD and D3PD.

  • ESD : low pT photons are saved in separate container
Collection:    TopoSeededPhotonCollection
Details:        TopoSeededPhotonDetails

  • AOD : low pT photons are added to PhotonAODCollection which can then be obtained via
this will return true if the object was found by low pT photon reconstruction

  • D3PD : low pt photons can be selected via ph_author==128 (bit 7 is set)

Shower shapes, Four momentum and Object Quality are calculated for these photons as is done for standard photons but not the PID (thus no isEM for low pT photons).

-- TulayCuhadarDonszelmann - 08 Jan 2009

Topic attachments
I Attachment History Action Size Date Who Comment
JPEGjpg ep.jpg r1 manage 49.0 K 2009-08-16 - 16:10 TulayCuhadarDonszelmann  
JPEGjpg ep1.jpg r1 manage 50.7 K 2009-08-16 - 16:21 TulayCuhadarDonszelmann  
JPEGjpg ep2.jpg r1 manage 47.2 K 2009-08-16 - 16:24 TulayCuhadarDonszelmann  
JPEGjpg epnew.jpg r1 manage 62.9 K 2009-08-16 - 20:36 TulayCuhadarDonszelmann  
Edit | Attach | Watch | Print version | History: r10 < r9 < r8 < r7 < r6 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r10 - 2012-06-11 - TulayCuhadarDonszelmann
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    Main All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback