EgammaHitsCalibration Class Reference

General Interface for calibrations at the LVL2 Egamma Calo Fex algo. More...

#include <EgammaHitsCalibration.h>

Inheritance diagram for EgammaHitsCalibration:

Collaboration diagram for EgammaHitsCalibration:

Public Member Functions
virtual StatusCode	initialize () override
	Initialization of the tool. More...
virtual StatusCode	finalize () override
	Finalization of the tool. More...
virtual void	makeCorrection (xAOD::TrigEMCluster *, const void *v=nullptr) const override
	method to perform the correction. More...

Private Attributes
Constant< CxxUtils::Array< 3 > >	m_correction { this, "correction" }
Constant< CxxUtils::Array< 2 > >	m_sampling_depth { this, "sampling_depth" }
Constant< float >	m_eta_start_crack { this, "eta_start_crack" }
Constant< float >	m_eta_end_crack { this, "eta_end_crack" }
Constant< float >	m_etamax { this, "etamax" }
Constant< bool >	m_use_raw_eta { this, "use_raw_eta" }
Constant< bool >	m_preserve_offset { this, "preserve_offset" }
Constant< bool >	m_fix_v6_pathologies { this, "fix_v6_pathologies" }

Detailed Description

General Interface for calibrations at the LVL2 Egamma Calo Fex algo.

Definition at line 24 of file EgammaHitsCalibration.h.

Member Function Documentation

finalize()

StatusCode EgammaHitsCalibration::finalize ( )

overridevirtual

Finalization of the tool.

Definition at line 38 of file EgammaHitsCalibration.cxx.

                                          {
    ATH_MSG_DEBUG( "Finalize Tool : " << name()  );
    return StatusCode::SUCCESS;
}

initialize()

StatusCode EgammaHitsCalibration::initialize ( )

overridevirtual

Initialization of the tool.

Definition at line 27 of file EgammaHitsCalibration.cxx.

                                            {
 
    CHECK (base_class::initialize());
 
    ATH_MSG_DEBUG( "Initialize Tool : " << name()  );
 
 
 
    return StatusCode::SUCCESS;
}

makeCorrection()

void EgammaHitsCalibration::makeCorrection ( xAOD::TrigEMCluster *  clus, const void *  v = nullptr  ) const

overridevirtual

method to perform the correction.

The correction type is defined by the tool which also uses this interface. In some cases, the tool needs more than the cluster to perform the calibration. This can be passed via the void pointer

Definition at line 43 of file EgammaHitsCalibration.cxx.

                         {
 
    const float etamax = m_etamax();
    const float eta_start_crack = m_eta_start_crack();
    const float eta_end_crack = m_eta_end_crack();
    const CxxUtils::Array<3> correction = m_correction();
 
    float the_aeta = (clus->eta());
 
    the_aeta = (the_aeta>0)?(the_aeta):(- the_aeta);
 
    if (the_aeta >= etamax) return; 
 
#ifndef NDEBUG
    ATH_MSG_DEBUG(  "************************************************************************************************"  );
    ATH_MSG_DEBUG(  " USING CALIBHITS CALIBRATION "  );
    ATH_MSG_DEBUG(  "************************************************************************************************"  );
#endif
    int si;
    if (the_aeta < eta_start_crack)
        si = 0; //barrel
    else if (the_aeta > eta_end_crack)
        si = 1; //endcap
    else {
        // No corrections are applied for the crack region.
        return;
    }
 
     
    int ibin = (static_cast<int> (the_aeta / etamax * 100));
    int ibin_frontCorr = ibin; 
    if (m_fix_v6_pathologies()) {
      if ( the_aeta>2.35) 
        ibin_frontCorr = (static_cast<int> (235. / etamax )); 
   }
 
#ifndef NDEBUG
        ATH_MSG_DEBUG( "Check etas -------------------------------------------------------------------" );
        ATH_MSG_DEBUG( "Eta --> " << the_aeta << "  Bin --> " << ibin <<" Cluster eta = " << clus->eta()  );
        
        ATH_MSG_DEBUG( "Check calibration coefficients -----------------------------------------------" );
        ATH_MSG_DEBUG( "Accordion   :  " << correction[0][ibin][0] <<"  " << correction[0][ibin][1] << " "  << correction[0][ibin][2] << " "  << correction[0][ibin][3]  );
        ATH_MSG_DEBUG( "OutOfCOne   :  " << correction[1][ibin][0] <<"  " << correction[1][ibin][1] << " "  << correction[1][ibin][2] << " "  << correction[1][ibin][3]  );
        ATH_MSG_DEBUG( "Leakage     :  " << correction[2][ibin][0] <<"  " << correction[2][ibin][1] << " " << correction[2][ibin][2] << " "  << correction[2][ibin][3]  );
        ATH_MSG_DEBUG( "Front offset:  " << correction[3][ibin_frontCorr][0] <<"  " <<correction[3][ibin_frontCorr][1]  << " " << correction[3][ibin_frontCorr][2] << " "  << correction[3][ibin_frontCorr][3]  );
        ATH_MSG_DEBUG( "Front Slope  :  " << correction[4][ibin_frontCorr][0] <<"  " << correction[4][ibin_frontCorr][1] << " " << correction[4][ibin_frontCorr][2] << " "  << correction[4][ibin_frontCorr][3]  );
        ATH_MSG_DEBUG( "Second order:  " << correction[5][ibin_frontCorr][0] << "  " << correction[5][ibin_frontCorr][1] << "  "  << correction[5][ibin_frontCorr][2] << " " << correction[5][ibin_frontCorr][3]  );
#endif
 
    EgammaHitsShowerDepth showerDepth;
    float shower_lbary = showerDepth.depth(the_aeta,
                                             eta_start_crack,
                                             eta_end_crack,
                                             m_sampling_depth(),
                                             etamax,
                                             clus);
    if (!shower_lbary)
      return;
 
    if (shower_lbary < 5. || shower_lbary > 25.){
      shower_lbary = 15.;
#ifndef NDEBUG
      ATH_MSG_DEBUG( "replace pathological depth by 15 X0"  );
#endif
    }
 
    // Compute total energy in the Accordion (eacc_base) and
    // presampler (eps_base)
 
    float eacc_base = 0;
    for (int sampling=1; sampling<4; sampling++) {
        eacc_base += clus->energy(EgammaHitsShowerDepth::m_samps[si][sampling]);
#ifndef NDEBUG
        ATH_MSG_DEBUG( "Barrel/endcap = " << si << "  Sampling = " <<   sampling << "   Energy -->> " <<  clus->energy(EgammaHitsShowerDepth::m_samps[si][sampling])  );
#endif
    }
    float eps_base = clus->energy (EgammaHitsShowerDepth::m_samps[si][0]);
 
#ifndef NDEBUG
    ATH_MSG_DEBUG( "E accordion base --->>>> "  << eacc_base <<  "  Eps base " << eps_base  );
#endif
 
 
   // Add a protection against large longitudinal barycenter for clusters from
   // hadrons which may cause over-calibration. A energy dependent upper limit
   // on the long bary is introduced (from 20 at 0 to 23 at 1 TeV)    
 
   float depth_max = 20. + 3. * (eacc_base+eps_base) * 1e-6; // divided by TeV
#ifndef NDEBUG
   ATH_MSG_DEBUG( "Raw energy ---->> " << (eacc_base+eps_base)  );
   ATH_MSG_DEBUG( "Bary max for this event ---->> " << depth_max  );
#endif
 
   if ( shower_lbary > depth_max ) {
     shower_lbary = 15.;
     //shower_lbary = depth_max;
#ifndef NDEBUG
     ATH_MSG_DEBUG( " replace pathological depth by 15 X0 " );
#endif
   } 
    // -------------------------------------------------------------
    // Now calibrate the Accordion
    // -------------------------------------------------------------
 
    float acc_corr = correction[0][ibin][1] + correction[0][ibin][2] * shower_lbary + correction[0][ibin][3] * shower_lbary * shower_lbary ;
 
    float e_out_perc = 0;
    if (the_aeta < eta_start_crack) {
        e_out_perc = correction[1][ibin][1] + correction[1][ibin][2] * shower_lbary + correction[1][ibin][3] * shower_lbary * shower_lbary ;
    }
    else {
        e_out_perc = correction[1][ibin][1] + correction[1][ibin][2] * shower_lbary + correction[1][ibin][3] / shower_lbary ;
    }   
 
    float e_acc_reco=acc_corr*(eacc_base )*(1+(e_out_perc)/100);
 
    // -------------------------------------------------------------
    // Now estimate the longitudinal leakage
    // -------------------------------------------------------------
 
    float e_leak_perc = 0;
    e_leak_perc = correction[2][ibin][3] + correction[2][ibin][1] * shower_lbary + correction[2][ibin][2] * exp(shower_lbary);
 
    if (e_leak_perc < 0 ) e_leak_perc = 0.;
    if (e_leak_perc > 100.) e_leak_perc = 100.;
    float e_leak_reco = e_leak_perc * (e_acc_reco)/100;
 
    // -------------------------------------------------------------
    // Now estimate the energy lost in front of the calorimeter
    // -------------------------------------------------------------
 
    float raw_energy=e_acc_reco*.001;
    float e_front_reco = eps_base;
 
    if (raw_energy <= 1) {
        // Protect against log() going negative in pow();
        // also protect against sqrt() of a negative number and div-by-zero.
    }
    else if (the_aeta < 1.8) {
 
        if (the_aeta < eta_start_crack) {
            float WpsOff      = correction[3][ibin_frontCorr][1] + correction[3][ibin_frontCorr][2] * raw_energy + correction[3][ibin_frontCorr][3] * raw_energy * raw_energy;
            float WpsSlo      = correction[4][ibin_frontCorr][1]  * pow(log(raw_energy),correction[4][ibin_frontCorr][2]) + correction[4][ibin_frontCorr][3] *sqrt( raw_energy );
            e_front_reco=WpsOff + WpsSlo*(eps_base);
 
#ifndef NDEBUG
            ATH_MSG_DEBUG( " raw event  " << raw_energy  );
            ATH_MSG_DEBUG( " froffset coeff " << correction[3][ibin_frontCorr][1] << " " << correction[3][ibin_frontCorr][2] << " " << correction[3][ibin_frontCorr][3]   );
            ATH_MSG_DEBUG( " frslope coeff  " << correction[4][ibin_frontCorr][1] << " " << correction[4][ibin_frontCorr][2] << " " << correction[4][ibin_frontCorr][3]  );
            ATH_MSG_DEBUG( " WpsOff,WpsSlo " << WpsOff << " " << WpsSlo  );
            ATH_MSG_DEBUG( " eps_base, efront_reco " << eps_base << " " << e_front_reco  );
#endif
        }
        else{
 
 
            if (raw_energy<20.) raw_energy=20.;
 
            float WpsOff      = correction[3][ibin_frontCorr][1] + correction[3][ibin_frontCorr][2] * raw_energy + correction[3][ibin_frontCorr][3] * sqrt(raw_energy);
            float WpsSlo      = correction[4][ibin_frontCorr][1]  + correction[4][ibin_frontCorr][2] * log(raw_energy) + correction[4][ibin_frontCorr][3] * sqrt(raw_energy);
            float WpsSlo2     = correction[5][ibin_frontCorr][1]  + correction[5][ibin_frontCorr][2] * raw_energy - correction[5][ibin_frontCorr][3] / (raw_energy * raw_energy) ;
            e_front_reco=WpsOff + WpsSlo*(eps_base) + WpsSlo2*(eps_base)*(eps_base);
            if (e_front_reco<0.) e_front_reco= eps_base;
#ifndef NDEBUG
            ATH_MSG_DEBUG( " raw energy " << raw_energy  );
            ATH_MSG_DEBUG( "p1 " << correction[3][ibin_frontCorr][1] << " " << correction[3][ibin_frontCorr][2] << " " << correction[3][ibin_frontCorr][3] <<  " " << WpsOff  );
            ATH_MSG_DEBUG( "p2 " << correction[4][ibin_frontCorr][1] << " " << correction[4][ibin_frontCorr][2] << " " << correction[4][ibin_frontCorr][3] << " " << WpsSlo  );
            ATH_MSG_DEBUG( "p3 " << correction[5][ibin_frontCorr][1] << " " << correction[5][ibin_frontCorr][2] << " " << correction[5][ibin_frontCorr][3] << " " << WpsSlo2  );
            ATH_MSG_DEBUG( " WpsOff, WpsSlo, WpsSlo2 " << WpsOff << " " << WpsSlo << " " << WpsSlo2  );
            ATH_MSG_DEBUG( " eps_base, efront_reco " << eps_base << " " <<  e_front_reco  );
#endif
 
        }
 
    }
    else {
        float p1 =  correction[3][ibin_frontCorr][1] + correction[3][ibin_frontCorr][2] * raw_energy  + correction[3][ibin_frontCorr][3] * raw_energy  * raw_energy ;
        float p2 =  correction[4][ibin_frontCorr][1] + correction[4][ibin_frontCorr][2] * raw_energy  + correction[4][ibin_frontCorr][3] * raw_energy  * raw_energy ;
        float p3 =  correction[5][ibin_frontCorr][1] + correction[5][ibin_frontCorr][2] * raw_energy  + correction[5][ibin_frontCorr][3] * raw_energy  * raw_energy ;
 
#ifndef NDEBUG
        ATH_MSG_DEBUG( "p1 " << correction[3][ibin_frontCorr][1] << " " << correction[3][ibin_frontCorr][2] << " " << correction[3][ibin_frontCorr][3]  );
        ATH_MSG_DEBUG( "p2 " << correction[4][ibin_frontCorr][1] << " " << correction[4][ibin_frontCorr][2] << " " << correction[4][ibin_frontCorr][3]  );
        ATH_MSG_DEBUG( "p3 " << correction[5][ibin_frontCorr][1] << " " << correction[5][ibin_frontCorr][2] << " " << correction[5][ibin_frontCorr][3]  );
#endif
 
        e_front_reco= (p1 + p2 * shower_lbary + p3 * shower_lbary * shower_lbary);
        if (e_front_reco<0.) e_front_reco=eps_base;
    }
 
    // -------------------------------------------------------------
    // Now compute the total energy and finally update the cluster energies
    // -------------------------------------------------------------
 
    float e_calo_reco =e_front_reco + e_leak_reco + e_acc_reco;
 
#ifndef NDEBUG
    ATH_MSG_DEBUG( "CaloSwCalibrationHits::Final reco energy ---------------------- " << e_calo_reco  );
    ATH_MSG_DEBUG( "CaloSwCalibrationHits::Front ---------------------- " <<  e_front_reco  );
    ATH_MSG_DEBUG( "CaloSwCalibrationHits::Accordion ------------------ " <<  e_acc_reco  );
    ATH_MSG_DEBUG( "CaloSwCalibrationHits::out of cone ---------------- " <<  acc_corr*(eacc_base )*(e_out_perc)/100  );
    ATH_MSG_DEBUG( "CaloSwCalibrationHits::Leakage -------------------- " <<  e_leak_reco  );
#endif
 
 
 
    /*
       -------------------------------------------------------------
       Conventions discussed in Dec 2007 larg week to fill energies in the samplings:
       E = total energy including all corrections
       E0 = total energy in front of the accordion :
       presampler energy + energy lost in front + energy lost between PS and strips
       E1 = energy in the strips with no out of cone corrections
       E2 = energy in the middle with no out of cone corrections
       E3 = energy in the back with no out of cone correction + longitudinal leakage
 
       Please note that E != E0+E1+E2+E3
       -------------------------------------------------------------
     */
 
    // presampler
    clus->setEnergy (EgammaHitsShowerDepth::m_samps[si][0], e_front_reco );
 
    // strips and middle
    for (int sampling=1; sampling<=2; sampling++){
        clus->setEnergy (EgammaHitsShowerDepth::m_samps[si][sampling], clus->energy(EgammaHitsShowerDepth::m_samps[si][sampling]) * acc_corr );
    }
 
    // back
    clus->setEnergy (EgammaHitsShowerDepth::m_samps[si][3], clus->energy (EgammaHitsShowerDepth::m_samps[si][3]) * acc_corr + e_leak_reco);
 
#ifndef NDEBUG
    // total energy
    float e_temp = 0;
    for (int nl = 0 ; nl< 4 ; nl++) e_temp +=  clus->energy (EgammaHitsShowerDepth::m_samps[si][nl]);
    ATH_MSG_DEBUG( "----------  Sum of the sampling energy ---  >> " << e_temp << "  EcaloReco = " << e_calo_reco  );
    ATH_MSG_DEBUG( "CaloSwCalibHitsCalibration Energy after  correction --> " <<  clus->energy()  );
#endif
 
    clus->setEnergy(e_calo_reco);
 
    clus->setEt(clus->energy()/cosh(the_aeta));
 
}

Member Data Documentation

m_correction

Constant<CxxUtils::Array<3> > EgammaHitsCalibration::m_correction { this, "correction" }

private

Definition at line 47 of file EgammaHitsCalibration.h.

m_eta_end_crack

Constant<float> EgammaHitsCalibration::m_eta_end_crack { this, "eta_end_crack" }

private

Definition at line 50 of file EgammaHitsCalibration.h.

m_eta_start_crack

Constant<float> EgammaHitsCalibration::m_eta_start_crack { this, "eta_start_crack" }

private

Definition at line 49 of file EgammaHitsCalibration.h.

m_etamax

Constant<float> EgammaHitsCalibration::m_etamax { this, "etamax" }

private

Definition at line 51 of file EgammaHitsCalibration.h.

m_fix_v6_pathologies

Constant<bool> EgammaHitsCalibration::m_fix_v6_pathologies { this, "fix_v6_pathologies" }

private

Definition at line 54 of file EgammaHitsCalibration.h.

m_preserve_offset

Constant<bool> EgammaHitsCalibration::m_preserve_offset { this, "preserve_offset" }

private

Definition at line 53 of file EgammaHitsCalibration.h.

m_sampling_depth

Constant<CxxUtils::Array<2> > EgammaHitsCalibration::m_sampling_depth { this, "sampling_depth" }

private

Definition at line 48 of file EgammaHitsCalibration.h.

m_use_raw_eta

Constant<bool> EgammaHitsCalibration::m_use_raw_eta { this, "use_raw_eta" }

private

Definition at line 52 of file EgammaHitsCalibration.h.

---

The documentation for this class was generated from the following files:

• EgammaHitsCalibration.h
• EgammaHitsCalibration.cxx