#include <EcalUncalibRecHitWorkerMultiFit.h>

Inheritance diagram for EcalUncalibRecHitWorkerMultiFit:

Public Member Functions
	EcalUncalibRecHitWorkerMultiFit (const edm::ParameterSet &, edm::ConsumesCollector &c)

bool	run (const edm::Event &evt, const EcalDigiCollection::const_iterator &digi, EcalUncalibratedRecHitCollection &result)

void	set (const edm::EventSetup &es)

virtual	~EcalUncalibRecHitWorkerMultiFit ()

Public Member Functions inherited from EcalUncalibRecHitWorkerBaseClass
	EcalUncalibRecHitWorkerBaseClass (const edm::ParameterSet &, edm::ConsumesCollector &c)

	EcalUncalibRecHitWorkerBaseClass (const edm::ParameterSet &)

virtual	~EcalUncalibRecHitWorkerBaseClass ()

Protected Member Functions
const SampleMatrix &	noisecor (bool barrel, int gain) const

double	timeCorrection (float ampli, const std::vector< float > &amplitudeBins, const std::vector< float > &shiftBins)

Protected Attributes
BXVector	activeBX

bool	ampErrorCalculation_

double	chi2ThreshEB_

double	chi2ThreshEE_

std::vector< double >	EBamplitudeFitParameters_

std::vector< double >	ebPulseShape_

double	EBtimeConstantTerm_

std::pair< double, double >	EBtimeFitLimits_

std::vector< double >	EBtimeFitParameters_

edm::ParameterSet	EcalPulseShapeParameters_

std::vector< double >	EEamplitudeFitParameters_

std::vector< double >	eePulseShape_

double	EEtimeConstantTerm_

std::pair< double, double >	EEtimeFitLimits_

std::vector< double >	EEtimeFitParameters_

FullSampleMatrix	fullpulsecovEB

FullSampleMatrix	fullpulsecovEE

FullSampleVector	fullpulseEB

FullSampleVector	fullpulseEE

double	gainRatios [3]

edm::ESHandle< EcalGainRatios >	gains

edm::ESHandle < EcalWeightXtalGroups >	grps

edm::ESHandle < EcalTimeCalibConstants >	itime

bool	kPoorRecoFlagEB_

bool	kPoorRecoFlagEE_

EcalUncalibRecHitLeadingEdgeAlgo < EBDataFrame >	leadingEdgeMethod_barrel_

EcalUncalibRecHitLeadingEdgeAlgo < EEDataFrame >	leadingEdgeMethod_endcap_

EcalUncalibRecHitMultiFitAlgo	multiFitMethod_

SampleMatrix	noisecorEBg1

SampleMatrix	noisecorEBg12

SampleMatrix	noisecorEBg6

SampleMatrix	noisecorEEg1

SampleMatrix	noisecorEEg12

SampleMatrix	noisecorEEg6

edm::ESHandle < EcalTimeOffsetConstant >	offtime

double	pedRMSVec [3]

edm::ESHandle< EcalPedestals >	peds

double	pedVec [3]

EcalUncalibRecHitRatioMethodAlgo < EBDataFrame >	ratioMethod_barrel_

EcalUncalibRecHitRatioMethodAlgo < EEDataFrame >	ratioMethod_endcap_

edm::ESHandle< EcalSampleMask >	sampleMaskHand_

std::string	timealgo_

edm::ESHandle < EcalTimeBiasCorrections >	timeCorrBias_

const EcalWeightSet::EcalWeightMatrix *	weights [2]

EcalUncalibRecHitTimeWeightsAlgo < EBDataFrame >	weightsMethod_barrel_

EcalUncalibRecHitTimeWeightsAlgo < EEDataFrame >	weightsMethod_endcap_

edm::ESHandle< EcalTBWeights >	wgts

Private Member Functions
void	fillInputs (const edm::ParameterSet &params)

Detailed Description

Definition at line 37 of file EcalUncalibRecHitWorkerMultiFit.h.

Constructor & Destructor Documentation

EcalUncalibRecHitWorkerMultiFit::EcalUncalibRecHitWorkerMultiFit	(	const edm::ParameterSet &	ps,
		edm::ConsumesCollector &	c
	)

Definition at line 18 of file EcalUncalibRecHitWorkerMultiFit.cc.

References activeBX, ampErrorCalculation_, chi2ThreshEB_, chi2ThreshEE_, EBamplitudeFitParameters_, ebPulseShape_, EBtimeConstantTerm_, EBtimeFitLimits_, EBtimeFitParameters_, EcalPulseShapeParameters_, EEamplitudeFitParameters_, eePulseShape_, EEtimeConstantTerm_, EEtimeFitLimits_, EEtimeFitParameters_, fillInputs(), edm::ParameterSet::getParameter(), kPoorRecoFlagEB_, kPoorRecoFlagEE_, BXVector< T >::resize(), AlCaHLTBitMon_QueryRunRegistry::string, and timealgo_.

                                                                                                                  :
   EcalUncalibRecHitWorkerBaseClass(ps,c),
   noisecorEBg12(SampleMatrix::Zero()), noisecorEEg12(SampleMatrix::Zero()),
   noisecorEBg6(SampleMatrix::Zero()), noisecorEEg6(SampleMatrix::Zero()),
   noisecorEBg1(SampleMatrix::Zero()), noisecorEEg1(SampleMatrix::Zero()),
   fullpulseEB(FullSampleVector::Zero()),fullpulseEE(FullSampleVector::Zero()),
   fullpulsecovEB(FullSampleMatrix::Zero()),fullpulsecovEE(FullSampleMatrix::Zero()) {
 
   // get the pulse shape, amplitude covariances and noise correlations
   EcalPulseShapeParameters_ = ps.getParameter<edm::ParameterSet>("EcalPulseShapeParameters");
   fillInputs(EcalPulseShapeParameters_);
 
   // get the BX for the pulses to be activated
   std::vector<int32_t> activeBXs = ps.getParameter< std::vector<int32_t> >("activeBXs");
   activeBX.resize(activeBXs.size());
   for (unsigned int ibx=0; ibx<activeBXs.size(); ++ibx) {
     activeBX.coeffRef(ibx) = activeBXs[ibx];
   }
 
   // uncertainty calculation (CPU intensive)
   ampErrorCalculation_ = ps.getParameter<bool>("ampErrorCalculation");
 
   // algorithm to be used for timing
   timealgo_ = ps.getParameter<std::string>("timealgo");
   
   // ratio method parameters
   EBtimeFitParameters_ = ps.getParameter<std::vector<double> >("EBtimeFitParameters"); 
   EEtimeFitParameters_ = ps.getParameter<std::vector<double> >("EEtimeFitParameters"); 
   EBamplitudeFitParameters_ = ps.getParameter<std::vector<double> >("EBamplitudeFitParameters");
   EEamplitudeFitParameters_ = ps.getParameter<std::vector<double> >("EEamplitudeFitParameters");
   EBtimeFitLimits_.first  = ps.getParameter<double>("EBtimeFitLimits_Lower");
   EBtimeFitLimits_.second = ps.getParameter<double>("EBtimeFitLimits_Upper");
   EEtimeFitLimits_.first  = ps.getParameter<double>("EEtimeFitLimits_Lower");
   EEtimeFitLimits_.second = ps.getParameter<double>("EEtimeFitLimits_Upper");
   EBtimeConstantTerm_=ps.getParameter<double>("EBtimeConstantTerm");
   EEtimeConstantTerm_=ps.getParameter<double>("EEtimeConstantTerm");
 
   // leading edge parameters
   ebPulseShape_ = ps.getParameter<std::vector<double> >("ebPulseShape");
   eePulseShape_ = ps.getParameter<std::vector<double> >("eePulseShape");
 
   // chi2 parameters for flags determination
   kPoorRecoFlagEB_ = ps.getParameter<bool>("kPoorRecoFlagEB");
   kPoorRecoFlagEE_ = ps.getParameter<bool>("kPoorRecoFlagEE");;
   chi2ThreshEB_=ps.getParameter<double>("chi2ThreshEB_");
   chi2ThreshEE_=ps.getParameter<double>("chi2ThreshEE_");
 
 }

virtual EcalUncalibRecHitWorkerMultiFit::~EcalUncalibRecHitWorkerMultiFit ( )

inlinevirtual

Definition at line 42 of file EcalUncalibRecHitWorkerMultiFit.h.

42 {};

Member Function Documentation

void EcalUncalibRecHitWorkerMultiFit::fillInputs ( const edm::ParameterSet & params )

private

Definition at line 437 of file EcalUncalibRecHitWorkerMultiFit.cc.

References funct::abs(), fullpulsecovEB, fullpulsecovEE, fullpulseEB, fullpulseEE, edm::ParameterSet::getParameter(), i, j, gen::k, noisecorEBg1, noisecorEBg12, noisecorEBg6, noisecorEEg1, noisecorEEg12, noisecorEEg6, and funct::pow().

Referenced by EcalUncalibRecHitWorkerMultiFit().

                                                                               {
 
   const std::vector<double> ebCorMatG12 = params.getParameter< std::vector<double> >("EBCorrNoiseMatrixG12");
   const std::vector<double> eeCorMatG12 = params.getParameter< std::vector<double> >("EECorrNoiseMatrixG12");
   const std::vector<double> ebCorMatG06 = params.getParameter< std::vector<double> >("EBCorrNoiseMatrixG06");
   const std::vector<double> eeCorMatG06 = params.getParameter< std::vector<double> >("EECorrNoiseMatrixG06");
   const std::vector<double> ebCorMatG01 = params.getParameter< std::vector<double> >("EBCorrNoiseMatrixG01");
   const std::vector<double> eeCorMatG01 = params.getParameter< std::vector<double> >("EECorrNoiseMatrixG01");
   
   int nnoise = ebCorMatG12.size();
 
   // fill correlation matrices: noise (HF (+) LF)
   for (int i=0; i<nnoise; ++i) {
     for (int j=0; j<nnoise; ++j) {
       int vidx = std::abs(j-i);
       noisecorEBg12(i,j) = ebCorMatG12[vidx];
       noisecorEEg12(i,j) = eeCorMatG12[vidx];
       noisecorEBg6(i,j)  = ebCorMatG06[vidx];
       noisecorEEg6(i,j)  = eeCorMatG06[vidx];
       noisecorEBg1(i,j)  = ebCorMatG01[vidx];
       noisecorEEg1(i,j)  = eeCorMatG01[vidx];        
     }
   }
   
   // fill shape: from simulation for samples 3-9, from alpha/beta shape for 10-14
   const std::vector<double> ebPulse = params.getParameter< std::vector<double> >("EBPulseShapeTemplate");
   const std::vector<double> eePulse = params.getParameter< std::vector<double> >("EEPulseShapeTemplate");
   int nShapeSamples = ebPulse.size();
   for (int i=0; i<nShapeSamples; ++i) {
     fullpulseEB(i+7) = ebPulse[i];
     fullpulseEE(i+7) = eePulse[i];
   }
 
   const std::vector<double> ebPulseCov = params.getParameter< std::vector<double> >("EBPulseShapeCovariance");
   const std::vector<double> eePulseCov = params.getParameter< std::vector<double> >("EEPulseShapeCovariance");
   for(int k=0; k<std::pow(nShapeSamples,2); ++k) {
     int i = k/nShapeSamples;
     int j = k%nShapeSamples;
     fullpulsecovEB(i+7,j+7) = ebPulseCov[k];
     fullpulsecovEE(i+7,j+7) = eePulseCov[k];
   }
 
 }

const SampleMatrix & EcalUncalibRecHitWorkerMultiFit::noisecor	(	bool	barrel,
		int	gain
	)		const

protected

Definition at line 409 of file EcalUncalibRecHitWorkerMultiFit.cc.

References noisecorEBg1, noisecorEBg12, noisecorEBg6, noisecorEEg1, noisecorEEg12, and noisecorEEg6.

Referenced by run().

                                                                                          {
   if (barrel) {
     if (gain==6) {
       return noisecorEBg6;
     }
     else if (gain==1) {
       return noisecorEBg1;
     }
     else {
       return noisecorEBg12;
     }    
   }
   else {
     if (gain==6) {
       return noisecorEEg6;
     }
     else if (gain==1) {
       return noisecorEEg1;
     }
     else {
       return noisecorEEg12;
     }        
   }
   
   return noisecorEBg12;
   
 }

bool EcalUncalibRecHitWorkerMultiFit::run	(	const edm::Event &	evt,
		const EcalDigiCollection::const_iterator &	digi,
		EcalUncalibratedRecHitCollection &	result
	)

virtual

Implements EcalUncalibRecHitWorkerBaseClass.

Definition at line 196 of file EcalUncalibRecHitWorkerMultiFit.cc.

 {
         DetId detid(itdg->id());
 
         const EcalSampleMask  *sampleMask_ = sampleMaskHand_.product();                
         
         // intelligence for recHit computation
         EcalUncalibratedRecHit uncalibRecHit;
         
         
         const EcalPedestals::Item * aped = 0;
         const EcalMGPAGainRatio * aGain = 0;
         const EcalXtalGroupId * gid = 0;
 
         if (detid.subdetId()==EcalEndcap) {
                 unsigned int hashedIndex = EEDetId(detid).hashedIndex();
                 aped  = &peds->endcap(hashedIndex);
                 aGain = &gains->endcap(hashedIndex);
                 gid   = &grps->endcap(hashedIndex);
         } else {
                 unsigned int hashedIndex = EBDetId(detid).hashedIndex();
                 aped  = &peds->barrel(hashedIndex);
                 aGain = &gains->barrel(hashedIndex);
                 gid   = &grps->barrel(hashedIndex);
         }
 
         pedVec[0] = aped->mean_x12;
         pedVec[1] = aped->mean_x6;
         pedVec[2] = aped->mean_x1;
         pedRMSVec[0] = aped->rms_x12;
         pedRMSVec[1] = aped->rms_x6;
         pedRMSVec[2] = aped->rms_x1;
         gainRatios[0] = 1.;
         gainRatios[1] = aGain->gain12Over6();
         gainRatios[2] = aGain->gain6Over1()*aGain->gain12Over6();
 
         
         // === amplitude computation ===
         int leadingSample = ((EcalDataFrame)(*itdg)).lastUnsaturatedSample();
 
         if ( leadingSample >= 0 ) { // saturation
                 if ( leadingSample != 4 ) {
                         // all samples different from the fifth are not reliable for the amplitude estimation
                         // put by default the energy at the saturation threshold and flag as saturated
                         float sratio = 1;
                         if ( detid.subdetId()==EcalBarrel) {
                                 sratio = ebPulseShape_[5] / ebPulseShape_[4];
                         } else {
                                 sratio = eePulseShape_[5] / eePulseShape_[4];
                         }
             uncalibRecHit = EcalUncalibratedRecHit( (*itdg).id(), 4095*12*sratio, 0, 0, 0);
                         uncalibRecHit.setFlagBit( EcalUncalibratedRecHit::kSaturated );
                 } else {
                         // float clockToNsConstant = 25.;
                         // reconstruct the rechit
                         if (detid.subdetId()==EcalEndcap) {
                                 leadingEdgeMethod_endcap_.setPulseShape( eePulseShape_ );
                                 // float mult = (float)eePulseShape_.size() / (float)(*itdg).size();
                                 // bin (or some analogous mapping) will be used instead of the leadingSample
                                 //int bin  = (int)(( (mult * leadingSample + mult/2) * clockToNsConstant + itimeconst ) / clockToNsConstant);
                                 // bin is not uset for the moment
                                 leadingEdgeMethod_endcap_.setLeadingEdgeSample( leadingSample );
                                 uncalibRecHit = leadingEdgeMethod_endcap_.makeRecHit(*itdg, pedVec, gainRatios, 0, 0);
                                 uncalibRecHit.setFlagBit( EcalUncalibratedRecHit::kLeadingEdgeRecovered );
                                 leadingEdgeMethod_endcap_.setLeadingEdgeSample( -1 );
                         } else {
                                 leadingEdgeMethod_barrel_.setPulseShape( ebPulseShape_ );
                                 // float mult = (float)ebPulseShape_.size() / (float)(*itdg).size();
                                 // bin (or some analogous mapping) will be used instead of the leadingSample
                                 //int bin  = (int)(( (mult * leadingSample + mult/2) * clockToNsConstant + itimeconst ) / clockToNsConstant);
                                 // bin is not uset for the moment
                                 leadingEdgeMethod_barrel_.setLeadingEdgeSample( leadingSample );
                                 uncalibRecHit = leadingEdgeMethod_barrel_.makeRecHit(*itdg, pedVec, gainRatios, 0, 0);
                                 uncalibRecHit.setFlagBit( EcalUncalibratedRecHit::kLeadingEdgeRecovered );
                                 leadingEdgeMethod_barrel_.setLeadingEdgeSample( -1 );
                         }
                 }
         // do not propagate the default chi2 = -1 value to the calib rechit (mapped to 64), set it to 0 when saturation
                 uncalibRecHit.setChi2(0);
         } else {
                 // multifit
                 bool barrel = detid.subdetId()==EcalBarrel;
                 int gain = 12;
                 if (((EcalDataFrame)(*itdg)).hasSwitchToGain6()) {
                   gain = 6;
                 }
                 if (((EcalDataFrame)(*itdg)).hasSwitchToGain1()) {
                   gain = 1;
                 }
                 const SampleMatrix &noisecormat = noisecor(barrel,gain);
                 const FullSampleVector &fullpulse = barrel ? fullpulseEB : fullpulseEE;
                 const FullSampleMatrix &fullpulsecov = barrel ? fullpulsecovEB : fullpulsecovEE;
                                 
                 uncalibRecHit = multiFitMethod_.makeRecHit(*itdg, aped, aGain, noisecormat,fullpulse,fullpulsecov,activeBX);
                 
                 // === time computation ===
                 if(timealgo_.compare("RatioMethod")==0) {
                   // ratio method
                   float const clockToNsConstant = 25.;
                   if (detid.subdetId()==EcalEndcap) {
                     ratioMethod_endcap_.init( *itdg, *sampleMask_, pedVec, pedRMSVec, gainRatios );
                     ratioMethod_endcap_.computeTime( EEtimeFitParameters_, EEtimeFitLimits_, EEamplitudeFitParameters_ );
                     ratioMethod_endcap_.computeAmplitude( EEamplitudeFitParameters_);
                     EcalUncalibRecHitRatioMethodAlgo<EEDataFrame>::CalculatedRecHit crh = ratioMethod_endcap_.getCalculatedRecHit();
                     double theTimeCorrectionEE = timeCorrection(uncalibRecHit.amplitude(),
                                                                 timeCorrBias_->EETimeCorrAmplitudeBins, timeCorrBias_->EETimeCorrShiftBins);
                     
                     uncalibRecHit.setJitter( crh.timeMax - 5 + theTimeCorrectionEE);
                     uncalibRecHit.setJitterError( std::sqrt(pow(crh.timeError,2) + std::pow(EEtimeConstantTerm_,2)/std::pow(clockToNsConstant,2)) );
                     
                   } else {
                     ratioMethod_barrel_.init( *itdg, *sampleMask_, pedVec, pedRMSVec, gainRatios );
                     ratioMethod_barrel_.fixMGPAslew(*itdg);
                     ratioMethod_barrel_.computeTime( EBtimeFitParameters_, EBtimeFitLimits_, EBamplitudeFitParameters_ );
                     ratioMethod_barrel_.computeAmplitude( EBamplitudeFitParameters_);
                     EcalUncalibRecHitRatioMethodAlgo<EBDataFrame>::CalculatedRecHit crh = ratioMethod_barrel_.getCalculatedRecHit();
                     
                     double theTimeCorrectionEB = timeCorrection(uncalibRecHit.amplitude(),
                                                                 timeCorrBias_->EBTimeCorrAmplitudeBins, timeCorrBias_->EBTimeCorrShiftBins);
                     
                     uncalibRecHit.setJitter( crh.timeMax - 5 + theTimeCorrectionEB);
                     
                     uncalibRecHit.setJitterError( std::sqrt(std::pow(crh.timeError,2) + std::pow(EBtimeConstantTerm_,2)/std::pow(clockToNsConstant,2)) );
                   }
                 } else if(timealgo_.compare("WeightsMethod")==0) {
                   //  weights method on the PU subtracted pulse shape
                   std::vector<double> amplitudes;
                   for(unsigned int ibx=0; ibx<activeBX.size(); ++ibx) amplitudes.push_back(uncalibRecHit.outOfTimeAmplitude(ibx));
                   
                   EcalTBWeights::EcalTDCId tdcid(1);
                   EcalTBWeights::EcalTBWeightMap const & wgtsMap = wgts->getMap();
                   EcalTBWeights::EcalTBWeightMap::const_iterator wit;
                   wit = wgtsMap.find( std::make_pair(*gid,tdcid) );
                   if( wit == wgtsMap.end() ) {
                     edm::LogError("EcalUncalibRecHitError") << "No weights found for EcalGroupId: " 
                                                             << gid->id() << " and  EcalTDCId: " << tdcid
                                                             << "\n  skipping digi with id: " << detid.rawId();
                     
                     return false;
                   }
                   const EcalWeightSet& wset = wit->second; // this is the EcalWeightSet
                   
                   const EcalWeightSet::EcalWeightMatrix& mat1 = wset.getWeightsBeforeGainSwitch();
                   const EcalWeightSet::EcalWeightMatrix& mat2 = wset.getWeightsAfterGainSwitch();
                   
                   weights[0] = &mat1;
                   weights[1] = &mat2;
                   
                   double timerh;
                   if (detid.subdetId()==EcalEndcap) { 
                     timerh = weightsMethod_endcap_.time( *itdg, amplitudes, aped, aGain, fullpulse, weights);
                   } else {
                     timerh = weightsMethod_barrel_.time( *itdg, amplitudes, aped, aGain, fullpulse, weights);
                   }
                   uncalibRecHit.setJitter( timerh );
                   uncalibRecHit.setJitterError( 0. ); // not computed with weights
                 }  else if(timealgo_.compare("None")==0) {
                   uncalibRecHit.setJitter( 0. );
                   uncalibRecHit.setJitterError( 0. );                  
                 } else {
                   edm::LogError("EcalUncalibRecHitError") << "No time estimation algorithm called " 
                                                           << timealgo_
                                                           << "\n  setting jitter to 0. and jitter uncertainty to 10000. ";
                   
                   uncalibRecHit.setJitter( 0. );
                   uncalibRecHit.setJitterError( 10000. );
                 }
         }
 
     // set flags if gain switch has occurred
     if( ((EcalDataFrame)(*itdg)).hasSwitchToGain6()  ) uncalibRecHit.setFlagBit( EcalUncalibratedRecHit::kHasSwitchToGain6 );
     if( ((EcalDataFrame)(*itdg)).hasSwitchToGain1()  ) uncalibRecHit.setFlagBit( EcalUncalibratedRecHit::kHasSwitchToGain1 );
 
         // set quality flags based on chi2 of the the fit
         /*
         if(detid.subdetId()==EcalEndcap) { 
           if(kPoorRecoFlagEE_ && uncalibRecHit.chi2()>chi2ThreshEE_) {
           bool samplesok = true;
           for (int sample =0; sample < EcalDataFrame::MAXSAMPLES; ++sample) {
             if (!sampleMask_->useSampleEE(sample)) {
               samplesok = false;
               break;
             }
           }
           if (samplesok) uncalibRecHit.setFlagBit(EcalUncalibratedRecHit::kPoorReco);
           }
         } else {
           if(kPoorRecoFlagEB_ && uncalibRecHit.chi2()>chi2ThreshEB_) {
           bool samplesok = true;
           for (int sample =0; sample < EcalDataFrame::MAXSAMPLES; ++sample) {
             if (!sampleMask_->useSampleEB(sample)) {
               samplesok = false;
               break;
             }
           }
           if (samplesok) uncalibRecHit.setFlagBit(EcalUncalibratedRecHit::kPoorReco);
           }
         }
         */
 
         // put the recHit in the collection
         if (detid.subdetId()==EcalEndcap) {
                 result.push_back( uncalibRecHit );
         } else {
                 result.push_back( uncalibRecHit );
         }
 
         return true;
 }

void EcalUncalibRecHitWorkerMultiFit::set ( const edm::EventSetup & es )

virtual

Implements EcalUncalibRecHitWorkerBaseClass.

Definition at line 103 of file EcalUncalibRecHitWorkerMultiFit.cc.

References ampErrorCalculation_, EcalUncalibRecHitMultiFitAlgo::disableErrorCalculation(), gains, edm::EventSetup::get(), grps, itime, multiFitMethod_, offtime, peds, sampleMaskHand_, timeCorrBias_, and wgts.

Referenced by betterConfigParser.BetterConfigParser::getGeneral().

 {
 
         // common setup
         es.get<EcalGainRatiosRcd>().get(gains);
         es.get<EcalPedestalsRcd>().get(peds);
 
         // for the multifit method
         if(!ampErrorCalculation_) multiFitMethod_.disableErrorCalculation();
 
         // weights parameters for the time
         es.get<EcalWeightXtalGroupsRcd>().get(grps);
         es.get<EcalTBWeightsRcd>().get(wgts);
 
     // which of the samples need be used
     es.get<EcalSampleMaskRcd>().get(sampleMaskHand_);
 
         // for the ratio method
 
         // for the leading edge method
         es.get<EcalTimeCalibConstantsRcd>().get(itime);
         es.get<EcalTimeOffsetConstantRcd>().get(offtime);
 
         // for the time correction methods
         es.get<EcalTimeBiasCorrectionsRcd>().get(timeCorrBias_);
 }

double EcalUncalibRecHitWorkerMultiFit::timeCorrection	(	float	ampli,
		const std::vector< float > &	amplitudeBins,
		const std::vector< float > &	shiftBins
	)

protected

Amplitude-dependent time corrections; EE and EB have separate corrections: EXtimeCorrAmplitudes (ADC) and EXtimeCorrShifts (ns) need to have the same number of elements Bins must be ordered in amplitude. First-last bins take care of under-overflows.

The algorithm is the same for EE and EB, only the correction vectors are different.

Returns: Jitter (in clock cycles) which will be added to UncalibRechit.setJitter(), 0 if no correction is applied.

Definition at line 139 of file EcalUncalibRecHitWorkerMultiFit.cc.

References newFWLiteAna::bin.

Referenced by run().

                                        {
 
   // computed initially in ns. Than turned in the BX's, as
   // EcalUncalibratedRecHit need be.
   double theCorrection = 0;
 
   // sanity check for arrays
   if (amplitudeBins.size() == 0) {
     edm::LogError("EcalRecHitError")
         << "timeCorrAmplitudeBins is empty, forcing no time bias corrections.";
 
     return 0;
   }
 
   if (amplitudeBins.size() != shiftBins.size()) {
     edm::LogError("EcalRecHitError")
         << "Size of timeCorrAmplitudeBins different from "
            "timeCorrShiftBins. Forcing no time bias corrections. ";
 
     return 0;
   }
 
   int myBin = -1;
   for (int bin = 0; bin < (int) amplitudeBins.size(); bin++) {
     if (ampli > amplitudeBins.at(bin)) {
       myBin = bin;
     } else {
       break;
     }
   }
 
   if (myBin == -1) {
     theCorrection = shiftBins.at(0);
   } else if (myBin == ((int)(amplitudeBins.size() - 1))) {
     theCorrection = shiftBins.at(myBin);
   } else if (-1 < myBin && myBin < ((int) amplitudeBins.size() - 1)) {
     // interpolate linearly between two assingned points
     theCorrection = (shiftBins.at(myBin + 1) - shiftBins.at(myBin));
     theCorrection *= (((double) ampli) - amplitudeBins.at(myBin)) /
                      (amplitudeBins.at(myBin + 1) - amplitudeBins.at(myBin));
     theCorrection += shiftBins.at(myBin);
   } else {
     edm::LogError("EcalRecHitError")
         << "Assigning time correction impossible. Setting it to 0 ";
     theCorrection = 0.;
   }
 
   // convert ns into clocks
   return theCorrection / 25.;
 }