---

CSCWireElectronicsSim.cc

Go to the documentation of this file.

#include "FWCore/MessageLogger/interface/MessageLogger.h"
#include "SimMuon/CSCDigitizer/src/CSCWireElectronicsSim.h"
#include "SimMuon/CSCDigitizer/src/CSCAnalogSignal.h"
#include "DataFormats/CSCDigi/interface/CSCWireDigi.h"
#include "Geometry/CSCGeometry/interface/CSCLayer.h"
#include "Geometry/CSCGeometry/interface/CSCLayerGeometry.h"
#include "Geometry/CSCGeometry/interface/CSCChamberSpecs.h"
#include "CLHEP/Units/PhysicalConstants.h"
#include <iostream>


CSCWireElectronicsSim::CSCWireElectronicsSim(const edm::ParameterSet & p) 
 : CSCBaseElectronicsSim(p),
   theFraction(0.5),
   theWireNoise(0.0),
   theWireThreshold(0.),
   theTimingCalibrationError(p.getParameter<double>("wireTimingError"))
{
  fillAmpResponse();
}


void CSCWireElectronicsSim::initParameters() {
  theLayerGeometry = theLayer->geometry();
  nElements = theLayerGeometry->numberOfWireGroups();
  theWireNoise = theSpecs->wireNoise(theShapingTime)
                * e_SI * pow(10.0,15);
  theWireThreshold = theWireNoise * 8;
}


int CSCWireElectronicsSim::readoutElement(int element) const {
  return theLayerGeometry->wireGroup(element);
}

void CSCWireElectronicsSim::fillDigis(CSCWireDigiCollection & digis) {

  if(theSignalMap.empty()) {
    return;
  }

  // Loop over analog signals, run the fractional discriminator on each one,
  // and save the DIGI in the layer.
  for(CSCSignalMap::iterator mapI = theSignalMap.begin(),
      lastSignal = theSignalMap.end();
      mapI != lastSignal; ++mapI) 
  {
    int wireGroup = (*mapI).first;
    const CSCAnalogSignal & signal = (*mapI).second;
    LogTrace("CSCWireElectronicsSim") << "CSCWireElectronicsSim: dump of wire signal follows... " 
       <<  signal;
    int signalSize = signal.getSize();

    int timeWord = 0; // and this will remain if too early or late (<bx-6 or >bx+9)

    // the way we handle noise in this chamber is by randomly varying
    // the threshold
    float threshold = theWireThreshold;
    if (doNoise_) {
       threshold += theRandGaussQ->fire() * theWireNoise;
    }
    for(int ibin = 0; ibin < signalSize; ++ibin)
    {
      if(signal.getBinValue(ibin) > threshold) 
      {
        // jackpot.  Now define this signal as everything up until
        // the signal goes below zero.
        int lastbin = signalSize;
        int i;
        for(i = ibin; i < signalSize; ++i) {
          if(signal.getBinValue(i) < 0.) {
            lastbin = i;
            break;
          }
        }

        float qMax = 0.0;
        // in this loop, find the max charge and the 'fifth' electron arrival
        for ( i = ibin; i < lastbin; ++i)
        {
          float next_charge = signal.getBinValue(i);
          if(next_charge > qMax) {
            qMax = next_charge;
          }
        }
     
        int bin_firing_FD = 0;
        for ( i = ibin; i < lastbin; ++i)
        {
          if( bin_firing_FD == 0 && signal.getBinValue(i) >= qMax * theFraction )
          {
             bin_firing_FD = i;
          }
        } 
        float tofOffset = timeOfFlightCalibration(wireGroup);
        int chamberType = theSpecs->chamberType();

        // Note that CSCAnalogSignal::superimpose does not reset theTimeOffset to the earliest
        // of the two signal's time offsets. If it did then we could handle signals from any
        // time range e.g. form pileup events many bx's from the signal bx (bx=0).
        // But then we would be wastefully storing signals over times which we can never
        // see in the real detector, because only hits within a few bx's of bx=0 are read out.
        // Instead, the working time range for wire hits is always started from
        // theSignalStartTime, set as a parameter in the config file.
        // On the other hand, if any of the overlapped CSCAnalogSignals happens to have
        // a timeOffset earlier than theSignalStartTime (which is currently set to -100 ns)
        // then we're in trouble. For pileup events this would mean events from collisions
        // earlier than 4 bx before the signal bx.

        float fdTime = theSignalStartTime + theSamplingTime*bin_firing_FD;
        if(doNoise_) {
          fdTime += theTimingCalibrationError * theRandGaussQ->fire();
        }

        float bxFloat = (fdTime - tofOffset- theBunchTimingOffsets[chamberType]) / theBunchSpacing
                           + theOffsetOfBxZero;
        if(bxFloat >= 0 && bxFloat < 16)
        {
           timeWord |= (1 << static_cast<int>(bxFloat) );
        }

        // Wire digi as of Oct-2006 adapted to real data: time word has 16 bits with set bit
        // flagging appropriate bunch crossing, and bx 0 corresponding to the 7th bit, 'bit 6':
  
        //      1st bit set (bit 0) <-> bx -6
        //      2nd              1  <-> bx -5
        //      ...           ...       ....
        //      7th              6  <-> bx  0
        //      8th              7  <-> bx +1
        //      ...           ...       ....
        //     16th             15  <-> bx +9

        // skip over all the time bins used for this digi
        ibin = lastbin;
      } // if over threshold
    } // loop over time bins in signal

    // Only create a wire digi if there is a wire hit within [-6 bx, +9 bx]
    if(timeWord != 0)
    {
      CSCWireDigi newDigi(wireGroup, timeWord);
      LogTrace("CSCWireElectronicsSim") << newDigi;
      digis.insertDigi(layerId(), newDigi);
      addLinks(channelIndex(wireGroup));
    }
  } // loop over wire signals   
}


float CSCWireElectronicsSim::calculateAmpResponse(float t) const {
  static const float fC_by_ns = 1000000;
  static const float resistor = 20000;
  static const float amplifier_pole               = 1/7.5;
  static const float fastest_chamber_exp_risetime = 10.;
  static const float p0=amplifier_pole;
  static const float p1=1/fastest_chamber_exp_risetime;

  static const float dp = p0 - p1;

  // ENABLE DISC:

  static const float norm = -12 * resistor * p1 * pow(p0/dp, 4) / fC_by_ns;

  float enable_disc_volts = norm*(  exp(-p0*t) *(1          +
                                                 t*dp       +
                                                 pow(t*dp,2)/2 +
                                                 pow(t*dp,3)/6  )
                                    - exp(-p1*t) );
  static const float collectionFraction = 0.12;
  static const float igain = 1./0.005; // volts per fC
  return enable_disc_volts * igain * collectionFraction;
}                                                                               


float CSCWireElectronicsSim::signalDelay(int element, float pos) const {
  // readout is on right edge of chamber, signal speed is c
  // zero calibrated to chamber center
  // pos is assumed to be in wire coordinates, not local
  float distance = -1. * pos; // in cm
  float speed = c_light / cm;
  float delay = distance / speed;
  return delay;
}

float CSCWireElectronicsSim::timeOfFlightCalibration(int wireGroup) const {
  // calibration is done for groups of 8 wire groups, facetiously
  // called wireGroupGroups
  int middleWireGroup = wireGroup - wireGroup%8 + 4;
  int numberOfWireGroups = theLayerGeometry->numberOfWireGroups();
  if(middleWireGroup > numberOfWireGroups) 
     middleWireGroup = numberOfWireGroups;

  GlobalPoint centerOfGroupGroup = theLayer->centerOfWireGroup(middleWireGroup);
  float averageDist = centerOfGroupGroup.mag();
  float averageTOF  = averageDist * cm / c_light; // Units of c_light: mm/ns

  LogTrace("CSCWireElectronicsSim") << "CSCWireElectronicsSim: TofCalib  wg = " << wireGroup << 
       " mid wg = " << middleWireGroup << 
       " av dist = " << averageDist << 
      " av tof = " << averageTOF;
  
  return averageTOF;
}
 

---