BGV MC Digitization Progress since BGV #27 Plamen Hopchev CERN BE-BI-BL BGV meeting #29 12 Mar 2014 1 / 8
Summary Event Model Classes The needed classes are ready: SciFiChannelID SciFiLiteCluster SciFiCluster Exact same bit assignment and functionality as the Velo versions Raw Bank Conversion The needed algorithms are ready: Decode/Encode SciFiClusters from/to Raw Bank Exact same behavior as the Velo versions Detector Element The basic functionality is ready The LHCb SciFi tracker code used as prototype 2 / 8
Summary (2) Digitization Getting close to a first version Initially, make a direct translation of the MCHit position to ChannelID, set fixed cluster size (2), no energy sharing between SiPM cells, etc. First raw files could be available soon, allowing to Get expected event size Start developing pattern recognition algorithms 3 / 8
Packages The suggestion is to put all BGV SciFi packages under the same hat: SciFi Allows to have an easier overview of the package organization Might be beneficial for the BGV code portability Might need to consider other hats too in case our package number becomes large Hat / Name Copy from Purpose CURRENT Event / DigiEvent, SciFi / SciFiEvent Event Model classes Kernel / LHCbKernel SciFi / SciFiDAQ Velo / VeloDAQ Raw bank decoding and encoding Detector element (c++ representation of SciFi / SciFiDet Det / FTDet the xml geometry description) STILL TO COME MCHit energy deposit –> strip ADCs SciFi / SciFiSim (LHCb SciFi) (effects of the signal acquisition) Tracking Vetexing HLT ... 4 / 8
SciFi modules Panoramix views 5 / 8
Next steps Fix the TO-DOs in the currently implemented packages Add REC ↔ MC links Implement signal acquisition effects: light sharing, attenuation, SiPM QE, cross-talk Event size, pattern reco, track fit, vertexing 6 / 8
Additional Slides
«particle physics» software chain Interaction conditions: Material description: Detector description: Detector description: proton energy, target nucleus (un-)sensitive volumes, position-channel info, channel-position info, type, etc. material types, etc. signal response, noise, alignment constants digitization (possibly, add misalignments) Alignment Monitoring histograms, Simulation trends Generator Reconstruction (GEANT) Digitization (HIJING) data decoding, track finding, generates material transform MC hits into raw track fitting, vertex finding, generates the primary interactions, multiple data vertex fitting interaction scattering, energy deposits, etc MC particle «decay tree»: MC Hit bank: Raw data: High level objects: store particle type, deposited energy, ADCs and channel IDs tracks, vertices include include 4momentum vector, origin entry/exit points vtx, decay vtx, daughter links, time-of-flight, ... SAME AS REAL DATA (apart from MC truth) NB: simplification compared to LHCb: - probably do not need spillover treatment - pile-up treatment is not a priority 3 BGV meeting 29-May-2013 CERN Massimiliano Ferro-Luzzi 8 / 8
Recommend
More recommend
