MDC Analysis Scheme

The MDC analysis running fom raw data on hld level to itted tracks with momentum determination and Vertex finding can be generalized as the following:
I. Depending on the input configuration the flow of analysis is different. Reading from hld file involves the unpackers, whereas reading from a root file can basicly start from each data level contained in the root input.
II. For simulation unpacking from raw data on hld level is not performed, instead of this the GEANT input is read from root file input. Those input files are the output of a HGEANT simulation. The HMdcDigitizer will read GEANT information about hits from HGeantMdc and convert them into drift time measurements folding in the response of the MDCs. The output of this procedure will be HMdcCal1Sim, looking like a real measurement but keeping additional infomations from simulation.


In Data Analysis after unpacking the HMdcRaw level is filled. This data level contains the measured data in a hardware oriented way, which means the measured information is stored by hardware addresses sector, module, motherboard and tdc channel. For the further analysis this coordinate system is not very usefull, therefore the information has to be mapped to an other coordinate system during calibration. Second the raw time measurements from the tdcs have to be converted to calibrated drift times, which will be later used for track reconstruction. This tasks are performed by the HMdcCalibrater1 reconstructor.

HMdcCalibrater1 performs the following tasks:
I. Calibration of the drift times from tdc channels on HMdcRaw level to ns on HMdcCal1 level.
II. Removing jitter from Trigger by substracting the start time taken from HStartHit.
III. Cutting noise by applying time cuts in time1, time2 and time2-time1 taken from HMdcTimeCuts
IV. Fill statistics of cutted cells in HMdcCutStats.
V. Map from sector,module,motherboard,tdc channel to sector, module,layer,cell.

HMdcCalibrater2 fills the HMdcCal2 data level, which contains distance from sense wire in mm obtained from caculations using HMdcCal2Par for converting drift times as input of the SANTIAGO tracking procedure. This transformation uses a fixed impacted angle of the traversing particle track, because the impact angle at this point of analysis is unknown. As the transformation from drift time to distance from wire in general is depending on the impact angle of the particle track this is a rather crude approximation. The DUBNA tracking procedure starts from HMdcCal1 and transforms during the tracking when this parameters are know from drift time to distance to wire to obtain the points in space.


The DUBNA tracking procedure is structured in different steps:
I. Finding of candidates of hits (cluster search). This step is performed by HMdcTrackfinder and HMdc34ClFinder. The input for the candidate search is taken from HMdcCal1 data level. The wires which belong to one cluster are found by doing projections of all wires for one module or one Segment depending on the cluster search mode. The output of the procedure will be stored in HMdcClus, which is by default non persistent.
II. Fitting of clusters. Input to the procedure is a list of all wires belonging to one cluster. The fitting algorithm converts the measured drift times from HMdcCal1 into points in space using the xt-relation stored in HMdcCal2ParSim (distance->time) and HMdcCal2Par (time->distance). The minimization procedure of the fitter uses tukey weights and will decide which wires belong to a hit and which are just noise or contribution from other hits. The output of the procedure will be HMdcSeg (information about hits of 2 modules belonging to oone segment (inner/outer)), HMdcHit (infomation about one hit in one module), HMdcClusInf (information about one hit (clustersize/number of contributing wires before fit), used in Close Pair Rejection algorithm (CPR))and HMdcTrkCand (information about combination of matched inner/out segment of Mdcs), which will be input to further track reconstruction involving momentum reconstruction and matching with hits in the META detector and the RICH. Some helper data levels like HMdcClusFit and HMdcWireFit, which contains detailed information about the fit ,can be filled optional.


The reconstruction of the vertex event by event is performed by HVertexFinder. This reconstructor loops over HMdcSeg category for the inner segments and tries to find a common vertex for all segments of the same event. To suppress tracks not coming from target the minimazation procedure uses tukey weights.


The energyloss of an particle on its way through the MDCs is calulated by HMdcDeDxMaker. The procedure starts from HMdcTrkCand data level, retrieving the indices of the match inner/outer segments inside HMdcSeg category. For each wire inside the segment the Time Above Threshold is calucaled (TAB = time2-time1) normalized and averaged for the segment.The result of the procedure are written to the mdc dEdx related data members of HMdcTrkCand.


-- JochenMarkert - 24 Feb 2004
Topic attachments
I Attachment Action Size Date Who Comment
mdc_ana_input_scheme.gif.drawdraw mdc_ana_input_scheme.gif.draw manage 7 K 2004-02-23 - 09:44 JochenMarkert TWikiDraw draw file
mdc_ana_input_scheme.gif.gifgif mdc_ana_input_scheme.gif.gif manage 13 K 2004-02-23 - 09:44 JochenMarkert TWikiDraw GIF file
mdc_ana_mdcdedx.gif.drawdraw mdc_ana_mdcdedx.gif.draw manage 3 K 2004-02-24 - 13:43 JochenMarkert TWikiDraw draw file
mdc_ana_mdcdedx.gif.gifgif mdc_ana_mdcdedx.gif.gif manage 7 K 2004-02-24 - 13:43 JochenMarkert TWikiDraw GIF file
mdc_ana_scheme.gif.drawdraw mdc_ana_scheme.gif.draw manage 4 K 2004-02-23 - 10:05 JochenMarkert TWikiDraw draw file
mdc_ana_scheme.gif.gifgif mdc_ana_scheme.gif.gif manage 8 K 2004-02-23 - 10:05 JochenMarkert TWikiDraw GIF file
mdc_ana_tracking.gif.drawdraw mdc_ana_tracking.gif.draw manage 11 K 2004-02-24 - 13:59 JochenMarkert TWikiDraw draw file
mdc_ana_tracking.gif.gifgif mdc_ana_tracking.gif.gif manage 23 K 2004-02-24 - 13:59 JochenMarkert TWikiDraw GIF file
mdc_ana_vertex.gif.drawdraw mdc_ana_vertex.gif.draw manage 1 K 2004-02-24 - 13:53 JochenMarkert TWikiDraw draw file
mdc_ana_vertex.gif.gifgif mdc_ana_vertex.gif.gif manage 3 K 2004-02-24 - 13:53 JochenMarkert TWikiDraw GIF file
resolution_vs_p_combined.pdfpdf resolution_vs_p_combined.pdf manage 8 K 2004-02-23 - 17:35 JochenMarkert  
tdc_modes_signal.pdfpdf tdc_modes_signal.pdf manage 140 K 2004-02-23 - 16:09 JochenMarkert  
Topic revision: r3 - 2004-03-01, JochenMarkert
Copyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki Send feedback | Imprint | Privacy Policy (in German)