Physics Performance
CBM Progress Report 2013
Hit finding efficiency and J/ψ reconstruction efficiency of the MUCH detector K. Dey1 , S. Chattopadhyay2 , and B. Bhattacharjee1 1
Department of Physics, Gauhati University, Guwahati, India; 2 Variable Energy Cyclotron Centre, Kolkata, India
The proposed muon chamber (MUCH) of the FAIRCBM experiment aims to detect di-muons from the decay of both low-mass vector mesons and J/ψ [1]. The use of a thick absorber layer as in most of the existing muon detectors is not a good choice. As an alternative, a multilayer absorber system has been proposed. In this configuration, triplet-detector chambers are situated between the absorber layers placed over a considerable distance downstream of the STS detector. The sector-shaped gaseous chambers will be used as sensitive detectors for MUCH. The large-acceptance detectors of MUCH will give access to almost the entire forward phase space depending on the particle type. The angular acceptance of the detector spans from 5◦ to 25◦ , which corresponds to the pseudorapidity range 3.18 – 1.5. It is expected that the particle density varies largely over the detector acceptance. Such a variation of particle density might result in a variation of the hit and track finding efficiencies of the detector in different pseudo-rapidity regions. Hence a study on pseudorapidity dependent hit efficiency of the detector is of considerable significance. Hit efficiency is defined as the ratio of the number of reconstructed hits to the number of MC point, the latter representing a charged track crossing a sensitive detector element. The simulation was performed for Au+Au collisions at Elab = 10A GeV with the CBMROOT framework (DEC13) for the SIS-100 (3 absorber) configuration. Background particles and signal muons (from J/ψ decay) were generated using the UrQMD [2] and PLUTO [3] event generators, respectively. To account for a realistic detector geAu+Au @ 10 AGeV, central, 1st station
Table 1: J/ψ reconstruction efficiency for different cluster deconvolution algorithms and segmentation schemes
1
Hit-Efficiency
ometry, the readout modules are segmented in pads for obtaining the final detector response. In case of a sector geometry, projective pads of radially increasing size are implemented, the dimensions of the pads being determined by the angular separation in the transverse plane. Fired pads having a charge deposition greater than the threshold value are called digits, which are grouped into clusters using a suitable clustering algorithm. Based on particle multiplicity and associated cluster overlap, the clusters are broken into sub-clusters, which are treated as hits (advanced hit finder), or each cluster is treated as one hit (simple hit finder). The hits thus generated are used for the tracking. Fig. 1 shows the hit efficiency for different segmentation schemes as well as for different cluster deconvolution algorithms as a function of pseudo-rapidity. The hit efficiency decreases towards mid-rapidity where the particle density is maximum. Furthermore, the hit efficiency is found to be larger for a pad angle of 0.5◦ because of the decrease of multi-hit probability. For both segmentation schemes, i. e. 0.5◦ and 1◦ , the algorithm ‘local maxima finder’ (LMF) seems to perform better as far as hit efficiency is concerned. In addition to the results above, the J/ψ reconstruction efficiency was calculated for different segmentation schemes as well as for different cluster deconvolution algorithms. Table 1 shows it to be almost independent of the segmentation scheme considered. Thus, as far as cost is concerned, the pad angle of 1◦ seems preferrable. For both segmentation schemes, the LFM algorithm gives the maximum signal efficiency. We conclude that 1◦ pad angle and the LMF algorithm seem to be the optimum choice.
Pad angle
0.8
0.50
0.6
Local Maxima (LMF) LocalFinder Maxima Finder (LMF) 0.4
Deconvolution with charge threshold Deconvolution with Charge (DCT) Threshold (DCT)
10
One Hit Per One Cluster Hit Per(OHPC) Cluster (OHPC)
0.2
1.5
2
2.5
Pseudorapidity (η)
Algorithm OHPC DCT LMF OHPC DCT LMF
efficiency (embedded) 14.0 % 14.2 % 14.3 % 14.2 % 14.2 % 14.3 %
3
References Figure 1: Hit efficiency as function of pseudo-rapidity in the 1st station for different cluster deconvolution algorithms. The solid symbols correspond to a pad angle of 0.5◦ , the open symbols to a pad angle of 1◦ .
[1] P. Senger, J. Phys. G 28 (2002) 1869 [2] S. A. Bass et al., Prog. Part. Nucl. Phys. 41 (1998) 255 [3] http://www-hades.gsi.de/computing/pluto
118