1 / 12

Algorithm design for MAPS clustering

Algorithm design for MAPS clustering. Bradley Hopkinson / Birmingham. Introduction: Using LDC01 detector model in Mokka-06-01 MAPS Mokka geometry (Yoshi Mikami) Using MARLIN (v09-04) with MarlinReco (v00-02). Compare default 1cm analogue Si with 50 m m MAPS (15 m m epi layer)

freya-patrick
Télécharger la présentation

Algorithm design for MAPS clustering

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Algorithm design for MAPS clustering Bradley Hopkinson / Birmingham Introduction: • Using LDC01 detector model in Mokka-06-01 • MAPS Mokka geometry (Yoshi Mikami) • Using MARLIN (v09-04) with MarlinReco (v00-02). • Compare default 1cm analogue Si with 50mm MAPS (15mm epi layer) • Event samples – several k events for each of: • single electron events c/o Yoshi Mikami (default geometry & MAPS geometry) – should be just 1 large cluster • Z  qq~ (uds quarks only) events at 91 GeV – should be ~20 small clusters CALICE MAPS / RAL 06-Sep-2006 1

  2. Efficiency and Purity – definition used Efficiency = number of reconstructed hits in a cluster (right and wrong) number of possible hits in the 'True Cluster' Purity = correct hits in a cluster total hits in the cluster Efficiency of an event = number of reconstructed clusters number of possible 'True Clusters' 2

  3. Example Z qq event

  4. Example Z qq event

  5. Example events (analogue Si)

  6. Clustering merges where not natural? Example events (MAPS geom.)

  7. Example events (MAPS geom.)

  8. Z  qq events Analogue Si (1cm pitch) Efficiency of each cluster ~107%

  9. MAPS (50mm pitch) Efficiency of each cluster ~106%

  10. Analogue Si (1cm pitch) Efficiency of event ~62%

  11. MAPS (50mm pitch) Efficiency of event ~61%

  12. Conclusion and Outlook First study of real clustering for MAPS simulated events, realistic sensor geometry Trackwise clustering algorithm performs well on single electron events, in both the analogue Si and MAPS cases In MAPS case, clustering does not use energy information On more complex events, e.g.jets in the Z -> qq events, the trackwise clustering algorithm has comparable performance to the default analogue Si case, cluster reconstruction efficiency of ~ 61%. Can be tuned further. The processor I have written can be adapted to work with different clustering algorithms. Have not successfully used it on Mark Thomson's PandoraPFA as yet. Would be interesting to see how a newer algorithm like this would fare with MAPS data.

More Related