Skip to content

Add HLT2 single muon line with isolation; finalise preliminary QEE HLT2 single muon programme

Ross John Hunter requested to merge rjhunter-try-SUMCONE into qee_upgrade

FYI @wbarter @ncooke @olupton @yqiu @cvazquez @hyin @dzuliani

As pointed out in the context of !560 (merged), the rate of the SingleHighPtMuonLine is too high, suggesting it is letting quite a lot of background in. The next step was then to apply some isolation requirements on the muon. This has been achieved using the SUMCONE LoKi functor, which takes particle flow input particles, and put into the new Hlt2SingleHighPtIsoMuonLine, which is the same as above but with an arbitrary cut of 10 GeV on the PT within a cone of \Delta R = 0.5.

Within this MR we then finalise 3 lines for the QEE single-muon programme:

  • a line with no isolation and no prescale with p_{T}^{\mu} > 15 GeV.
  • a line with no prescale, a isolation cut of p_{T}^{\textrm{cone}} < 10 GeV and the lower threshold fo p_{T}^{\mu} > 12.5 GeV.
  • a line with a 5% prescale and no isolation with the even lower p_{T}^{\mu} > 10 GeV.

Efficiencies on 10k W+jet events:

------------------------------------------------------------------------------------
Hlt2 integrated efficiencies for the lines with denominator: HighTruthPtMuon
------------------------------------------------------------------------------------
Line:	 Hlt2SingleHighPtMuonIsoLineDecision	 Efficiency:	 0.733 +/- 0.005
Line:	 Hlt2SingleHighPtMuonLineDecision   	 Efficiency:	 0.905 +/- 0.003
------------------------------------------------------------------------------------
Finished printing Hlt2 integrated efficiencies for denominator: HighTruthPtMuon
------------------------------------------------------------------------------------

(the prescaled line is omitted as the efficiency is very low, although a line with a p_{T}^{\mu} > 10 GeV threshold and no prescale was found to have a 90.9% efficiency on this denominator.)

Screenshot_2020-08-20_at_15.13.06

The rates of all 3 lines, as measured on 300k min bias:

Hlt2 rates: 
----------------------------------------------------------------------------------------------------
Line:	 Hlt2SingleHighPtMuonIsoLineDecision       	 Incl: 0.0566666666667 +/- 0.01 kHz, Excl: 0.00666666666667 +/- 0.00 kHz
Line:	 Hlt2SingleHighPtMuonLineDecision          	 Incl: 0.0833333333333 +/- 0.01 kHz, Excl: 0.0633333333333 +/- 0.01 kHz
Line:	 Hlt2SingleHighPtMuonPrescaleLineDecision  	 Incl: 0.0133333333333 +/- 0.00 kHz, Excl: 0.01 +/- 0.00 kHz
----------------------------------------------------------------------------------------------------
Hlt2 Total:                                            Rate:	 1 +/- 0.1 kHz
----------------------------------------------------------------------------------------------------

TODO:

  • Try some different cuts, get their rate and signal efficiency. Settle on a value.
  • Move the make_particleflow_for_<jets/isolation> function live in a shared QEE builder script, rather than duplicate the code from the jet line.

Based on !560 (merged) so requires that to be merged to master and qee_upgrade before this can be merged. The diff will look horrible until that is done too. Thanks @olupton for the functors help.

Edited by Ross John Hunter

Merge request reports