MDC3: HBT Physics Working Group

---

1. BNL-Attendance - By my count, HBT expects about 8-10 folks on location, including resident BNL-ers. I think we all have adequate office/terminal access, but I have put out a ping on that.
2. Goals - In MDC3, we will be performing several mock analyses that hopefully will give us an idea of what more needs to be done before data comes. These include:
   • "standard" pi-pi (like and unlike sign)
     • pull out multiplicity dependence of Radii
     • how much momentum resolution hurts
     • track-splitting (Testing of Randy's algorithm) (see note below)
     • can we reconstruct big sources?
     • is our Coulomb-correction algorithm sufficient?
   • event-by-event HBT
     • At what multiplicity does it break down?
     • tag tools
     • tests of various fitting techniques, including global fit of piplus-piplus, piminus-piminus, piplus-piminus
   • 3-particle correlation analysis - test code which being implemented into StHbt - what signals are mocked up by 2-particle correlations (which is all what is in the events) + detector effects - how many events can be analyzed in lifetime of the universe?
   • charged kaon and proton-proton HBT - largely test of statistics/PID
   • proton-Lambda and K0-K0 HBT - can we extract these interesting signal with exiting cuts and expected statistics?
   • phi, K*, and other resonances - better understanding of efficiency/acceptance for various strictness cuts - generation of 1D and 2D spectra with StHbt
3. There are also issues of general infrastructure:
   • We would like to run in conjuntion with other PWGs (EbyE/flow and strangeness) in a PWG-Maker chain, to generate uDSTs. Question is how well does this work in a large scale?
   • For some analyses, we would like to generate HBT uDSTs (either ASCII or binary) and analyze them with the standalone StHbt. Is this practical/desirable?
   • How well does the track-split removal algorithm of Randy Wells clean up the signal for HBT? Is it neccessary/sufficient? (see note below)
   • In general, what types of pairwise cuts must we implement? (Including issues of merging/splitting/parent-daughter incest in track-V0 and V0-V0 correlations.)
   • Test of tag submission and retrieval (event-by-event HBT)
   • Do we have sufficient information in our StHbtEvent structures to perform adequate cuts and do analysis?
4. Experiences - Asking around, I would say that most folks know about the STAR disks, but not everybody is familiar with LSF (including me).

Note related to infrastructure: We want to try to use track topology to attack potential tracksplit problems that come up. This is why I proposed the bitpattern hitmap in StTrack. Currently this is not being filled by global (and so not copied into StTrack). We would like DST production of HBT data to proceed after this information is filled. Helen says she will do it on global end tonite, and I hope you can make the next step shortly after so it gets into production.

Text by Mike Lisa, Feb 17, 2000