Addition of new algorithms to TOFEstimator processor #88

dudarboh · 2021-02-08T14:51:27Z

BEGINRELEASENOTES

TOFEstimators processor: adding new algorithms to calculate the ToF, track length and momentum.
- New output parameters:
  - TOFClosest --- based on the closest ECAL hit
  - TOFFastest --- based on the fastest ECAL hit
  - TOFCylFit --- extrapolated time from the ECAL hits within a cylinder of a shower core
  - TOFClosestFit --- extrapolated time from the ECAL hits closest to the linear continuation of the track inside the ECAL
  - FlightLength --- the helix track length based on the track parameters from the TrackState at the ECAL
  - MomAtCal --- the momentum based on the track parameters from the TrackState at the ECAL
- New steering parameters:
  - ProcessorVersion --- changes output between the idr version (before this patch) and the dev (after this patch). Default: idr
  - CylRadius --- the radius within which to select hits for the TOFCylFit method. Default: 5 mm

ENDRELEASENOTES

Important notes

Currently new algorithms work only with simple cases of PFO, meaning: only one cluster and only one track associated with PFO. Other PFOs are ignored.

Flight length calculation are done using the assumption of less than 2*pi change in the phi angle. Meaning that the flight length gives the wrong results for the tracks with more than one helix curls.

Thus, the new algorithms are still optimized to work in the barrel part only. Performance of the TOF estimators in the endcap is currently unknown.

What's new

We added new TOF estimators that show a better performance in terms of less bias in the measurement of pions, kaons and protons masses.

Steering parameter

New steering parameter ProcessorVersion has been added with the possible values: "idr" (default) and "dev".
The previous behavior is set by default for the backwards compatibility.

Output

In the new version the output parameters are modified to:
{"TOFClosest", "TOFFastest", "TOFCylFit", "TOFClosestFit", "FlightLength", "MomAtCalo"}

TOF* --- are TimeOfFlight estimated by different algorithms (see bellow).
FlightLength --- is a track length of a particle from IP (assuming (0, 0, 0)) to ECAL surface
MomAtCalo --- Track momentum at ECAL surface.

Using these parameters the mass of a particle can be calculated.

TOF estimators

The four estimators are available in the new version:

TOFClosest --- time of the ECAL hit closest to the extrapolated track position at the ECAL surface with correction for distance between impact point and center of ECAL cell assuming speed of light propagation.
TOFFastest --- time of the ECAL hit with the smallest time with correction for distance between impact point and center of ECAL cell assuming speed of light propagation.
TOFCylFit --- time estimated with a linear fit of the ECAL hits times in the 5mm cylinder with the main axis parallel to track direction at the ECAL surface vs distance to the impact point of each hit. Then TOF is the bias parameter of a linear fit.
TOFClosestFit --- time estimated with a linear fit of the ECAL hits times which are closest to the track line estimated at the ECAL surface (one closest hit per layer) vs distance to the impact point of each hit. Then TOF is the bias parameter of a linear fit.

rete · 2021-02-08T14:56:37Z

@dudarboh Can you:

fill the release notes between BEGINRELEASENOTE S and ENDRELEASENOTE S
rebase on the ilcsoft master. The CI has been updated

?

…s are resolved.

dudarboh · 2021-02-08T15:52:49Z

Hi, @rete.

Done. Please check.

rete · 2021-02-08T15:58:48Z

@dudarboh When you open the pull request there is message:


BEGINRELEASENOTE S
- Thank you for writing the text to appear in the release notes. It will show up
  exactly as it appears between the two bold lines
- ...

ENDRELEASENOTE S