version 1.4 – Nov 2000 – Paul Balm ( p.balm@nikhef.nl )
This information is supposed to eventually mimic the information in the class L3MuonPhysicsResults, that will be written out to the L3Chunk.
The information we will store, is a selection of properties of each muon in the event as they are measured by the L3 system. This means
We will store phi, eta and pT of the muon. The most accurate measurement possible, occurs when we have the muon matched to a central track, and we can rely on the measurement of the variables from the central tracker.
To store these values with the accuracy with which we can measure them, we will need 32 bits. Additionally, we will need 1 bit for the charge of the muon.
For the track in the muon system, we will store the number of wirehits on the track for both the A segment and the BC segment (track segments in B- and C-layers combined). The number of hits for an A-segment runs from 0 to 4, and for a BC-segment it’s maximally 6. To store these numbers, we’ll reserve 6 bits.
The quality of the match between the track in the muon system and the central track will be stored as the distance between the central track propagated to the A-layer, and the local muon track in the A-layer. The units are cm, and the value is truncated at 31 cm. We will need 5 bits to store that value.
For the timing information from the muon scintillators, we would like to store the velocity of the muon in the range from beta is 0.5 to 1.5. We assume we can measure time of passage in a scintillator with an accuracy of 1 ns. Roughly, the distance from the interaction point to the farthest scintillator is 8m. The time difference for a particle with beta is 0.5 and beta is 1.5, is roughly 27 ns. With an accuracy of 1 ns, we will need 5 bits, plus one bit to store the sign of the velocity. Particle having a velocity lower than beta is 0.5 or higher than 1.5 will be considered out-of-time.
Also calorimeter information will be stored. First, we will store the total energy deposited in the hadronic calorimeter cells associated with the track. The reason for including the hadronic layers only, is that the energy deposit in the electromagnetic layers is relatively small and more susceptible to large fluctuations (sometimes due to other particles). The deposited energy range starts from 0 and is truncated at 12.7 GeV with an accuracy of the order of 0.1 GeV. Hence we need 7 bits to store the energy information. Secondly, the number of hadronic layers traversed by the particle with energy > 0, as a fraction of the total number of hadronic layers traversed will be stored. This number is usually referred to as ‘hfrac’. Depending on the trajectory of the particle, the number of hadronic layers a particle traverses, is 3, 4 or 5, but for large |z| vertices, it can also be 6 or 7. To store all possibilities for the numerator of hfrac, we’ll need 3 bits, and to store all possibilities for the denominator, we will need 3 bits as well. The total number of bits for hfrac will then be 6.
The detailed proposal accounts for the following numbers of bits for different pieces of information:
This accumulates to a total of 63 bits. Additionally, we would like to reserve 1 bit for information we have not accounted for, bringing the information for Level 3 Muon to a total of 64 bits.