Because it accommodates such a wide range of requirements, the ALICE Central Trigger Processor (CTP) is, arguably, the most complex among the CTPs of the LHC experiments:
• The ALICE CTP generates three levels of hierarchical hardware triggers – L0, L1 and L2, before an event is accepted, transmitted to the Data Acquisition System (DAQ), and copied to the High Level Trigger (HLT) for further software assessments.
• At any time, the 24 sub-detectors of the ALICE experiment are dynamically partitioned into up to 6 independent clusters, with an additional, software triggered test cluster, configured on demand “on the fly”; the cluster configuration is fully arbitrary – clusters could be exclusive, but are more likely to overlap.
• Due to the complexity of ALICE events, a success or otherwise of the pattern recognition task is strongly dependant upon the level of event pile-up. The level is controlled by the Past-future Protection – a procedure that selects events with either no pile-up at all in a programmable time interval before and after the interaction, or with a number of pile-up interactions up to a programmable limit. The Past-future Protection operates independently for each cluster and is performed at all three trigger levels.
• Among the LHC experiments, the ALICE CTP generates the highest data traffic over the TTC system. Channel A is used for transmission of the timecritical L1 signal, in a way similar to most other experiments, but, for each trigger sequence that is confirmed at the L1 level, the following data are also transmitted via the TTC’s Channel B: the L1 Message - 5 16-bit words; the RoI Message (Region of Interest option) - 4 words; the L2a Message - 8 words. Concurrently, in a way similar to the other systems, Channel B is also used for transmission of the LHC Orbit and of the calibration Pre-pulse signal.