Speaker
Norman Graf
(SLAC)
Description
Maximizing the physics performance of detectors being designed for the International Linear Collide (ILC), while remaining sensitive to cost constraints, requires a powerful, efficient, and flexible simulation, reconstruction and analysis environment to study the capabilities of a large number of different detector designs. The preparation of letters of intent for the ILC involved the detailed study of dozens of detector options, layouts and readout technologies; the final physics benchmarking studies required the reconstruction and analysis of hundreds of millions of events.
We describe the Java-based software toolkit (org.lcsim) which was used for full event reconstruction and analysis. The components are fully modular and are available for tasks from digitization of tracking detector signals through to cluster finding, pattern recognition, track-fitting, calorimeter clustering, individual particle reconstruction, jet-finding, and analysis. The detector is defined by the same xml input files used for the detector
response simulation, ensuring the simulation and reconstruction geometries are always commensurate by construction. We discuss the architecture as well as the performance.
This toolkit has been used primarily for the characterization of the ILC Silicon Detector Concept. It therefore features packages which allow very sophisticated simulations of the response of silicon detectors to the passage of charged particles. The sensor classes allow very detailed descriptions of charge carrier movement in silicon detectors, e.g. one can list the collecting, absorbing and reflecting regions, properties of silicon (doping, mobility, diffusion length and so on), and electric and magnetic fields (including TCAD maps). After the charge carriers are generated and collected, the electronics simulation processes this charge into digital signals. We have defined an interface to specify how any such simulation should communicate with other parts of package. Since details of signal processing are very sensor specific, it is anticipated that any sensor option will have its own class handling such processing, but we have implemented a number of readout technologies of interest to the ILC detectors, such as CCDs and active pixel devices. Common to all the specific electronics simulation are the addition of electronics noise, propagation of the signal to readout, thresholding, and digitization of the
signal. The final output is then a list of electronics channels with their corresponding ADC counts, and optionally the time for the signal, replicating the readout from a real detector.
In addition to the ILC LOI studies, we describe the use of the org.lcsim software at CERN for CLIC studies, its application to the ATLAS tracker upgrade, dual readout crystal calorimeter detector R&D at Fermilab, and detector design for proposed “heavy photon” experiments at JLAB.
Author
Norman Graf
(SLAC)
