Using Mock Calibration for Reconstruction

Introduction

Falaise currently only provides a mocking model for digitization and calibration of data after the Geant4 part of the simulation has completed. This is due to the developing nature of the physical electronic system at the time of writing, and thus the digitization/calibration is currently provided as an optional "preprocessing" step within flreconstruct.

This allows a basic model to be used, and does not prevent further validation and implementation of new digitization and calibration models.

Using the Mock Calibration Models

To use the mocking model as the first step of the pipeline, two modules need to be run; snemo::processing::mock_tracker_s2c_module and snemo::processing::mock_calorimeter_s2c_module. The following example script shows how these may be run

# - Configuration
#@description Simple pipeline using a chain
#@key_label   "name"
#@meta_label  "type"

# - Module load section
#[name="flreconstruct.plugins" type=""]
#plugins : string[1] = "MyModule"

# Must define "pipeline" as this is the module flreconstruct will use
[name="pipeline" type="dpp::chain_module"]
modules : string[2] = "MockCalibration" "Dump"

[name="MockCalibration" type="dpp::chain_module"]
modules : string[2] = "CalibrateTracker" "CalibrateCalorimeters"

[name="CalibrateTracker" type="snemo::processing::mock_tracker_s2c_module"]
Geo_label : string = "geometry"

[name="CalibrateCalorimeters" type="snemo::processing::mock_calorimeter_s2c_module"]
Geo_label : string = "geometry"
hit_categories  : string[3]  = "calo" "xcalo" "gveto"

[name="Dump" type="dpp::dump_module"]

This simply chains together the two modules, with a simple dump module as the last stage to demonstrate that the "CD" (C alibrated D ata) data bank is added. It is this bank that should be used for any reconstruction or analysis task, and it should be read as an instance of snemo::datamodel::calibrated_data. The documentation of that object should be consulted for information on accessing Geiger and Calorimeter data.

You can replace the dump module in the above example with a chain of your own reconstruction and analysis modules.

Future Development

The mocking modules implement both digitization and calibration steps. In future, the digitization step will be refactored into flsimulate as the intent is for the simulation to output data in an identical format to the physical experiment(s). The calibration step will be refactored into a core module of flreconstruct.

Neither of these refactorings affect you if you wish to develop and propose new digitization or calibration models. You should however take care to not overwrite the "official" data banks for simulated and calibrated data ("SD" and "CD" respectively). This is allow easy comparison and validation of any new model with the "official" one, as both "official" and "new" data banks will be available to downstream modules.

Specific Topics for Development of Digitization

You should implement custom digitization as a processing module in flreconstruct, using the mctools::simulated_dataSD" data bank" output from flsimulate as the input. You should create any needed custom types to represent the digitized data and write these as a new bank in the datatools::things instance representing the event.

TODO : Define bank name for digital data.

Specific Topics for Development of Calibration

You should implement custom calibration as a processing module in flreconstruct. Your calibration model must output data in the form of an instance of snemo::datamodel::calibrated_data. You would write this into the datatools::things event instance as a new data bank named, say, "CD_NewModel" to avoid clashes with the official "CD" bank. Downstream modules can access both banks for comparison and validation.