Reconstruction

1. Signals -> Hits
   The raw output from the detector is read in and converted into hits i.e. small track segments within active areas of the detector. Channel based calibrations are applied. Hits typically carry timing, charge/pulse height and position which may be inferred from the position of the detector element possibly in conjunction with timing.

2. Hits -> vertices, tracks, showers
   At this stage 3D objects are built from hits. This may involve matching 2D hits. Pattern recognition and fitting algorithms are applied to groups of hits in an attempt to identify the global structures of the event which represent the event interactions. Calibrations that require full spatial position are applied at this stage.

3. vertices, tracks, showers -> Events
   Further pattern recognition algorithms are now applied to the event topology, trying to separate events, either statistically or individually, into different physics hypotheses.

1. Calibration
   Throughout the operation of the detector, calibrations are taken of all the detector characteristics that could be time dependent. Normally this will include both electronic and detector response functions in order to convert signal height into energy. Calibrations of timing and even positions may also be necessary,

2. Database
   A database is needed to record all the time dependent calibrations so that event reconstruction has a standard source of detector constants to be applied to events as they are processed. It also holds other event related data such as run summaries which can be used as the starting point of data processing book-keeping system.

Reconstruction Software:-

• Is normally written in C++ or Java although there are still some Fortran codes.
• Are often built on to frameworks such as ROOT
• Require Software Development
• Require a Database