Topology in the future ATLAS Level-1 Trigger

The ATLAS experiment examines the decays of high energetic particles produced in proton-proton collisions at the Large Hadron Collider (LHC). Resuming operation at the beginning of 2015 for Run 2 the LHC will work with an increased center-of-mass energy of $13-14 \mathrm{TeV}$, which will result in an unprecedented luminosity. The first trigger level (Level-1) of the ATLAS trigger system, based on custom-made electronic modules, needs to be upgraded to control the increased trigger rate, while keeping good efficiency for interesting physics events. In Run 1 the Level-1 trigger decisions were mostly based only on multiplicities of trigger objects (such as electrons / photons, jets, hadrons, muons above energy/momentum thresholds) observed in the calorimeters and the muon spectrometer. As part of the Level-1 upgrade a new trigger module, the topological processor (L1Topo) will be included into the trigger chain. It is intended to evaluate the spatial correlations of trigger objects and to perform more complex kinematic calculations such as an invariant mass. This requires to have as much as possible of the trigger object information in one place for maximum flexibility in the evaluation of the topological information. Therefore the topological processor consists of four FPGAs with the highest available input bandwidth of up to $1 \mathrm{TBit/s}$. The processing of the incoming data takes place in real-time and in less than the budgeted $200 \mathrm{ns}$. In this contribution the design of the L1Topo modules is discussed and test results of the prototypes of the L1Topo modules are presented.