Discrete Clustering in CMS HGCaI

The CMS Collaboration is proposing to build a high granularity calorimeter (HGCal) to replace the existing endcap calorimeter for the High Luminosity LHC. In this project I developed a discrete version of the clustering algorithm that is already in place (CLUE) in the CMS reconstruction framework (CMSSW). The algorithm exploits the topology of the detector to find neighbors instead of using a more traditional distance-based approach. I show, both quantitatively and qualitatively, that the discrete algorithm produces very similar clustering results as CLUE. Even though in its fastest implementation it is slower than CLUE, it can be made much faster with optimizations in the topology and hence, it could represent a suitable replacement for the distance-based options that are available now.