A combination of analysis techniques for efficient track reconstruction of high multiplicity events in silicon detectors

A novel combination of established data analysis techniques for reconstructing all charged-particle tracks in high energy collisions is proposed. It uses all information available in a collision event while keeping competing choices open as long as possible. Suitable track candidates are selected by transforming measured hits to a binned, three- or four-dimensional, track parameter space. It is accomplished by the use of templates taking advantage of the translational and rotational symmetries of the detectors. Track candidates and their corresponding hits, the nodes, form a usually highly connected network, a bipartite graph, where we allow for multiple hit to track assignments, edges. The graph is cut into very many minigraphs by removing a few of its vulnerable components, edged and nodes. Finally the hits are distributed among the track candidates by exploring a deterministic decision tree. A depth-limited search is performed maximising the number of hits on tracks, and minimising the sum of track-fit $\chi^2$. Simplified models of LHC silicon trackers, as well as the relevant physics processes, are employed to study the performance (efficiency, purity, timing) of the proposed method in the case of single or many simultaneous proton-proton collisions (high pileup), and for single heavy-ion collisions at the highest available energies.