A New Observable for Identifying Dijet Resonances

The development of techniques for identifying hadronic signals from the overwhelming multi-jet backgrounds is an important part of the Large Hadron Collider (LHC) program. Of prime importance are resonances decaying into a pair of partons, such as the Higgs and $\rm W$/$\rm Z$ bosons, as well as hypothetical new particles. We present a simple observable to help discriminate a dijet resonance from background that is effective even when the decaying resonance is not strongly boosted. We find consistent performance of the observable over a variety of processes and degree of boosts, and show that it leads to a reduction of the background by a factor of $3-5$ relative to signal at the price of $10-20\%$ signal efficiency. This approach represents a significant increase in sensitivity for Standard Model (SM) measurements and searches for new physics that are dominated by systematic uncertainties, which is true of many analyses involving jets - particularly in the high-luminosity running of the LHC.
Comment: 6 pages, 6 figures. Version to appear in PRL