SoftKiller

The SoftKiller algorithm is a pileup mitigation technique for jet reconstruction in high-energy physics. It is designed to suppress the effect pileup by removing low-$p_T$ particles before jet clustering.

Overview

SoftKiller divides the rapidity-azimuth ($y\text{-}\phi$) plane into a regular grid. In each grid cell, it determines a $p_T$ threshold such that at least one particle per cell survives. All particles with $p_T$ below this threshold are removed. This approach efficiently reduces pileup contamination while preserving the hard event structure.

The method is described in Cacciari, Salam, Soyez, Eur. Phys. J. C75 (2015) 59.

Interface

The main interface for SoftKiller is the softkiller function, which is used as follows:

using JetReconstruction

sk = SoftKiller(rapmax=5.0, grid_size=0.4)
filtered_particles, pt_threshold = softkiller(sk, particles)

particles is a collection of PseudoJet objects (see Input Particle Types).
rapmax sets the maximum rapidity considered.
grid_size sets the size of each grid cell in rapidity and azimuthal angle.

The function returns the filtered list of particles and the computed $p_T$ threshold.

Example

A typical workflow using SoftKiller is:

using JetReconstruction

# Read event particles (see examples/softkiller/softkiller_plots.jl)
particles = read_final_state_particles("event.hepmc3", T=PseudoJet)[1]

# Set up SoftKiller
sk = SoftKiller(rapmax=5.0, grid_size=0.4)

# Apply SoftKiller
filtered_particles, pt_threshold = softkiller(sk, particles)

# Cluster jets as usual
cs = jet_reconstruct(filtered_particles; algorithm=JetAlgorithm.AntiKt, R=0.4)
jets = inclusive_jets(cs, ptmin=25.0)

See [examples/softkiller/softkiller_plots.jl] and [examples/softkiller/softkiller_runtime.jl] for more specific working examples.

Parameters

Parameter	Description	Default
`rapmax`	Maximum rapidity for grid	5.0
`grid_size`	Size of grid cell in rapidity-azimuth	0.4

Integration

SoftKiller is typically used as a preprocessing step before jet clustering. It is compatible with all jet algorithms provided by this package and is especially useful in high pileup environments.

References

M. Cacciari, G. P. Salam, G. Soyez, Eur. Phys. J. C75 (2015) 59
SoftKiller in FastJet and (https://phab.hepforge.org/source/fastjetsvn/browse/contrib/contribs/SoftKiller/tags/1.0.0/)

Authors

The SoftKiller implementation in this package follows the approach described in the original paper and is integrated with the native Julia jet reconstruction algorithms.