University of Minnesota
School of Physics & Astronomy


Particle Reconstruction at High Pile-up

Kalafut,Sean_TA Award_2014 Update 2015.jpg
Sean Kalafut
Annie Bartels

Sean Kalafut is a graduate student working on the Compact Muon Solenoid (CMS) experiment at the Large Hadron Collider (LHC) at CERN under the direction of Professor Roger Rusack. Kalafut’s research focuses on particle reconstruction in high pile-up. Kalafut uses a physical analogy to describe high pile-up. “Imagine two people are each standing four feet in front of you, and the distance between the two people is four feet. Each person is holding a camera with a flash, and each person takes a photo of you with the flash turned on.

The time between the two photos is 1 second. You can easily tell where each flash came from, and the intensity of the two camera flashes relative to each other. Now, imagine trying to identify the brightest flash with 150 people and cameras packed into the same physical space, and the time between each camera flash is less than 1% of the time it takes a human to blink an eyelid. That is the challenge we will be facing in 10 years at the high energy and luminosity LHC.” In this analogy the intensity of one flash represents the energy of a particle produced by an interaction between two colliding protons. Kalafut is working on developing new software algorithms to identify the brightest camera flashes when the time between each flash is approximately 1/100th of a nanosecond. “In 10 years we will have about 150 ‘cameras’ going off in less than 1 nanosecond.” Kalafut and his advisor, Roger Rusack, are studying different ways to improve the detector so that they can improve the ability of CMS physicists to see those flashes.

Kalafut says that Rusack’s group has an idea for how to upgrade the CMS detector to better identify high energy particles produced by proton proton collisions in the future high pile up environment of the LHC. In the last 12 months Kalafut and other physicists at Minnesota and around the world have implemented this upgrade detector in simulation software and are studying how the hypothetical detector responds to different particle physics signals, like a high energy electron or photon that comes from the decay of a Higgs boson. Particle physics detectors at a basic level measure the energy of particles that travel through the detector, the time those particles hit the detector relative to the approximate time when the initial protons collided, and the path taken by particles as they travel through the detector. Kalafut says that those parameters are not enough to distinguish fundamentally different particle physics signatures, like an electron produced from the decay of a Z boson and a photon produced by the decay of a Higgs boson. The CMS detector uses a huge conglomerate of software called particle reconstruction code to accurately identify fundamental particles from different particle physics signatures, like electrons from Z boson decays and photons from Higgs boson decays, using the three raw outputs (energy, trajectory, and time) from the detector. The reconstruction software looks at all energy deposits measured by the detector in a 25 nanosecond window, and assigns a particle to each cluster of energy deposits based on the physical locations, times, and energy magnitudes of those deposits. Kalafut says that the reconstruction code and CMS detector has done a phenomenal job so far in advancing particle physics, but both need to be updated for CMS physicists to continue to advance the state of particle physics using data collected in the future LHC. “Our ability to sift raw detector output information into reconstructed particles is extraordinary. Now we want to make the physical detector and reconstruction code better because in the future we will have many more particles to sort.”