Use of fast tracking electronics systems for real-time decision making

Domain topic

Particle Physics

Supervisors

Motivation

Dark matter is one of the biggest mysteries of our universe. At the Large Hadron Collider (LHC), dark matter particles could be produced by colliding protons. Since dark matter processes are very rare, the proton-proton collision rate at the LHC needs to be very high to ensure that we are able to produce those elusive particles. To cope with this, the ATLAS experiment is designing and building new tracking systems to reconstruct the trajectories of the charged particles from the collisions. These projects concern two different elements related to tracking in particle physics:

  • the ATLAS Fast Tracker (FTK), a unique hardware processor made of several custom electronic boards based on Field Programmable Gate Arrays (FPGA) that will reconstruct track trajectories in real-time and guide the decision on which collisions are interesting and should be recorded.

  • the high-granularity, radiation-hard ATLAS Inner Tracker (ITk) that will substitute the current tracking system after the next upgrade of the LHC. Upcoming particle physics data analyses concerning dark matter particles will heavily rely on these novel tracking systems. Having a precise tracking system and fast decisions based on tracking information are both essential points to distinguish dark matter signals and background from known particles.

The volume of data available to both research and industry is ever-increasing. This acceleration in data collection is not always matched by a comparable acceleration in data storage and data utilization: in the upcoming LHC data taking, one needs to decide whether an event can be saved to disk for further analysis or discarded, all within milliseconds. Searches for rare processes with high-rate backgrounds such as dark matter-related particles suffer in this paradigm where data is first collected and then analysed. For this reason, real-time analysis of data and fast decision making are crucial for the understanding of the data, as well as crucial for reducing the time-to-insight in other fields of physics and industry. The FTK is the first system at the LHC that allows information from particle tracks to be broadly available in a milliseconds timescale, enabling background rejection from other known particles as well as real-time reconstruction of detector signals that are ready for physics analysis.

Project Description

The ATLAS FTK is based on a database of track patterns used to perform real-time decisions. In this project, the student will implement a fast version of the FTK software for simulation, which will utilise a simplified version of the database. Such an improved fast simulation will simplify the analysis of trigger-level data collected with the help of the FTK. The FTK will be operational from early 2021, and LHC data processes with FTK information will be used to search for dark matter using real-time reconstruction and analysis.

Methodological keywords

Electronics, simulation start-to-end & optimization, feedback loops based on close-to-real-time data analysis, industrial instrumentation