Development of a monitoring system for the muon trigger of the ATLAS detector

The Muon-to-Central Trigger Processor Interface (MUCTPI) is a fundamental component
of the Trigger and Data Acquisition (TDAQ) system of the ATLAS (A Toroidal LHC Apparatus) detector. ATLAS is a general-purpose particle physics experiment at the Large Hadron Collider (LHC) at CERN. The LHC collides protons at a rate of 40 MHz, but only a small fraction of the events will contain interesting characteristics for physics and should be recorded. For this reason the detector is equipped with the TDAQ system, which decides, in real time, whether to read out or discard the measurements of each observed interaction for offline analyses. The decision is based on the presence of defined
signals in the detector, e.g. a muon exceeding a specific transverse momentum threshold. The ATLAS Trigger is based on a two-stage selection: The first is hardware-based with custom electronics, and the second is software-based with algorithms running on a large farm of off-the-shelf computers. The first stage is the Level-1 (L1) trigger and the second is the High-Level-Trigger (HLT). The L1 trigger processes information coming from the calorimeters, muon trigger chambers and forward detectors. The muon trigger detecting systems are based on Resistive Plate Chambers (RPC) in the central region, and Thin-Gap Chambers (TGC), small-strip Thin-Gap Chambers (sTGC), and Micromegas (MM) in the forward region of the detector. With the information coming from the muon system, the MUCTPI calculates muon candidate multiplicities for programmable transverse momentum thresholds. The MUCTPI then sends the muon multiplicity information to the Central Trigger Processor (CTP) and trigger objects to the L1 Topological processor (L1Topo). For every bunch-crossing, the CTP combines the information coming from the MUCTPI, the calorimeters, and the forward detectors to elaborate the L1 trigger decision. The output rate of the CTP is limited to 100 kHz.
During the Long Shutdown that preceded the LHC Run-3 begun in July 2022, the ATLAS detector was subject to a number of upgrades in order to improve its capabilities. These upgrades are known as of Phase-I upgrades. With higher rates and pile-up levels (additional pp interactions happening in the same bunch-crossing) with respect to its design values, the ATLAS TDAQ system needed to adapt to the detector’s changes, with a number of planned upgrades. Among the TDAQ upgrades the MUCTPI is receiving a completely new design in terms of electronics and software. The 18 9U VME modules
which composed the MUCTPI of the previous Run-1 and Run-2 have been replaced by a single ATCA blade featuring three state-of-the-art FPGAs and a System-on-chip (SoC). Two FPGAs are responsible for receiving the trigger information from the muon trigger sectors, doing overlap removal for duplicate muon candidates, calculating the transverse momentum threshold multiplicities, and sending the trigger objects to L1Topo. The third FPGA calculates trigger multiplicities and sends them to the CTP and trigger data to the DAQ system. The SoC is used for control, configuration, and monitoring of the hardware and operations of the MUCTPI. The high-level services provided by the SoC are based on the usage of a collection of common libraries non-specific to the detector, known as TDAQ online software. On the MUCTPI, a cross-compiled version of the TDAQ software allows the running of run applications in the ATLAS run control framework. The TDAQ software is used by the applications to realize the configuration, control and monitoring tasks on the MUCTPI.       The main aim of my thesis was to develop MUCTPI run control monitoring applications, which compose the MuctpiMonitoring package of the ATLAS TDAQ software. There are several objects being monitored: the rate of muon candidates coming from muon trigger sectors, the rate of multiplicities sent to the CTP, and the busy sources in the MUCTPI. The rate applications are also implemented for per-bunch rates. I also developed a separate application running on a host PC in the same ATLAS counting cavern where the MUCTPI is installed. This application publishes histograms based on per-bunch data. Developing these applications required the implementation of low-level software which provides
an application programming interface (API) to the relevant FPGA firmware registers. Additionally I designed and implemented schemas and infrastructure software for the integration of the monitoring applications in the ATLAS run control system. Finally I developed a number of web monitoring pages in order to make the monitoring operations
in the ATLAS Control Room possible for human users. Their features are presented in this work.
The software I wrote during the thesis work has been exercised during laboratory tests and the commissioning phase. Currently it is executed in the ATLAS Control Room for the LHC Run-3.