SMuK 2023 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

HK: Fachverband Physik der Hadronen und Kerne

HK 2: Instrumentation I

HK 2.5: Vortrag

Montag, 20. März 2023, 17:45–18:00, SCH/A251

A Feature Extraction Ansatz for the PANDA Forward-Endcap EMC — •Celina Frenkel — HISKP, Uni Bonn

The forward endcap of the electromagnetic calorimeter of the PANDA experiment consists of 3856 lead tungstate crystals. These are either readout by VPTTs in the high rate regime (θ 13^∘) or by 2 APDs per crystal at larger angles. The signals are then digitized using sampling ADCs.

A feature extraction algorithm implemented on the FPGAs on the SADC boards is used to extract energy and time information from the SADC signals online in real-time. The expected particle rate in forward direction of the experiment reaches ∼500 kHz such that pile-up is a relevant aspect. A signal deconvolution can be used to shorten the pulse in time and reduce the probability of pile-up.

The central topic of this talk is the investigation of an online feature extraction ansatz making use of the Pulse Shape Deconvolution (PSD). With this ansatz it is possible to even remove unwanted properties of the signal shape and provide a clean signal for the following extraction of the signal’s features using a peakfinder algorithm.

Finally, the performance of the new PSD ansatz is compared to the less FPGA resource demanding moving window deconvolution based feature extraction, currently implemented.