Akademska digitalna zbirka SLovenije - logo
E-viri
Celotno besedilo
  • Lessard, G.; Arsenault, P.; Martel-Dion, P.; Prentice, S.; Rossignol, T.; Retiere, F.; Charlebois, S. A.; Pratte, J.-F.

    2023 IEEE Nuclear Science Symposium, Medical Imaging Conference and International Symposium on Room-Temperature Semiconductor Detectors (NSS MIC RTSD), 2023-Nov.-4
    Conference Proceeding

    Particle physics experiments, like nEXO and ARGO, require a large-scale data acquisition system capable of supporting high data rates. One can expect a total data rate of approximately 400 Gb/s for nEXO, whereas ARGO estimates are on a scale of Pb/s. To support those experiments, which require a low-power and high-bandwidth communication (≈1 Gb/s per link), a modular Silicon-Photonics (SiP) communication module is in development to connect the front-end electronics to the data acquisition system (DAQ). This research focuses on the interface between more than 100 optical transceivers and the DAQ. To eliminate the need for custom hardware in the servers of the DAQ, the system is compatible with a standard Ethernet network. Targetting an Ethernet interface allows the use of commodity off-the-shelf equipment to connect custom electronics to the DAQ and simplifies integration, deployment, and maintenance. This system is composed of a Zynq system-on-chip (SoC), where the FPGA receives data from a set of transceivers and wraps the frame received in UDP datagrams sent to the DAQ; the processor handles various configurations and commands. To match the requirements on data rates and modularity, this study proposes a proof of concept demonstrating a 100 Gb/s link using a FPGA to transfer the data from the custom SiP transceivers to a server running DAQ software, implemented using the MIDAS framework. Following this work, the implemented Ethernet link will constitute a system ready to integrate with the SiP communication module. This integration will provide a platform to deploy large-scale and high data-rate DAQs for the targeted experiments, namely nEXO and ARGO.