root@delano:/home/Research/Algorithms/$ ls
The real-time algorithm I have worked on is hosted on custom hardware that we designed in-house at UCLA using powerful, state-of-the-art FPGAs. The algorithm constructs high-quality muons in real time using a Kalman filter approach in the barrel region of the CMS detector. We utilize on-detector electronics in the barrel region to provide raw detector information to the algorithm, which blends many layers together and reconstructs muons using physics-informed, complex matrix algebra in five dimensions. Since both the algorithms and the hardware on which they are deployed are custom-built and developed entirely at UCLA by the Bachtis High-Energy Physics Group, deploying the algorithms as firmware onto the FPGAs presents a significant challenge. We host complex, payload-agnostic firmware infrastructure on the FPGAs that accepts the custom algorithms as payloads. We use Vivado and Vitis HLS to ensure that the algorithms are deployed correctly onto the hardware.