Published July 11, 2025
| Version v1
Thesis
Open
Emerging Technologies and Development Methodologies for High-Throughput Data Acquisition Systems
Authors/Creators
- 1. CERN
-
2.
Aix-Marseille University
-
3.
Center for Particle Physics of Marseilles
Description
High Energy Physics experiments depend heavily on FPGA-based data acquisition systems to manage the massive data volumes produced by particle detectors.
With CERN upgrading its accelerator complex, experiments such as LHCb must handle higher throughput, necessitating the development of advanced FPGA-based readout systems.
To meet these demands, more efficient design methodologies are required to streamline development while maintaining system reliability.
This work explores modern approaches to improving FPGA gateware development, focusing on verification-driven design and open-source solutions.
Functional verification frameworks such as UVVM, OSVVM, and VUnit were evaluated and implemented to automate testing, improve test coverage, and create robust verification environments.
Formal verification techniques were applied to critical components, demonstrating their ability to detect design flaws early and significantly reduce debugging time.
A major contribution of this research is the development of the colibri library, an open-source, vendor-independent collection of reusable FPGA components.
By standardizing commonly used functionalities, colibri improves portability across different hardware platforms and simplifies integration.
The library was successfully applied in various projects, including a high-speed Ethernet-based readout system for detector data acquisition.
High-level synthesis was also explored as a means to simplify FPGA development by allowing designs to be written in C++ instead of traditional hardware description languages.
While this approach showed potential for accelerating trigger and reconstruction algorithms, current high-level synthesis tools still present challenges in achieving the performance and reliability required for large-scale applications.
The findings of this work highlight the benefits of open-source methodologies, structured verification strategies, and reusable component libraries in the development of FPGA-based readout systems for high-energy physics experiments.
These results will help guide the design of next-generation data acquisition systems at LHCb.
Files
aperro-these-v6.pdf
Files
(11.3 MB)
| Name | Size | Download all |
|---|---|---|
|
md5:ad47a2aed9560a9b92bf608246e13b71
|
11.3 MB | Preview Download |
Additional details
CERN
- Department
- EP - Experimental Physics Department
- Programme
- CERN Doctoral Student Program
- Accelerator
- CERN LHC
- Experiment
- LHCb