Joel Brogan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Joel Brogan has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Amir Sadovnik
Oak Ridge National Laboratory
1 shared publication - Arka Daw
Oak Ridge National Laboratory
1 shared publication - David Cornett
Oak Ridge National Laboratory 1 Bethel Valley Rd. Oak Ridge Tennessee
1 shared publication - David S. Bolme
Oak Ridge National Laboratory
1 shared publication - Edmon Begoli
Oak Ridge National Laboratory
1 shared publication - Matthew Yohe
Oak Ridge National Laboratory 1 Bethel Valley Rd. Oak Ridge Tennessee
1 shared publication
Latest Publications
- 26 July 2025
- Published by Springer Nature
- p. 289-302
Bridging the Gap from Research to Reality: Methods for Fortifying Mitigation Measures Against Adversarial AI
- 26 July 2025
- Published by Springer Nature
- p. 289-302
Bridging the Gap from Research to Reality: Methods for Fortifying Mitigation Measures Against Adversarial AI
- Open Access
CHAMP: A Configurable, Hot-Swappable Edge Architecture for Adaptive Biometric Tasks
What if you could piece together your own custom biometrics and AI analysis system, a bit like LEGO blocks? We aim to bring that technology to field operators in the field who require flexible, high-performance edge AI system that can be adapted on a moment's notice. This paper introduces CHAMP (Configurable Hot-swappable Architecture for Machine Perception), a modular edge computing platform that allows operators to dynamically swap in specialized AI "capability cartridges" for tasks like face recognition, object tracking, and document analysis. CHAMP leverages low-power FPGA-based accelerators on a high-throughput bus, orchestrated by a custom operating system (VDiSK) to enable plug-and-play AI pipelines and cryptographically secured biometric datasets. In this paper we describe the CHAMP design, including its modular scaling with multiple accelerators and the VDiSK operating system for runtime reconfiguration, along with its cryptographic capabilities to keep data stored on modules safe and private. Experiments demonstrate near-linear throughput scaling from 1 to 5 neural compute accelerators, highlighting both the performance gains and saturation limits of the USB3-based bus. Finally, we discuss applications of CHAMP in field biometrics, surveillance, and disaster response, and outline future improvements in bus protocols, cartridge capabilities, and system software.
Open AccessCHAMP: A Configurable, Hot-Swappable Edge Architecture for Adaptive Biometric Tasks
What if you could piece together your own custom biometrics and AI analysis system, a bit like LEGO blocks? We aim to bring that technology to field operators in the field who require flexible, high-performance edge AI system that can be adapted on a moment's notice. This paper introduces CHAMP (Configurable Hot-swappable Architecture for Machine Perception), a modular edge computing platform that allows operators to dynamically swap in specialized AI "capability cartridges" for tasks like face recognition, object tracking, and document analysis. CHAMP leverages low-power FPGA-based accelerators on a high-throughput bus, orchestrated by a custom operating system (VDiSK) to enable plug-and-play AI pipelines and cryptographically secured biometric datasets. In this paper we describe the CHAMP design, including its modular scaling with multiple accelerators and the VDiSK operating system for runtime reconfiguration, along with its cryptographic capabilities to keep data stored on modules safe and private. Experiments demonstrate near-linear throughput scaling from 1 to 5 neural compute accelerators, highlighting both the performance gains and saturation limits of the USB3-based bus. Finally, we discuss applications of CHAMP in field biometrics, surveillance, and disaster response, and outline future improvements in bus protocols, cartridge capabilities, and system software.