NIPS-CERN is proud to announce that five papers from our group have been accepted for presentation at the 10th International Symposium on Instrumentation Systems, Circuits, and Transducers (INSCIT 2026). The accepted works reflect the breadth of research carried out by students and collaborators across instrumentation, signal processing, and real-time systems for high-energy physics.
The first paper, by Melissa S. Aguiar, Dabson F. Santos, Marcos V. S. Oliveira, and Luciano M. A. Filho, presents the HLS-based firmware redesign of the ATLAS Liquid Argon Calorimeter digital trigger back-end. By replacing the legacy VHDL serial architecture with a parallel switch matrix implemented through High-Level Synthesis, the new design reduces logic complexity and cuts FPGA utilisation from 69% to 64%, while enabling faster development cycles and systematic design exploration.
The second paper, by Melissa S. Aguiar, Eder B. Kapisch, Ualison R. F. Dias, Huacheng Cai, Denis O. Damazio, Marcos V. S. Oliveira, and Luciano M. A. Filho, proposes a free-running online energy reconstruction technique based on the Moore-Penrose pseudoinverse. The method reduces pile-up reconstruction error by up to 62.5% compared to the conventional Optimal Filter, while naturally providing impulsive signal recovery and a pile-up independent design suitable for FPGA implementation.
The third paper, by Leandro da G. Ribeiro, Ualison R. F. Dias, Thiago C. A. Paschoalin, Tiago M. Quirino, and Luciano M. A. Filho, investigates the combination of Pole-Zero Cancellation (PZC) as a pre-processing stage with Wiener filtering for calorimeter energy reconstruction. The PZC stage compensates for the long-duration exponential tail introduced by the shaping circuit, stabilising the signal baseline and dramatically reducing systematic bias under high event-rate conditions.
The fourth paper, by Fábio C. Luna, Thiago C. A. Paschoalin, Tiago M. Quirino, and Luciano M. A. Filho, presents a digital IIR-based signal conditioning stage implemented on FPGA for nuclear instrumentation pulse simulation. By discretising the analogue transfer function via the Impulse Invariant Method and decomposing it into parallel first- and second-order sections with fixed-point arithmetic, the design faithfully reproduces the long-tail shaper response using less than 1% of the target FPGA resources.
The fifth paper, by Nelson Cevidanes N. de Assis, Guilherme Barroso Morett, Tiago Aparecido Teixeira, Luciano M. A. Filho, Bernardo Sotto Maior Peralva, and Gustavo Barbosa Libotte, extends the Lorenzetti framework with continuous signal flow simulation. By reinterpreting Minimum Bias data as a free-running collision stream driven by a Poisson model, the work enables the study of first-level trigger algorithms on realistic, uninterrupted data streams -- a capability not previously available in standard event-driven simulators.
We congratulate Melissa S. Aguiar, Dabson F. Santos, Eder B. Kapisch, Ualison R. F. Dias, Leandro da G. Ribeiro, Thiago C. A. Paschoalin, Tiago M. Quirino, Fábio C. Luna, Nelson Cevidanes N. de Assis, and Tiago Aparecido Teixeira, along with all co-authors and collaborators, on this outstanding set of contributions. These five papers reflect the depth and scope of work at NIPS-CERN, from ATLAS trigger firmware operating at CERN to open simulation frameworks built for the next generation of high-energy physics detectors.