Júnia is a 40th-cycle Ph.D. candidate in Electronics, Telecommunications, and Information Technologies Engineering at the University of Bologna. She also holds a research fellowship at the “Ercole De Castro” Advanced Research Center on Electronic Systems for Information and Communication Engineering, ARCES. Her work focuses on cybersecurity, artificial intelligence, and embedded-system protection.
Her doctoral research develops artificial intelligence methods for detecting code-reuse attacks through instruction-level execution traces. The work covers data acquisition and processing, benign and malicious sample generation, compact neural-network design, semi-supervised learning, quantization, and evaluation on resource-constrained platforms.
Education
Since 2024, she has been enrolled in the Ph.D. programme in Electronics, Telecommunications, and Information Technologies Engineering at the University of Bologna.
She received an M.Sc. in Computer Science from the Federal University of Minas Gerais, Brazil, in 2024. Her dissertation presented a modular framework based on the Network Data Analytics Function, NWDAF, for data analysis and anomaly detection in 5G networks.
She received a B.Sc. in Telecommunications Engineering in 2021 and a B.Sc. in Science and Technology in 2019 from the Federal University of São João del-Rei, Brazil.
Academic career and research activities
As a research fellow at ARCES, she works on machine-learning methods for monitoring and protecting cyber-physical and embedded systems.
Since November 2024, she has contributed to the European KDT JU ISOLDE project at the University of Bologna. Her activities concern software-behaviour analysis using execution data and artificial intelligence models.
Since 2023, she has collaborated with the Programmability, Orchestration and Virtualization for 5G project, which addresses programmable network infrastructures. From 2023 to 2025, she provided technical support to the Slicing Future Internet Infrastructures project.
From October 2023 to January 2024, she conducted research at the University of Udine on indoor positioning, wireless communications, and signal analysis under multipath propagation conditions.
From May 2021 to September 2023, she held a research grant within the Analytical Capabilities Program of the Public Ministry of Minas Gerais, contributing to machine-learning solutions for operational data analysis.
From 2021 to 2022, she collaborated with a Federal University of Minas Gerais project focused on connectivity for Indigenous and Quilombola communities.
Teaching activities
From February 2023 to August 2024, she worked as a teaching tutor in 5G networks at the Federal University of Minas Gerais. She supported students, prepared teaching materials, and covered topics related to mobile networks, network-function virtualization, and data analytics.
Scientific activities
Her research interests include embedded-system security, anomaly detection, machine learning for cybersecurity, efficient neural networks, edge AI, wireless communications, and 5G-and-beyond networks.
She has collaborated with research groups in Italy, Brazil, Spain, and Portugal. Her work has been presented at international conferences in networking, embedded systems, and cybersecurity.
Peer-review activities
She has served as a reviewer for scientific conferences, including IEEE NOMS, IEEE LATINCOM, ICECET, CSBC, and SBRC. She was a member of the Brazilian Computer Society from 2022 to 2024.
Skills
She has experience with Python, PyTorch, TensorFlow, C, C++, ONNX, and tools for program and execution-trace analysis. Her expertise includes deep learning, anomaly detection, explainable AI, quantization, robustness evaluation, network-data analysis, and neural-model design for resource-constrained devices.
Languages
Portuguese: native.
English: C1, IELTS 7.5.
Italian: B1.
Software registrations
Empreenda Já — Brazilian computer-program registration no. BR 512020002494-7, registered on 11 November 2020. The software includes application, support-tool, control, information-management, and report-generation functions. The registration was completed with other authors.
Doação Inteligente — Brazilian computer-program registration no. BR 512020001186-1, registered on 26 June 2020. The software includes application, control, information-management, database-management, and report-generation functions.
Awards and recognition
In 2022, she received second place in Latin America in the IEEE Communications Society “Internet for All” initiative.
In 2020, she received the Diploma of the Order of Builders of Progress from the Academy of Sciences and Letters of Conselheiro Lafaiete for the community activities conducted by the Rede Família project.
In 2018, she obtained first place in a research-project competition at the Federal Institute of Minas Gerais.
