NISLAB - Network and Information Security Laboratory

George Emil Palade University of Medicine, Pharmacy, Science and Technology of Târgu Mureş

Who are we?

The Network and Information Security Laboratory (NISLAB) is a young and dynamic team of professors, researchers, post-docs, PhD and MSc students. NISLAB activates inside the George Emil Palade University of Medicine, Pharmacy, Science and Technology of Târgu Mureş (UMFST), a research and education university in the heart of the Romanian Transylvania. As a research group, the NISLAB team focuses on researching an understanding complex problems on information and network security design, intrusion detection and monitoring systems, data privacy, and machine learning modeling.

Research

Networks and Information Security

Information and critical infrastructure systems can be seen as the backbone of today's modern digital ecosystems. NISLAB's expertise on this matter is on critical infrastructure and cyber-physical system protection, automotive and traditional networking systems. In terms of research topic, NISLAB is active in the domain of anomaly detection and intrusion detection, security protocols and security-aware network design.

Data Privacy

Correlating data points to user behaviours is an on-going issue that threats a system's user's privacy. The NISLAB team actively researches emerging data privacy concerns, by designing and developing methods focused on preserving the privacy and keeping the required data utility in time series data streams, privacy-oriented feature selection, scoring, and metrics. With the emerging trend and popularity of Artificial Intelligence models, the NISLAB team prioritises as well the implications of data privacy in this domain as well.

Machine Learning

NISLAB considers Machine Learning an essential tool, that can be employed to enhanced a system's security. Through Machine Learning-based modeling, NISLAB tackles to solve problems concerned with anomaly and outlier detection, fault detection, and data tampering attempts. Through statistical data analysis and feature selection methodology, NISLAB researches how to design distributed and fault tolerant ensembles of models and decision fusion methods.