Modernizing Network Defense at UFVJM: Practical Simulation Using Open-Source Tools and Artificial Intelligence

Abstract

Continuous financial constraints in Brazilian public universities impose a severe bottleneck on updating their information security assets. Concurrently, mandatory compliance with the General Data Protection Law (LGPD) demands rapid and precise responses to cyber incidents. To address this disparity between legal requirements and budgetary limitations, this study reports a practical simulation focusing on the use of open-source platforms (Wazuh and Zeek) combined with machine learning, applied to the context of the Federal University of the Jequitinhonha and Mucuri Valleys (UFVJM) - Campus Mucuri. The research replicated brute-force and distributed denial-of-service (DDoS) attacks within a virtualized environment mirroring the campus's actual topology. The analysis demonstrated that replacing purely static rule-based defenses with Artificial Intelligence algorithms reduces the triage time for false positives and autonomously blocks atypical anomalies. We conclude that the strategic employment of free software not only eliminates the reliance on costly public bidding processes for commercial system licensing but also ensures the protection of the institution's digital assets, even amidst human and financial resource scarcity.