An EVALUATION OF NETWORK SECURITY THREATS ON THE FEDERAL UNIVERSITY LOKOJA WEBSITE USING SNORT INTRUSION DETECTION SYSTEM.

Abstract

Network intrusion detection systems capable of accurately and efficiently detecting and responding to security threats are essential for modern cybersecurity infrastructure. This research presents the implementation and performance evaluation of a comprehensive Snort-based intrusion detection system (IDS) for the Federal University Lokoja website. The system was deployed on a Virtual Private Server (VPS) running Ubuntu Server 20.04 LTS with Snort version 3.1.74.0, incorporating real-time packet analysis, threat detection, and automated alert generation mechanisms. The implementation involved configuring custom rule sets for detecting SQL injection attempts, DDoS attacks, port scanning, and unauthorized access patterns. Performance evaluation revealed detection rates of % for SQL injection, 94.5% for DDoS attacks, 98.2% for port scanning, and 92.3% for unauthorized access attempts, with corresponding false positive rates of 2.1%, 3.2%, 1.8%, and 4.1% respectively. The automated alert notification system successfully delivered real-time alerts through email notifications with an average delivery time of 12 seconds and a success rate of 99.2%. Testing was conducted using simulated attack scenarios from Kali Linux virtual machines, demonstrating the system's effectiveness in protecting university web infrastructure against common cyber threats.