Log-Based Threat Detection Simulation for Private Cloud Security within a Zero-Trust Architecture Framework
Taonashe Trevor Madziva, Fungai Mukoko · International Journal of Computer Science and Mobile Computing · 2025
This paper presents a novel framework for log-based threat detection in private cloud environments using Zero-Trust Architecture (ZTA) principles. With the increasing adoption of private clouds, traditional perimeter-based security models have proven inadequate against sophisticated attacks. Our research addresses this gap by developing a simulated environment that integrates continuous log monitoring with machine learning-based anomaly detection. The system demonstrates 92.3% detection accuracy for simulated attacks, including brute force attempts and privilege escalation, while maintaining sub-second response times. The implementation provides a practical blueprint for organizations transitioning to ZTA in private cloud deployments. Furthermore, the study quantifies the efficacy of various log sources, demonstrating that network logs offer the strongest attack signals. Contextual analysis is shown to significantly reduce false positives, enhancing the system's reliability and operational efficiency.