Dynamic Trust Scoring for Zero Trust at the Edge: A Multi-Factor, Context-Aware Approach

Zero Trust architectures are increasingly critical for securing distributed, resource-constrained edge environments, where traditional perimeter-based models are insufficient. This paper presents a Dynamic Trust Scoring designed specifically for Zero Trust edge networks. Our system implements context-aware trust evaluation that adapts to device types, network conditions, and operational requirements. The framework incorporates four key components: (1) advanced trust factors (time, patch, certificate, and behavioral), (2) adaptive weighting system, (3) historical analysis, and (4) comprehensive visualization tools. Through extensive simulation of both normal and adversarial scenarios, we demonstrate the system's effectiveness in maintaining high trust scores for normal devices while rapidly detecting compromised behavior. The framework's ability to adapt to changing conditions and provide fine-grained trust evaluation makes it particularly suitable for mission-critical edge applications, such as autonomous vehicles, industrial IoT, and public safety communications. Our results show that the system can effectively distinguish between benign and compromised devices, with trust scores dropping by up to 50% for compromised devices while maintaining stable high scores for normal operation. The framework's reproducibility features and comprehensive statistical analysis capabilities further enhance its practical applicability in real-world Zero Trust deployments.