Adaptive Continuous Trust Framework for Zero Trust-Enabled Network Slice Allocation in 6G Networks

Abstract

Future 6G networks are designed to accommodate a large number of interconnected devices through technologies such as network slicing. However, sharing of common physical infrastructure creates significant security challenges that cannot be effectively addressed by conventional one-time authentication mechanisms. This paper proposes an adaptive continuous trust framework based on Zero Trust principles for secure slice allocation in 6G networks. The framework combines the machine learning-based behavioural analysis, continuous trust evaluation using Exponential Moving Average (EMA), trust-aware network slicing, and automated incident response using n8n orchestration. The framework was evaluated using the CICIDS2017 Wednesday WorkingHours dataset after Min-Max normalization. Four machine learning models were compared; the highest performing model was Random Forest with 94.87% accuracy and a 3.25% false positive rate. The results show that the proposed framework improves intrusion detection and allows secure and adaptive resource allocation in future 6G networks.