Quantum-enhanced trusted execution with dynamic policy optimization for zero-trust edge computing
Abstract
Summary Zero-trust edge computing faces escalating security challenges from quantum threats, heterogeneous resources, and dynamic operational conditions. This paper proposes Q-TEE-DSO, a collaborative architecture integrating quantum-enhanced trusted execution environments with post-quantum cryptography and reinforcement learning-based dynamic security optimization. By embedding quantum random number generation and lightweight quantum key distribution into hardware-enforced TEEs, the framework strengthens trust-root initialization and session-key establishment under quantum-aware threat models. Instruction-level optimization of lattice-based PQC algorithms, combined with enclave-aware memory isolation and side-channel mitigation, reduces cryptographic overhead on constrained edge devices. A reinforcement learning-driven DSO-Engine dynamically adapts encryption strength, access control policies, and privacy-preserving computation modes according to real-time system load and threat levels. Experimental evaluations on heterogeneous testbeds demonstrate a 37% reduction in cryptographic latency and 42% improvement in energy efficiency compared with conventional approaches, alongside 99.88% availability, robust heterogeneous compatibility, and stable long-term operation across diverse network topologies.