Efficient routing algorithm for trusted relay quantum key distribution networks via quantum reinforcement learning.

Area of Science:

• Quantum Information Science
• Network Security
• Artificial Intelligence

Background:

• Trusted relay quantum key distribution networks (TR-QKDNs) require efficient routing algorithms for scalability.
• Current routing methods face limitations in adaptability, parameter configuration, and computational complexity.

Purpose of the Study:

• To develop an intelligent routing algorithm for TR-QKDNs that dynamically optimizes decisions.
• To address the limitations of existing routing approaches in complex network environments.

Main Methods:

• Proposed a hybrid quantum deep deterministic policy gradient (HQ-DDPG) algorithm integrating a quantum neural network (QNN) with deep deterministic policy gradient (DDPG).
• Utilized a quantum single-source shortest path (QSSP) algorithm to reduce computational complexity.
• Leveraged distributed quantum computing for QNN implementation.

Main Results:

• HQ-DDPG demonstrated nearly double the training convergence speed and reduced resource requirements by approximately 45 times compared to DDPG.
• Achieved a quantum key delivery ratio above 91.35% under high-load demand, outperforming DDPG and optimized link state routing (OLSR).
• The QNN enabled efficient solutions for large-scale TR-QKDN problems with fewer quantum resources.

Conclusions:

• The HQ-DDPG algorithm offers a superior, efficient, and scalable solution for routing in TR-QKDNs.
• Quantum reinforcement learning shows significant potential for optimizing complex network operations.
• The proposed approach enhances network expressiveness and performance in demanding conditions.