A Research on Black-box Evaluation Methods for Quantum Key Distribution Networks

【Objective】The Quantum Key Distribution Network (QKDN) is the large-scale implementation of Quantum Key Distribution (QKD), which is typically deployed within optic fiber communication infrastructure networks. Based on the principles of quantum physics, the key distribution process in QKDN is considered to be theoretically absolutely secure. Any unauthorized interference or measurement will destroy the quantum state and be immediately detected, ensuring the security and integrity of communication. However, since quantum state storage technology is not yet mature, the expansion from point-to-point QKD to multi-point-to-multi-point QKDN needs to be realized with the help of classical technologies, which also brings new security threats. Denial-of-Service (DoS) attacks are a common form of attack in classical networks, widely exploited by attackers, and the application layer of QKDN also faces security challenges.【Methods】The paper proposes an Adaptive Range Threshold (ART) detection method to detect DoS attacks. It automatically adjusts the threshold based on abnormal fluctuations in network traffic patterns, thereby flexibly identifying and responding to DoS attacks.【Results】Simulation results show that the Detection Success Rate (DSR) of QKDN is effectively improved by the ART-based DoS attack detection method, and the DSR can reach 80% by setting reasonable parameters. Additionally, the network security of the QKDN is measured from two perspectives: delay time and request success rate.【Conclusion】The paper identifies potential network security threats by setting dynamic thresholds and evaluates network security from multiple dimensions. This provides an effective security assessment method for the actual operation of QKDN and lays the foundation for future development of security defense strategies for quantum information networks.