Study and Design of Efficient and Resilient Quantum Key Distribution System

Abstract

This work investigates the implementation of a secure Quantum Key Distribution (QKD) protocol, BB84, within an Optical Data Center Network (ODCN), focusing on key negotiation under varying noise levels and adversarial attacks. A Fully Classical Model, where encoding and noise were simulated using classical methods, and a Hybrid Model, which uses Qiskit SDK to simulate quantum noise and quantum-specific encoding, were developed to assess key metrics like Quantum Bit Error Rate(QBER) and Secure Key Rate(SKR). Two error-correction techniques were used: Hamming, and Cascade. Cascade was able to manage higher error rates; the QBER threshold was chosen liberally---0.25, particularly in the presence of an eavesdropper. The network’s performance, measured by the Success Ratio of Connection Requests (SRCR), Time-slot Utilization Ratio (TUR) and Network Security Performance (NSP) were similar to the values available in literature. The higher error rate and eavesdropping degraded the performance of key negotiation, which lowered the SKR, underscoring the system's vulnerability at high QBER values.