Chaoze Wang

Satellite-based quantum key distribution (QKD) holds the potential to establish global quantum communication networks. While satellite QKD has been extensively demonstrated using polarization encoding, time-bin encoding offers distinct advantages, such as simplifying polarization-maintaining telescope designs and being insensitive to satellite-ground relative motion. In this work, we developed a high-speed 625-MHz QKD light source using a robust Sagnac-interferometer-based modulation scheme and subsequently demonstrated time-bin QKD over a 7 km urban terrestrial free-space channel. This experiment successfully operated over a channel traversing an equivalent atmospheric thickness and loss exceeding that of typical satellite-to-ground links, achieving a low quantum bit error rate of 0.87% and a secure key rate of 134 bps@51.4 dB. Furthermore, by using a half-wave plate to simulate the polarization basis rotation inherent in such links, we verified the robustness of time-bin encoding for satellite scenarios. The results validate time-bin encoding as a compelling alternative and lay the technical foundation for future satellite QKD applications.

Qubit-based synchronization offers a novel approach for quantum key distribution (QKD), simplifying system architecture and reducing implementation costs by leveraging the exchanged qubits. However, the performance of such schemes hinges on accumulating sufficient qubit events, rendering them vulnerable to local clock drift—particularly in high-channel-loss scenarios. This study investigates the impact of frequency instability in non-ideal oscillators, emphasizing clock drift-induced deterioration on QKD performance in these challenging, lossy conditions. We develop a computational model to quantify the secure key rate of QKD systems as a function of clock frequency stability. Through simulations and experimental validation with two distinct clock configurations, we demonstrate that oscillator stability becomes a key bottleneck in high-loss scenarios. By integrating target frequency scanning and clock offset recovery method, we verify our model via Monte Carlo simulations. Experimental validation confirms these findings, demonstrating a secure key rate of 0.37 bps at 67 dB channel loss—empirically validating the trade-off relationship among clock frequency stability, channel loss and acquisition time.