Improving the performance of reference-frame-independent quantum key distribution through a turbulent atmosphere

Reference-frame-independent quantum key distribution (RFI-QKD) can dispense with the requirements of active alignment on reference frames between legitimate users, which is beneficial for free-space implementation. However, the fluctuating transmittance due to atmospheric turbulence still remains a great challenge for improving the performance and has been seldom addressed. In this paper we extend the recently proposed prefixed-threshold real-time selection method to RFI-QKD while combining practical consideration of the transmittance probability distribution model based on the finite aperture theory. Through numerical simulations, we present an estimation for the variance of the log-normal model with respect to distance and receiving aperture radius, and demonstrate the effectiveness of using this method in the RFI protocol. Considering the finite-key effects, simulation results show that the gap of the key rate with different reference frame deviations can be alleviated by increasing the data size. By adopting this method one can tolerate more serious transmission loss, especially in strong turbulence cases, which is helpful for future free-space experimental designs.