ESTIMATION OF QUANTUM TIME LENGTH FOR ROUND-ROBIN SCHEDULING ALGORITHM USING NEURAL NETWORKS

N.B. Karayiannis, R. Kretzschmar, Hans Richner

Toshihiko Sasaki, Masato Koashi

Abstract The round-robin differential phase shift (RRDPS) quantum key distribution (QKD) protocol is a unique QKD protocol whose security has not been understood through an information-disturbance trade-off relation, and a sufficient amount of privacy amplification was given independently of signal disturbance. Here, we discuss the security of the RRDPS protocol in the asymptotic regime when a good estimate of the bit error rate is available as a measure of signal disturbance. The uniqueness of the RRDPS protocol shows up as a peculiar form of information-disturbance trade-off curve. When the length of a block of pulses used for encoding and the signal disturbance are both small, it provides a significantly better key rate than that from the original security proof. On the other hand, when the block length is large, the use of the signal disturbance makes little improvement in the key rate. Our analysis will bridge a gap between the RRDPS protocol and the conventional QKD protocols.

憲明 幸田

Laxmidhar Behera

Rong Wang, Zhen−Qiang Yin, Shuang Wang +3 more

In recent years, round-robin-differential-phase-shift (RRDPS) quantum key distribution (QKD) has attracted great attention for its unique characteristics, i.e., the information leakage can be bounded without learning bit error rate. Though the RRDPS QKD has made a breakthrough, it is still a question of how RRDPS will perform with monitoring signal disturbance, e.g., decoy-state and error rate statistics are both used. Here, we present simulations to study RRDPS protocol while monitoring signal disturbance. To our excitement, when using the infinite decoy-states method, RRDPS protocol can outperform the commonly used Bennett and Brassard 1984 (BB84) protocol in terms of channel length under typical experimental parameters. In the case of finite decoy states, we find that only two decoy-states and one signal state are sufficient to obtain performance very close to the infinite decoy-states case. Our simulations prove that RRDPS is a competitive protocol in real-life situations.

Manuel Pegalájar Cuéllar, M.C. Pegalajar, L. G. B. Ruiz +1 more