Continuous-variable quantum secret sharing based on constellation shaping

Continuous-variable quantum secret sharing (CVQSS) based on discrete modulation is a viable alternative to Gaussian-modulated CVQSS due to its implementation simplicity. However, its theoretical performance remains far below the Gaussian-modulated counterparts. In this work we propose a CVQSS scheme based on constellation shaping, which enables near-Gaussian modulation effects by combining probabilistic and geometric constellation shaping with quadrature amplitude modulation (QAM). For the probabilistic shaping, binomial distribution and Boltzmann-Maxwell (BM) distribution are selected to adjust the occurrence probabilities of quantum states, and for the geometric shaping, triangular quadrature amplitude modulation (TQAM) is employed to optimize the geometric positions of the constellation points. Numerical simulations indicate that compared with other discrete-modulated CVQSS schemes, the proposed BM-distributed 16-TQAM scheme achieves the highest asymptotic secret key rate and the lowest detection error probability. Moreover, the optimal modulation variance of this scheme is higher than that of other schemes, demonstrating that the players can employ a larger modulation variance when preparing coherent states, thereby enhancing the signal-to-noise ratio of the whole system and providing a potential advantage in long-distance transmission.