High-throughput and low-cost LDPC reconciliation for quantum key distribution

Reconciliation is a crucial procedure in post-processing of quantum key distribution (QKD), which is used for correcting the error bits in sifted key strings. Although most studies about reconciliation of QKD focus on how to improve the efficiency, throughput optimizations have become the highlight in high-speed QKD systems. Many researchers adopt high-cost GPU implementations to improve the throughput. In this paper, an alternative high-throughput and high-efficiency solution implemented in low-cost CPU is proposed. The main contribution of the research is the design of a quantized LDPC decoder including improved RCBP-based check node processing and saturation-oriented variable node processing. Experiment results show that the throughput up to 60 Mbps is achieved using the bidirectional approach with reconciliation efficiency approaching to 1.1, which is the optimal combination of throughput and efficiency in discrete-variable QKD. Meanwhile, the performance remains stable when quantum bit error rate varies from 1 to 8%.