Study On Reconciliation Protocol For Continuous Variable Quantum Key Distribution

Continuous variable quantum key distribution(CVQKD) is a hot topic in quantum key distribution, because it has mature quantum signal source and measurement device, high communication capacity. However, there are some errors in the key because of the presence of noise in quantum channel and experiment device. Reconciliation is necessary for high efficient CVQKD. Efficient channel coding is essential to realize reconciliation for CVQKD. Polar codes which were proposed by Arikan in 2007 can achieve the channel capacity and have lower encoding and decoding complexity. Thus we study the reconciliation protocol based on Polar codes for CVQKD in this thesis.The main results are as follows:(1) In the thesis, a novel reconciliation protocol based on multidimensional reconciliation using Polar codes is proposed. This protocol consists of two components, one is the multidimensional algorithm section, and the other is error correction section using Polar codes. In the first section, the long sifted key is divided into a set of d-dimension vectors. Then every d-dimension vector is transformed to a binary data by normalization, sphere and rotation transform operation. In the other section, a specific decoding scheme with side information for Polar codes is presented, where the frozen bits locations are used as the reconciliation information. The receiving user decodes the final common key with a proper decoding algorithm after receiving the reconciliation information. Simulations show that the efficiency of the protocol is higher than that of protocol using LDPC under the same condition.(2) In addition, a high-speed CVQKD reconciliation protocol based on bi-direction error correction is proposed. In this protocol, each user divides his long sifted key into two substrings. Then both users encode one of their substrings at the same time and send the frozen bits location as reconciliation information to the other user. The other user decodes the corresponding substring to get the final common key at the help of the received reconciliation information. Numerical simulations show that this proposed protocol increases the throughput and key rate by nearly 2 fold, and the reconciliation throughput improves greatly as the key length is longer.(3) Finally, a novel reconciliation protocol for CVQKD based on multi-level coded(MLC) modulation using Polar cascade codes is proposed. The protocol introduces best quantization or uniform quantization method to quantify Bob’s continuous key into multi-level binary sequences. Then Bob chooses the code rate of Polar cascade codes for every level in MLC system and sends the syndrome to Alice. Alice finally decodes the common binary key with Bob using the multi-stage decode(MSD) algorithm. Numerical simulations show the proposed protocol can improve the efficiency compared to that using LDPC under the same condition.