Research On Continuous Variable Quantum Key Distribution Based On SC-LDPC Code

With the development of quantum science,the emergence of quantum computers has caused a serious crisis in traditional cryptography: the complex of key sequences based on mathematical computation will be easily cracked by quantum computers with powerful computational power.In the face of such a crisis,scientists have to study the new encryption method quantum key gradually into the focus of people's research.Due to their quantum physical properties,Quantum keys ensure that the communication process in the quantum channel is not acquired by the potential eavesdroppers according to the quantum uncertainty theorem and the quantum non-cloning theorem.Quantum Key Distribution(QKD)is divided into discrete variable quantum key distribution(DVQKD)and continuous variable quantum key distribution(CVQKD).From the first QKD protocol to the current CVQKD protocol,quantum Key distribution has been greatly developed,but its communication distance and key rate performance are still low,especially in the case of low signal-to-noise ratio.The QKD protocol is divided into four steps: Quantum transmission and measurement,sifting,reconciliation and secrecy amplification.Through the study of many experts,the DVQKD system has some difficult problems to solve,so that its transmission distance and key rate are unsatisfactory.The CVQKD system can achieve higher key rate and transmission distance.The transmission distance and key rate of the CVQKD system are closely related to the performance of the reconcile part,that is to say,the performance of the reconcile affects the performance of the whole CVQKD system directly.The reconcile protocol is fundamentally an error-correcting scheme,in which one party sends the coded key data as the reconciliation information to the decoding party,and the decoding party uses its own data as the edge information,then unifies the consultation information received to carry on the joint decoding.At present,the optimization of reconcile performance is mainly based on better error-correcting codes and higher efficiency decoding algorithms.Low-Density parity-check(LDPC)codes with good error correction performance is very suitable for quantum key distribution system,so the application of LDPC codes in quantum Key distribution has become a hot research topic.This paper mainly studies the reconcile part of CVQKD protocol based on spatially coupled(SC)-LDPC code.SC-LDPC code is a kind of LDPC code which can still have good error-correcting performance in low SNR.Moreover,the coding complexity and decoding complexity of the SC-LDPC code used in this paper are low,which is easy to realize by hardware.In order to solve the poor reconciliation efficiency and key generation rate performance of CVQKD system in low SNR,this paper tries to apply SC-LDPC code to CVQKD system,and proposes a reconciliation protocol for SC-LDPC code.The new decoding is based on the number of iterations in the construction of SC-LDPC code l The reconciliation information is divided into L section,and the correct key segment is reserved,and the key segment of decoding error is discarded.Under the new protocol,the error frame rate(FER)is changed,the FER under the new protocol reconcile will be less effective than the FER under the general agreement.In addition,the key rate will be greatly improved.Due to the good decoding error-correcting performance of SC-LDPC code in low SNR,the reconcile efficiency of CVQKD system will be greatly improved.In this paper,firstly,the error correcting performance of SC-LDPC code is simulated under BI AWGN channel,which verifies the good error-correcting performance of SC-LDPC code in low signal-to-noise ratio,and then simulates the multidimensional reconciliation based on SC-LDPC code.It verifies that the efficiency of reconcile based on SC-LDPC code is higher than that of General QC-LDPC code.Finally,in order to improve the speed of the quantum key reconciliation program,that is,the parallel acceleration of the reconcile decoding program,the multidimensional Reconciliation program is accelerated in parallel GPU,and the reconciliation rate is promoted effectively without sacrificing the reconciliation efficiency.