Research On Practical Performance Of Twin-field Quantum Key Distribution

Quantum Key Distribution(QKD)allows two remote users to share a series of random keys securely,and it is expected to realize unconditional secure communication at the level of information theory by combining with "One-Time Pad" encryption technology.Since the first QKD protocol,BB84,was proposed,the QKD field has experienced rapid development.However,for a long time,the attenuation problem of weak quantum signals in the quantum channel has always limited the communication distance of QKD,there is a so-called "linear bound" ceiling in the QKD field.However,this limit was overcome by the twin-field quantum key distribution(TFQKD).The TF-QKD is based on single-photon interference,and can improve the relationship between secret key rate and transmittance of fiber channel from O(η)to O((?)),this makes long-distance QKD communication can be expected soon.In spite of this,there are still some problems for the practical performance of TF-QKD protocol.The index of practical performance in QKD protocol mainly includes two aspects,one is the highest secret key rate that the protocol can achieve,and the other is the practical security of the protocol.While,the TF-QKD has not achieved the optimal effect in these two aspects.In the aspect of key rate,The sending or not seding TFQKD(SNS-TFQKD)variant protocol of TF-QKD does not make use of the effective components of signal pulse as much as possible in code mode,resulting in the key rate of the protocol is not optimal.In terms of the security of the protocol,the intensity modulator based on lithium niobate is usually used in the intensity modulation of the source signal.The defects of this kind of modulator may introduce side channels in the experimental system,which leads to the security problems of the protocol.In order to optimize the secret key rate of TF-QKD protocol and eliminate the side channels of the experimental system,this thesis modified the code mode of the SNS-TFQKD protocol,and studied the practical security problems of the decoy state method,studies the TF-QKD protocol from theoretical model and experimental.The main innovation of this thesis are:designing the multi-photon based SNS-TFQKD protocol,improving the overall key rate of the protocol;desinging the passive decoy state module for TF-QKD protocol to eliminate the side channels at the source end of the protocol;and studied the defect intensity modulator in the TF-QKD system,and a scheme of working point stabilization of the modulator is proposed.which eliminates the problem of side channel from the perspective of the modulator itself.The relevant research results are summarized as follows:(1)The two-photon based SNS-TFQKD protocol is proposed to improve the security key rate of the original SNS-TFQKD protocol.The analysis shows that,as an important variant of TF-QKD,In the security proof of the SNS-TFQKD protocol,only the single photon component of the signal pulse is used to generate the key,which makes the protocol not make full use of the effective component in the signal,resulting in the security rate of the original SNS-TFQKD protocol is not optimal.In order to improve the key rate of the SNS-TFQKD protocol,a protocol is proposed to extract the two-photon component from the signal pulses by designing different effective response event sets,so that the two-photon component can contribute to the key generation.Furthermore,the security proof and numerical simulation of the protocol are also given.The results show that the security key rate of the improved protocol is significantly improved compared with that of the original SNS-TFQKD at short range,and the protocol does not change the experimental system settings of the original protocol,so it is feasible in the experiment.(2)The passive decoy state based TF-QKD protocol is proposed to eliminate the side channel of decoy state modulation,avoiding the security problem caused by the active dimming of intensity modulator in TF-QKD experimental system.In the experimental system of TF-QKD,the decoy states are usually achieved by implementing intensity modulator to get different light intensities.However,due to the imperfection of the modulator used in the experiment,the modulated quantum state cannot perfectly conform to the requirements of the decoy state scheme,this introduces side channels into the experimental system and seriously threatens the security of the protocol.In the passive decoy state protocol proposed of this thesis a passive modulation optical module is designed to replace the intensity modulator.The different modes of the local detectors in the optical module are classically related to the photon number distribution of the signal pulses.Therefore,pulses with different photon number distributions can be prepared passively according to the different response modes of the local detector.Thus,the side channel induced by active dimming intensity with intensity modulator is avoided.The simulation results of the protocol show that the key rate of the passive decoy state scheme is the same as that of the active decoy state scheme in the case of four-decoy state,indicating that the protocol can improve the security of the system without damaging the key rate,and has practical feasibility.(3)In view of the side channel induced by the intensity modulator mentioned in the second point,the defects of the modulator is also studied in in-depth.An RC circuit model is used to describe the charge relaxation overpath quantitatively,furthermore,a multi-step bias stable scheme is protosed to control the bias working point.It is convincing that the bias drift of the modulator is caused by internal charge relaxation,and the RC circuit model is used to describe the charge relaxation overpath quantitatively,the effectiveness of the RC model is also verified by further experiments.Based on this,the multi-step bias stable scheme is protosed.The scheme inverts the regulation voltage scheme with required load according to the working point drift of the modulator,so as to obtain the stable optical power output at the output end.The actual experiments show that the scheme can keep the working point of the modulator in a long time range(30 to 50 hours),and the scheme does not require any feedback or detection equipment,so it does not increase the experimental complexity of the system.The simulation of TF-QKD protocol is also implemented to show the affect bias drift of the modulator.The results show that the key rate of TF-QKD protocol can be increased by one order of magnitude after the stabilization scheme is adopted compared with that before the operating point stabilization.