A method for characterizing light source defects in measurement-independent quantum key distribution systems

Abstract

The present invention proposes a method for quantitatively characterizing the actual light source defects in a measurement-independent quantum key distribution system based on two-photon interference. The present invention splits the light into two upper and lower branches, wherein the upper branch is encoded and modulated in a decoy state through a polarization modulation unit and an intensity modulation unit, and the lower branch is used as an unmodulated reference light. The light of the two branches interferes on the beam splitter of the monitoring module, and the monitoring and characterization of the light source defects are achieved through interference visibility. In terms of theoretical methods, the present invention establishes the relationship between interference visibility and fidelity, and characterizes the actual vulnerabilities existing at the transmitting end through interference visibility. Compared with the prior art, the present invention can characterize high-dimensional security vulnerabilities, adapt to the application scenarios of decoy states, effectively improve the security of the solution, and is close to the device-independent quantum key distribution protocol in terms of security. In addition, a longer transmission distance can still be obtained when there are large errors in the light source.

Description

Technical Field

[0001] The present invention relates to the technical field of quantum key distribution, and in particular to a method for characterizing light source defects in a measurement-independent quantum key distribution system. Background Art

[0002] Quantum key distribution (QKD) is a way to achieve secure communication between the communicating parties Alice and Bob by using quantum states as a carrier for key distribution. Unlike traditional key distribution algorithms based on mathematical complexity (such as the RSA algorithm), the security of QKD is based on the laws of quantum mechanics. In the QKD process, any eavesdropping behavior will leave traces. By tracking these traces and performing data post-processing, the amount of key information obtained by the eavesdropper can be reduced to almost zero. In other words, the security of QKD does not depend on the computing power of the eavesdropper Eve. Since Bennett and Brassard proposed the first BB84 protocol in 1984, people have done a lot of related research on both security and practicality.

[0003] However, there is still a gap or so-called security vulnerability between the security proof and actual implementation of QKD. In 2007, a device-independent quantum key distribution protocol was proposed (Phy. Rev. Lett. 98, 230501), which is naturally immune to all side channel attacks, but is limited in practicality due to the high requirements for detector efficiency and channel loss. As an alternative, the measurement device-independent quantum key distribution protocol proposed in 2012 eliminates all side channels of the vulnerable detection end (Phys. Rev. Lett. 108, 130503). The dual-field quantum key distribution protocol proposed in 2018 maintains the advantages of the measurement device-independent quantum key distribution protocol (Nat. 557, 400-403) and breaks the PLOB boundary. However, the light source end is still prone to modulation deviation and attack problems.

[0004] In recent years, in order to solve the security vulnerabilities of the light source end to improve the security of quantum key distribution, many works have been proposed, including the loss-tolerant method (Phys. Rev. A 90, 052314) proposed in 2019, which introduced a general phase error rate estimation method, under which the influence of the error of the state prepared by the phase randomized light source can be ignored. The reference technique (Phys. Rev. Applied 15, 034072) proposed in 2021 relatively completely characterizes the security vulnerabilities of the light source end through a parameter. In these works, the loss-tolerant method only considers the state preparation error in two-dimensional space, and the reference technology is difficult to achieve a practical secure coding rate. On the other hand, side channels caused by mode-dependent effects, such as classical pulse correlation (Quantum 5, 602) or the distinguishability of decoy states and signal states (Phys. Rev. A 98, 012330), require additional parameters to characterize light source security vulnerabilities. Although they have been solved in theory, these parameters are often difficult to measure in experiments (Sci. Bull. 67, 2167), making high-dimensional light source leakage difficult to quantitatively characterize in practice. In 2021, Duplinskiy et al. proposed a new method to characterize passive side channel information leakage by monitoring Hong-Ou-Mandel (HOM) interferometry visibility (Phys. Rev. A. 104, 012601), and implemented this method in the BB84 protocol, but did not consider the decoy state method and the distinguishability between signal states and decoy states. Summary of the invention

[0005] Purpose of the invention: The purpose of the present invention is to provide a method and system for characterizing light source defects in measurement-independent quantum key distribution systems (QKD), which solves the problem of low coding rate of existing quantum key distribution protocols while ensuring high security. The present invention provides a method for quantitatively characterizing light source defects by measuring the HOM interference visibility between a reference pulse and a modulated pulse, which can achieve security close to that of DI-QKD when combined with protocols such as MDI-QKD and TF-QKD. In order to adapt to the application scenarios of the decoy state, the present invention reconsiders the distinguishability of the signal state and the decoy state, and estimates the response rate and error rate of single-photon pulses. The feasible practical system designed by the present invention has similar equipment requirements to the original protocol and is highly practical.

[0006] Technical solution: The present invention proposes a method for characterizing light source defects in a measurement-independent quantum key distribution system (QKD), specifically comprising:

[0007] Step 1: Alice and Bob prepare phase-randomized weak coherent optical pulses respectively, and polarization encode the pulses respectively, and randomly select the basis, bit value and light source intensity;

[0008] Alice and Bob send the prepared signal state to the intermediate measurement device Charlie, and at the same time monitor the interference visibility of the light source and record the two-photon interference visibility value.

[0009] Step 2: Charlie performs Bell state measurement on the signal states from Alice and Bob, retains the events of successful Bell state measurement, and records the responding detector and the time when the response occurs;

[0010] Step 3, repeat steps 1 and 2 above. After the distribution is completed, Charlie publishes his response detector and response occurrence time at the effective time. Then Alice and Bob each publish the light intensity and basis vector information at the effective time, and take the same basis vector to form the original key under their respective basis vectors;

[0011] Step 4: Alice and Bob estimate the gain and quantum bit error rate. If the estimated gain and quantum bit error rate meet the coding requirements, the protocol continues; otherwise, the protocol is abandoned and the above process is repeated.

[0012] In step 5, Alice and Bob perform error correction and privacy amplification and generate the final security key.

[0013] Furthermore, the specific process of polarization encoding the pulse in step 1 is:

[0014] The polarization modulation options are {0°(|H>), 90°(|V>), 45°(|D>), 135°(|A>)}, where the Z basis prepares pulses with polarization information |H>, |V>, and the X basis prepares pulses with polarization information |D>, |A>. Alice and Bob respectively x and 1-p x The probability of choosing X basis and Z basis, where p x ∈[0,1]; Alice and Bob select the basis vectors and then select polarization information with equal probability to modulate the pulses. The pulses are emitted by three light sources with different intensities, p and μ 、p v With the probability of p0, a signal state pulse with intensity μ, a decoy state pulse with intensity v, and a pulse with intensity 0 are emitted.

[0015] Furthermore, we establish the fidelity F(ρ x ,ρ y ) and the connection between two-photon interference visibility;

[0016] F(ρx ,ρ y ) represents ρ x (λ) and ρ y (λ), x and y represent the pulse intensity, where x,y∈{μ,v,0}. The density matrix ρ of the light source ω Represented as its original density matrix and high-dimensional information ρ ω The tensor product of (λ), ρ ω It is expressed as:

[0017] Define the state Eve uses to distinguish between signal state and decoy state as ρ ω (λ), λ is the parameter measured by Eve, which refers to time or frequency, and ω is the intensity of the pulse, where ω∈{x,y};

[0018] F(ρ x ,ρ y ) The fidelity is calculated by the following formula:

[0019]

[0020] in Used to represent ρ x and ρ y The similarity between Represents the Poisson distribution satisfied by the number of pulse photons.

[0021] Furthermore, the calculation formulas for the lower bound of the single photon counting rate and the upper bound of the bit error rate are:

[0022]

[0023] Where Q xy and E. xy denote the total gain and bit error rate when the pulse strength from Alice is x and the pulse strength from Bob is y, respectively, x,y∈{μ,v,0}, where the upper and lower bounds of the parameters are indicated by overbars and underbars, respectively;

[0024] Classical decoy state techniques usually assume that the count rate and bit error rate of pulses are independent of intensity, so the count rate and bit error rate of pulses of different intensities are always equal. However, if Eve attacks Alice (Bob) in high dimensions and obtains relevant information, then the assumption is no longer accurate, as shown below: The state used by Eve to distinguish between the signal state and the decoy state is defined as ρ ω (λ), λ is the parameter measured by Eve, which can be time, frequency, etc., ω is the intensity of the pulse, D xy Used to represent ρ x (λ) and ρy (λ) is the trace distance, that is:

[0025] By the following constraints:

[0026]

[0027] in, It is expressed as the number of pulse photons sent by Alice is n, the number of pulse photons sent by Bob is m, and both are photon counting rates at x intensity;

[0028] Furthermore, the basis mismatch error is calculated by the following process:

[0029] First, the basis vector mismatch error △ is defined as:

[0030]

[0031] Among them, F X,Z is the fidelity between basis vectors, based on the Burse angle and triangle inequality:

[0032]

[0033] X i and Z j Represent the auxiliary matrix of X basis and the auxiliary matrix of Y basis respectively, where i,j∈{0,1}, B is used to represent X basis or Z basis;

[0034] Considering that the fidelity in the inequality is equal, and a worst case where all fidelity F′ is calculated by the minimum interference visibility V′, we get:

[0035]

[0036] Furthermore, the correction process of single-photon phase bit error rate considering basis vector mismatch is:

[0037] First, considering that Eve uses a lossless channel to attack, the basis mismatch is corrected to:

[0038]

[0039] Considering the corrected basis vector mismatch, the upper bound of the single-photon phase bit error rate is corrected to

[0040]

[0041] Furthermore, the coding rate formula is:

[0042]

[0043] Among them, P μz is the probability of being prepared in the Z basis with strength μ, is the probability that a weakly coherent light source emits a single-photon pulse with intensity μ, and In particular, the lower bound of the single photon counting rate and the upper bound of the corrected phase error rate, and They respectively represent the gain and bit error rate when both parties of the Z basis prepare the signal state, f is the error correction efficiency, H(x) is the Shannon entropy, and the expression is H(x)=-xlog2(x)-(1-x)log2(1-x).

[0044] The present invention also proposes a quantum key distribution system for monitoring a light source and quantitatively characterizing its defects, including a transmitting end Alice, a transmitting end Bob and an intermediate measuring end Charlie;

[0045] Both the sender Alice and the sender Bob include coherent light preparation modules, encoding and monitoring modules:

[0046] Among them, the coherent light preparation module is used to generate weak coherent pulses;

[0047] The encoding and monitoring module performs polarization modulation and intensity modulation on the pulses, monitors the modulated pulses for two-photon interference, and records the interference visibility;

[0048] The intermediate measurement module, Charlie, is used to perform interferometric measurements on the arriving weakly coherent pulses and record the responses.

[0049] Furthermore, the encoding and monitoring module includes,

[0050] Beam splitter BS1, used to split the pulse into two branches, the lower branch is used as the reference pulse;

[0051] A polarization modulation unit and an intensity modulation code are used to perform polarization modulation and intensity modulation on the upper branch pulse;

[0052] The beam splitter BS2 is used to split the polarization modulated and intensity modulated pulses in the upper branch into two parts, one part is used to transmit to the Charlie end, and the other part is used to measure the visibility of two-photon interference with the reference pulse;

[0053] An optical attenuation unit, used for attenuating the optical intensity of the lower branch pulse;

[0054] The reference pulse after the light attenuation unit is passed through the beam splitter BS3 and the pulse from the beam splitter BS2 is measured for two-photon interference visibility;

[0055] Detector D1 and detector D2 are used to detect the result of two-photon interference at beam splitter BS3.

[0056] Beneficial effects of the present invention:

[0057] (1) Quantitatively characterize the actual safety issues at the light source end: In the present invention, it is only necessary to measure the two-photon interference visibility of the reference pulse and the modulated pulse in the monitoring module, so that the defects of the actual light source can be quantitatively characterized by one parameter, and the measuring device is simple.

[0058] (2) Real-time monitoring of light source safety issues: Part of the modulated pulse that interferes with the reference pulse is used to quantitatively characterize the light source defects, and part is sent to the channel for transmission through a beam splitter at the same time, which can achieve the purpose of real-time monitoring.

[0059] (3) Improving the decoy state method and combining it with the solution: The present invention takes into account the distinguishability of the signal state and the decoy state, further modifies the decoy state method, and improves the security analysis.

[0060] (4) The solution has good generalization ability: it is applicable to protocols such as BB84, MDI-QKD and TF-QKD, and it improves security while retaining the advantages of the original protocols.

[0061] (5) Improving transmission distance and coding rate: When the present invention is combined with the MDI-QKD and TF-QKD protocols, it has a security similar to that of the device-free quantum key distribution protocol (DI-QKD), and has a long transmission distance and a high coding rate.

[0062] (6) The device of the present invention is simple, highly practical, and has extremely high adaptability and strong practicality for actual operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a schematic diagram of the system of the present invention.

[0064] Figure 2 It is a structural diagram of the encoding and monitoring module of the present invention.

[0065] Figure 3 This is a diagram showing the relationship between the coding rate and transmission distance simulated by the present invention in combination with MDI-QKD and TF-QKD.

[0066] Figure 4 It is a relationship diagram between the coding rate and the two-photon interference value simulated by combining the present invention with MDI-QKD and TF-QKD respectively. DETAILED DESCRIPTION

[0067] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. It should be pointed out that for ordinary technicians in this technical field, without departing from the technical principles of the present invention, several improvements and deformations can be made, and these improvements and deformations should also be regarded as the protection scope of the present invention.

[0068] like Figures 1 to 4 As shown, a method for quantitatively characterizing light source defects in a quantum key distribution system (QKD) is proposed, which is applicable to MDI-QKD. The method involves the transmitters Alice and Bob and the measuring end Charlie. The process includes the following steps:

[0069] (1) Alice and Bob prepare phase-randomized weak coherent light pulses respectively, and polarization encode the pulses they prepare respectively, randomly select the basis, bit value and light source intensity, and record the relevant information. Alice and Bob then send the prepared signal state to the intermediate measurement device Charlie, perform interference visibility test on the light source in real time, and record the two-photon interference visibility.

[0070] The polarization modulation options are {0°(|H>), 90°(|V>), 45°(|D>), 135°(|A>)}, where the Z basis prepares pulses with polarization information |H>, |V>, and the X basis prepares pulses with polarization information |D>, |A>. Alice and Bob respectively x and 1-p x The probability of choosing X basis and Z basis, where p x ∈[0,1]; Alice and Bob select the basis vectors and then select polarization information with equal probability to modulate the pulses. The pulses are emitted by three light sources with different intensities, p and μ 、p v With the probability of p0, a signal state pulse with intensity μ, a decoy state pulse with intensity v, and a pulse with intensity 0 are emitted.

[0071] (2) Charlie performs Bell state measurements on the signal states from Alice and Bob and retains the event of successful Bell state collapse, that is, satisfying And record the responding detector and the time when the response occurs.

[0072] (3) Repeat steps (1) and (2) multiple times. After the distribution is completed, Charlie publishes the response detector and the time when the response occurs at his effective time. Then Alice and Bob each publish the light intensity and basis vector information at the effective time. When the same basis vector is used, the original key under each basis vector is formed.

[0073] (4) Alice and Bob estimate the gain and quantum bit error rate. If the estimated gain and quantum bit error rate meet the coding requirements, the protocol continues; otherwise, the protocol is abandoned and steps (1) to (3) are repeated.

[0074] The calculation formulas for the lower bound of the single photon counting rate and the upper bound of the bit error rate are:

[0075]

[0076] Where Q xy and E xy denote the total gain and bit error rate when the pulse strength from Alice is x and the pulse strength from Bob is y, respectively, x,y∈{μ,v,0}.

[0077] The present invention takes into account the finite length effect and uses the Chernoff bound to estimate the parameter bound, wherein an upper bound and an underscore are used to represent the upper bound and the lower bound of the parameter respectively.

[0078] Classical decoy state techniques usually assume that the count rate and bit error rate of pulses are independent of intensity, so the count rate and bit error rate of pulses of different intensities are always equal. However, if Eve attacks Alice (Bob) in high dimensions and obtains relevant information, then the assumption is no longer accurate, as shown below: The state used by Eve to distinguish between the signal state and the decoy state is defined as ρ ω (λ), λ is the parameter measured by Eve, which can be time, frequency, etc., ω is the intensity of the pulse, ω∈{x,y}, x represents the intensity of the pulse from Alice, and x represents the intensity of the pulse from Bob. xy Used to represent ρ x (λ) and ρ y (λ) is the trace distance, that is:

[0079]

[0080] The following constraints can be used:

[0081]

[0082] in, It is expressed as the number of pulse photons sent by Alice is n, the number of pulse photons sent by Bob is m, and both are photon counting rates at x intensity; It is expressed as the number of pulse photons sent by Alice is n, the number of pulse photons sent by Bob is m, and both are photon counting rates at y intensity. ω Represented as its original density matrix and high-dimensional information ρ ω The tensor product of (λ), ρω It can be expressed as:

[0083]

[0084] And F(ρ x ,ρ y ) represents ρ x and ρ y Fidelity between.

[0085] F(ρ x ,ρ y ) The calculation of fidelity can be done in the following way. By defining fidelity, we can get the following formula, which establishes the connection between fidelity and two-photon interference visibility:

[0086]

[0087] in Used to represent ρ x and ρ y The similarity between HOM Represents ρ x and ρ y Two-photon interferometry visibility.

[0088] At this point, the basis vector mismatch error can be calculated by the following process:

[0089] First, the basis mismatch error △ is usually defined as:

[0090]

[0091] where F X,z is the fidelity between the X basis vectors and the Z basis vectors.

[0092] Based on the Burse angle and triangular inequality, we can get:

[0093]

[0094] HereX i and Z j Represent the auxiliary matrix of the X basis and the auxiliary matrix of the Z basis respectively, where i, j∈{0,1}, B is used to represent the X basis or the Z basis, and the construction of the auxiliary matrix refers to the work of Duplinskiy et al. in 2021 (Phys. Rev. A. 104, 012601), and we can get:

[0095]

[0096] Among them, F′ is considered to be combined with the decoy state scheme by the following method:

[0097] In the encoding and monitoring module, the state arriving through the upper branch passes through the polarization modulator and the intensity modulator, and the intensity can be modulated to μ or v, and the polarization can be modulated to one of |H>, |V>, |D>, |A>, resulting in a total of eight different modulated states, represented as: |H> μ ,|H> v ,|V> μ ,|V> v ,|D> v ,|D> v ,|A> μ ,|A> v Please note that the present invention does not consider the case where the pulse intensity is 0 and its encoding. The state reached through the lower branch is a reference signal with an intensity of μ and a polarization of |H>. In the encoding and monitoring module, there are eight possible cases for the measurement value of the two-photon interference visibility, which are represented as And the corresponding expected value is expressed as Here k∈{1, 2, ..., 8}. The present invention uses the symbol V k To represent the deviation between the eight sets of measured values ​​and the expected values. Considering that the maximum value of the two-photon visibility is 0.5 (i.e., the theoretical value of the two-photon interference visibility of two states with intensity μ and polarization |H>), the present invention uses the following formula to represent the deviation, and the minimum value is represented as V′:

[0098]

[0099] F′ represents the fidelity calculated for the minimum interference visibility V′, that is, when V HOM =V′, the fidelity is obtained according to formula (6).

[0100] Afterwards, the present invention considers the basis vector mismatch to correct the single-photon phase bit error rate as follows:

[0101] First, considering that Eve can use a lossless channel to attack, the basis mismatch is corrected as follows:

[0102]

[0103] Considering the corrected basis vector mismatch, the upper bound of the single-photon phase bit error rate is corrected to

[0104]

[0105] (5) Alice and Bob perform error correction and privacy amplification and generate the final security key:

[0106]

[0107] Among them, P μzis the probability of being prepared in the Z basis with strength μ, is the probability that a weakly coherent light source emits a single-photon pulse with intensity μ, and In particular, the lower bound of the single photon counting rate and the upper bound of the corrected phase error rate, and They represent the gain and bit error rate when both parties of the Z basis prepare the signal state, f is the error correction efficiency, H(x) is the Shannon entropy, and the expression is:

[0108] H(x)=-xlog2(x)-(1-x)log2(1-x) (13)

[0109] The parameters used are as follows:

[0110] <![CDATA[P d ]]> <![CDATA[e d ]]> η f ξ N <![CDATA[10 -9 ]]> 0.015 0.85 1.16 <![CDATA[10 -10 ]]> <![CDATA[10 14 ]]>

[0111] Among them, P d is the dark count rate of the detector, e d is the background bit error rate, η is the detector detection efficiency, f is the error correction efficiency, ξ is the error probability in the statistical fluctuation, and N is the total number of pulses sent by Alice or Bob.

[0112] like Figure 1 As shown, the present invention also includes a quantum key distribution system for monitoring an actual light source and quantitatively characterizing its defects, including a transmitting end Alice, a transmitting end Bob and an intermediate measuring end Charlie, wherein both the transmitting end Alice and the transmitting end Bob include a coherent light preparation module, an encoding and monitoring module: wherein the coherent light preparation module is used to generate weak coherent pulses; the encoding and monitoring module performs polarization modulation and intensity modulation on the pulses, monitors the modulated pulses, performs two-photon interference with the reference pulses and records the interference visibility. The intermediate measuring module Charlie is used to perform interference measurement on the arriving weak coherent pulses and record the response.

[0113] like Figure 2 As shown, the encoding and monitoring module of the present invention includes a beam splitter BS1, which is used to divide the pulse into two branches, the upper branch is encoded through a polarization modulation unit and intensity modulation, and the lower branch is used as a reference light. It interferes with the modulated pulse of the upper branch at the beam splitter BS3 at the monitoring module, and records the responses of the two-photon interference visibility detector D1 and the detector D2; the beam splitter BS2 is used to divide the modulated pulse of the upper branch into two parts, one part is used for emission, and the other part is used to interfere with the reference pulse at the monitoring module to monitor and quantitatively characterize the defects of the light source.

[0114] Figure 3 (a) is a diagram showing the relationship between the code rate and transmission distance of quantum key distribution when the present invention is combined with MDI-QKD. Figure 3(b) is a graph showing the relationship between the code rate and transmission distance of quantum key distribution when the present invention is combined with TF-QKD. It can be seen that the present invention has good performance while ensuring high security. Although the security key rate and transmission distance will decrease rapidly with the decrease of interference visibility, even when the interference visibility is a poor value of 0.47, the MDI-QKD and TF-QKD schemes still maintain good performance. Figure 4 (a) is a diagram showing the relationship between the quantum key distribution code rate and the visibility of two-photon interference in the MDI-QKD scheme of the present invention. Figure 4 (b) is a relationship diagram between the quantum key distribution code rate and the two-photon interference visibility in the TF-QKD scheme of the present invention. It can be seen that the security key rate gradually becomes smooth as the interference visibility decreases.

[0115] In summary, the technical solution of the present invention combines monitoring with coding at the light source end, and can quantitatively characterize the defects of the light source end through a measurement value, including modulation errors and errors introduced by high-dimensional information. At the same time, the present invention improves the decoy state method and combines it with MDI-QKD and TF-QKD, further improving the security of the solution. In addition, the security key generation of the present invention has nothing to do with whether the third-party measurement end is trustworthy, and the present solution is also highly secure when the measurement device is attacked by an eavesdropper. Based on the above advantages, the present invention has good performance while ensuring high security, and obtains a longer distance and a higher security code rate. At the same time, based on the above advantages, the present solution has high practicality.

[0116] The above specific embodiments further illustrate the purpose, technical solutions and beneficial results of the present invention. It should be understood that the current patent specification only introduces weak coherent state light sources and polarization coding as examples. For example, the method used in the specific embodiments of the present invention is also applicable to quantum key distribution systems based on other light sources or other coding methods, and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for characterizing light source defects in a measurement-independent quantum key distribution system, characterized in that: In the polarization-coded MDI-QKD scenario, Alice and Bob prepare phase-randomized weakly coherent optical pulses respectively, and polarization-encode their prepared pulses respectively, randomly selecting the basis, bit value, and light source intensity; The specific process of polarization encoding pulses is: The polarization modulation options are {0°(|H>), 90°(|V>), 45°(|D>), 135°(|A>)}, where the Z basis prepares pulses with polarization information |H>, |V>, and the X basis prepares pulses with polarization information |D>, |A>. Alice and Bob respectively x and 1-p x The probability of choosing X basis and Z basis, where p x ∈[0,1]; Alice and Bob select the basis vectors and then select polarization information with equal probability to modulate the pulses. The pulses are emitted by three light sources with different intensities, p and μ 、p v The probability of emitting a signal pulse with a light intensity of μ, a decoy pulse with a light intensity of v, and a pulse with a light intensity of 0 is Alice and Bob send the prepared signal state to the intermediate measurement device Charlie, monitor the interference visibility of the light source at the same time, and record the two-photon interference visibility value; Establish the fidelity F(ρ x ,ρ y ) and the connection between two-photon interference visibility; F(ρ x ,ρ y ) represents ρ x (λ) and ρ y (λ), x and y represent the pulse intensity, where x,y∈{μ,v,0}; the density matrix ρ of the light source ω Represented as its original density matrix and high-dimensional information ρ ω The tensor product of (λ), ρ ω It is expressed as: Define the state Eve uses to distinguish between signal state and decoy state as ρ ω (λ), λ is the parameter measured by Eve, which refers to time or frequency, and ω is the intensity of the pulse, where ω∈{x,y}; F(ρ x ,ρ y ) The fidelity is calculated by the following formula: in Used to represent ρ x and ρ y The similarity between HOM is x and ρ y The two-photon interference visibility, Represents the Poisson distribution satisfied by the number of pulse photons.

2. According to claim 1, a method for characterizing light source defects in a measurement-independent quantum key distribution system is characterized in that: Also includes, Charlie performs Bell state measurements on the signal states from Alice and Bob, retains the events of successful Bell state measurements, and records the responding detector and the time when the response occurs; Repeat the above state preparation and measurement process. After the distribution is completed, Charlie publishes his response detector and response occurrence time at the effective time. Then Alice and Bob each publish the light intensity and basis vector information at the effective time. When the same basis vector is used, the original key under each basis vector is formed. Alice and Bob estimate the gain and qubit error rate. If the estimated gain and qubit error rate meet the coding requirements, the protocol continues. Otherwise, abandon the agreement and repeat the above process; Alice and Bob perform error correction and privacy amplification and generate the final security key.

3. According to claim 2, a method for characterizing light source defects in a measurement-independent quantum key distribution system is characterized in that: The calculation formulas for the lower bound of the single photon count rate and the upper bound of the bit error rate in the application scenario of the adaptive decoy state are: Where Q xy and E. xy denote the total gain and bit error rate when the pulse strength from Alice is x and the pulse strength from Bob is y, respectively, x,y∈{μ,v,0}, where the upper and lower bounds of the parameters are indicated by overline and underline, respectively; D xy Used to represent ρ x (λ) and ρ y (λ) is the trace distance, that is: By the following constraints: in, It is expressed as the number of pulse photons sent by Alice is n, the number of pulse photons sent by Bob is m, and both are photon counting rates at x intensity.

4. According to claim 3, a method for characterizing light source defects in a measurement-independent quantum key distribution system is characterized in that: The basis mismatch error is calculated by the following process: First, the basis vector mismatch error △ is defined as: Where F(X,Z) is the fidelity between the X basis and the Z basis, based on the Burse angle and triangle inequality: X i and Z j Represent the auxiliary matrix of X basis and the auxiliary matrix of Z basis respectively, where i,j∈{0,1}, B is used to represent X basis or Z basis; Considering that the fidelity in the inequality is equal, and a worst case where all fidelity F′ is calculated by the minimum interference visibility V′, we get: Where F′ represents the fidelity calculated by the minimum interference visibility V′, that is, when V HOM =V′, F′ is obtained according to the fidelity calculation formula; In the encoding and monitoring module, the state arriving through the upper branch passes through the polarization modulator and the intensity modulator, and the intensity can be modulated to μ or v, and the polarization can be modulated to one of |H>, |V>, |D>, |A>, resulting in eight different modulated states, represented as: |H> μ ,|H> ν ,|V> μ ,|V> ν ,|D> μ ,|D> v ,|A> μ ,|A> v ; The state arriving through the lower branch is a reference signal with intensity μ and polarization |H>; in the encoding and monitoring module, there are eight possible cases for the measurement value of the two-photon interference visibility, which are expressed as And the corresponding expected value is expressed as Here k∈{1, 2, …, 8}; V k represents the deviation between the eight sets of measured values ​​and the expected values; considering that the maximum value of the two-photon visibility is 0.5, the following formula is used to represent the deviation, and the minimum value is represented as V′:

5. According to claim 4, a method for characterizing light source defects in a measurement-independent quantum key distribution system is characterized by: The correction process of single-photon phase bit error rate considering basis vector mismatch is: First, considering that Eve uses a lossless channel to attack, the basis mismatch is corrected to: Considering the corrected basis vector mismatch, the upper bound of the single-photon phase bit error rate is corrected to 6. The method for characterizing light source defects in a measurement-independent quantum key distribution system according to claim 5, characterized in that: The code rate formula is: Among them, P μz is the probability of being prepared in the Z basis with strength μ, is the probability that a weakly coherent light source emits a single-photon pulse with intensity μ, and In particular, the lower bound of the single photon counting rate and the upper bound of the corrected phase error rate, and They respectively represent the gain and bit error rate when both parties of the Z basis prepare the signal state, f is the error correction efficiency, H(x) is the Shannon entropy, and the expression is H(x)=-xlog2(x)-(1-x)log2(1-x).

7. A measurement-independent quantum key distribution system for monitoring a light source and quantitatively characterizing its defects, comprising a transmitting end Alice, a transmitting end Bob and an intermediate measuring end Charlie, characterized in that: The transmitting end Alice and the transmitting end Bob both include a coherent light preparation module, a coding and monitoring module: Among them, the coherent light preparation module is used to generate phase-randomized weak coherent pulses; The encoding and monitoring module performs polarization modulation and intensity modulation on the pulse, monitors the modulated pulse, performs two-photon interference with the reference pulse and records the interference visibility; The middle measurement module Charlie is used to perform interferometric measurement on the arriving weak coherent pulses and record the responses; The encoding and monitoring module includes: Beam splitter BS1, used to split the pulse into two branches, the lower branch is used as the reference pulse; A polarization modulation unit and an intensity modulation code are used to perform polarization modulation and intensity modulation on the upper branch pulse; The beam splitter BS2 is used to split the polarization modulated and intensity modulated pulses in the upper branch into two parts, one part is used to transmit to the Charlie end, and the other part is used to measure the visibility of two-photon interference with the reference pulse; An optical attenuation unit, used for attenuating the optical intensity of the lower branch pulse; The reference pulse after the light attenuation unit is passed through the beam splitter BS3 and the pulse from the beam splitter BS2 is measured for two-photon interference visibility; Detector D1 and detector D2 are used to detect the result of two-photon interference at beam splitter BS3.

Images