Polarization control method in continuous variable quantum key distribution system

Abstract

The invention discloses a polarization control method in a continuous variable quantum key distribution system, and belongs to the technical field of quantum communication of optical fiber channels. In order to solve the problem that the polarization state of an optical fiber drifts in an external field complex environment, a Bayesian parameter estimation method in deep learning is effectively integrated in an extended Kalman filter algorithm, Gaussian distribution is determined as prior distribution, and a polarization controller cascaded by four degrees of freedom has the best control effect on the polarization state of a target. A polarization state measured by a polarization analyzer is used as a feedback signal, and meanwhile, the algorithm is quickly operated on FPGA hardware, so that real-time tracking and control of the polarization state of the optical field in the optical fiber are realized. By means of the method, the problem that the polarization state of the light field drifts rapidly under the external field extreme environment can be effectively solved, and compared with an existing genetic algorithm, a simulated annealing algorithm and a particle swarm algorithm, the number of iterations is greatly reduced, and the program running speed is higher.

Description

Technical Field

[0001] The present invention belongs to the technical field of quantum communication of optical fiber channels, and in particular relates to a polarization control method in a continuous variable quantum key distribution system. Background Art

[0002] Quantum key distribution is based on the complexity of quantum physics rather than mathematical calculations. Combined with "one-time one-key", it can theoretically enable both parties in legitimate communication to share a set of unconditionally secure keys. In the face of the threat posed by quantum computing, Shor's algorithm, etc. to the security of modern cryptographic systems based on mathematical calculation complexity, the establishment of an absolutely secure quantum secure communication system is of great significance in the fields of national defense, military and finance. Quantum key distribution is mainly divided into discrete variable quantum key distribution protocol and continuous variable quantum key distribution protocol. Among them, the Gaussian modulated coherent state continuous variable quantum key distribution protocol has attracted extensive attention from researchers at home and abroad in recent years due to its advantages of easy preparation of light source, high security key rate over medium and short distances, low detection end cost, and good compatibility with existing optical fiber communication networks. In this protocol, the transmitter uses time division multiplexing and polarization multiplexing technology to transmit the signal light field and the local light field in the same optical fiber channel, and reaches the receiving end through a long-distance single-mode optical fiber in a complex external field environment. The receiving end performs polarization demultiplexing on the signal light field and the local light field through the polarization state control unit, and finally uses balanced zero-beat detection technology to detect the orthogonal components of the signal light field. Due to the influence of uncertain factors such as temperature and stress in the complex outdoor environment, single-mode optical fiber will produce a birefringence effect, that is, the circular symmetry of the original ideal optical fiber is broken, the original isotropic optical fiber becomes anisotropic, the refractive index in the x-polarization direction is different from the refractive index in the y-polarization direction, and the signal light field and the local light field polarization in the optical fiber change accordingly, affecting the balanced zero-beat detection results at the receiving end, causing the system to fail to operate normally. In order to further promote the system from the laboratory environment to the actual outdoor application, the real-time tracking and control of the polarization state of the optical field signal in the optical fiber is introduced.

[0003] The existing polarization control technology of the continuous variable quantum key distribution system (patent CN102916807A) uses a DC voltage signal as a feedback signal to adjust the squeezer of the polarization controller in turn, and repeat the adjustment process until the target value is reached. This process results in a long polarization control cycle, and it is impossible to achieve real-time tracking and control of the polarization state of the light field signal. The patent (CN109361515A) uses a simulated annealing algorithm to search for the target polarization state in real time, which can achieve polarization control in the case of pulsed light. Since the process of searching for the target polarization state is limited by the local optimal point, it is necessary to go through multiple iterations with random probability to jump out of the local optimal value to find the global optimal value. This process will cause the polarization control time to increase in extreme cases and reduce the ability of the polarization control unit to compensate for the polarization state. In view of the problem of random disturbance of polarization in the continuous variable quantum key distribution system, it is necessary to further reduce the complexity of the system or algorithm to improve the ability to quickly track and compensate for the polarization state. Summary of the invention

[0004] In order to solve the problem that the polarization state of optical fiber drifts under complex external environment, the present invention provides a polarization control method in a continuous variable quantum key distribution system.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A polarization control method in a continuous variable quantum key distribution system, the method comprising the following steps:

[0007] Step 1: Time division multiplexing and polarization multiplexing are used at the transmitter in the continuous variable quantum key distribution system to transmit the local light field and the signal light field in the same optical fiber and reach the receiver through the optical fiber channel in a long-distance and complex environment;

[0008] Step 2: The single-mode optical fiber at the receiving end is separated by a beam splitter to separate a portion of the optical field for measuring the polarization state change of the optical field in the current optical fiber, and the polarization control unit is used to perform polarization demultiplexing on the signal optical field and the local optical field;

[0009] Furthermore, in step 2, the polarization state change in the single-mode optical fiber is a continuous change, and a portion of the single-mode optical fiber light field is connected to a polarization analyzer to measure the real-time polarization state. It should be emphasized that the polarization analyzer can only measure the polarization state of the light field at the beam splitter in real time.

[0010] Furthermore, the integrating optical detector is used to convert the optical field signal received at the receiving end into the peak voltage of the output pulse in real time, and at the same time, the peak voltage is input into the polarization control unit as a feedback signal.

[0011] Furthermore, the polarization state change is represented by a point on the Poincare sphere, that is, the transformation relationship between the azimuth angle χ and the ellipticity ψ of the polarization state and the three-dimensional Cartesian coordinates (x, y, z):

[0012]

[0013] Furthermore, the polarization controller in step 3 is a polarization controller using four squeezers in cascade, and the azimuth angles are arranged in sequence of 0°, 45°, 0°, and 45°. The squeezing force is controlled by the driving voltage to change the polarization state of the optical fiber, and the extinction ratio of 30dB is used as the output objective function value.

[0014] Step 3: The polarization control unit uses the real-time polarization state measured by the polarization analyzer as a feedback signal, and uses the fast computing speed and reprogrammable characteristics of FPGA hardware, combined with an integral light detector and a polarization controller, to quickly run the extended Kalman filter algorithm incorporating the Bayesian parameter method on the FPGA hardware to track and control the polarization state;

[0015] Furthermore, the specific operations of step 3 are:

[0016] Step 3.1: Track the three-dimensional Cartesian coordinate (x, y, z) data in real time using the time update equation:

[0017] x k =Ax k-1 +Bu k

[0018] Among them, x k represents the k-th estimated polarization state, x k-1 represents the k-1th estimated value, u k is the k-th control noise term; A=(100) represents the state transfer matrix, and B represents the control matrix;

[0019] The prior and posterior covariance matrices incorporating the Bayesian parameter method during the k-th cycle are:

[0020] P k =AP k-1 A T +Q

[0021] Among them, P k represents the posterior error covariance matrix of the kth process; P k-1 represents the kth process prior error covariance matrix; Q represents the covariance matrix of the noise in the measurement process; A T represents the transpose of the state transfer matrix;

[0022] Step 3.2: Construct the state measurement equation in the x direction:

[0023] x k =x k-1 +K K (z k-Hx k-1 )

[0024] Among them, x k Represents the current polarization state estimate, x k-1 represents the k-1th estimated value, H = (1 0 0) represents the initial state transfer matrix, z k Represents the current polarization state coordinate measurement value, Kalman gain K K :

[0025] K K =P k H T (HP k-1 H T +R) -1

[0026] Where R is the measurement noise covariance matrix, which is updated in real time during the experiment;

[0027] Step 3.3: Based on the extended Kalman filter algorithm, perform first-order Taylor expansion on the three-dimensional Cartesian coordinate representation (x, y, z) of the polarization state; specifically:

[0028] Based on the relationship between the polarization state in the Poincare sphere coordinate system and the key parameters (χ, ψ) in the three-dimensional Cartesian coordinate system, the measurement state prediction equation is expressed as:

[0029]

[0030] Among them, d t Rsinχcosψ represents the control term at time t, a and b represent the predicted quantities, and w represents the real-time noise tracking term; x t represents the real-time estimated value of the polarization state in the x direction at time t after the extended Kalman filter method with Bayesian prior estimation is incorporated; t represents the estimated value of the azimuth in the spherical coordinate system at time t; ψ t It represents the estimated value of ellipticity in the spherical coordinate system at time t; the first-order Taylor expansion is obtained:

[0031] Rsinacosb+(aRcosχcosψ-bRsinχsinψ)

[0032] The gain parameter B obtained in the x direction x :

[0033]

[0034] Step 3.4: Get the gain parameter B in the y direction y :

[0035]

[0036] Get the gain parameter B in the z direction z :

[0037]

[0038] Step 3.5: At the initial moment, input the time update equation according to the measured value to generate a predicted value, substitute the predicted value into the state measurement equation, generate a voltage value to track and control the optical fiber polarization state in real time, and the target polarization state extinction ratio reaches 30dB.

[0039] Furthermore, in the measurement process of step 3.1, the Bayesian prior distribution adopts Gaussian distribution, which has the best effect on polarization state tracking.

[0040] Furthermore, during the measurement process of step 3.1, the noise Q value adopts different optimal Q values ​​for different polarization change rates.

[0041] Furthermore, the specific process of the extended Kalman filter algorithm incorporating the Bayesian parameter method is as follows: the current polarization state is analyzed in real time according to the polarization analyzer data to determine the Q value, the compensation voltage is calculated in real time by establishing a time update equation and a measurement state update equation, and whether the current voltage value needs to be updated is determined in real time based on whether the feedback signal light field extinction ratio reaches 30dB, and the polarization control unit program starts or ends synchronously with the overall operation of the system.

[0042] Among them, the Bayesian prior distribution adopts Gaussian distribution, the time update equation and the state measurement equation are executed alternately, and the measurement value at the initial moment is input into the time update equation to generate a preliminary prediction value. The prediction value enters the measurement equation for update, and then the update result is fed back to the time update equation, so that the system can reach a stable tracking state in this cycle. The polarization control unit uses the real-time coordinates of the optical fiber polarization state initially obtained by the polarization analyzer and the extended Kalman filter as the core algorithm. According to different polarization change rates, the optimal Q value is adjusted in real time to track and control the polarization state in the optical fiber in real time.

[0043] Compared with the prior art, the present invention has the following advantages:

[0044] Aiming at the problem of optical fiber polarization state drifting under complex external environment, the present invention designs a polarization tracking and control algorithm based on polarization control unit. After the long-distance optical fiber reaches the receiving end, a part of the light field is first distributed to enter the polarization analyzer. The Bayesian parameter estimation method in deep learning is integrated into the extended Kalman filter algorithm, and the Gaussian distribution is determined as the prior distribution. The polarization controller with 4 degrees of freedom cascade is used as the polarization control device, and the optimal Q value is adjusted in real time according to the different rates of change of polarization state in the optical fiber under complex environment. At the same time, the algorithm is quickly run on FPGA hardware to realize real-time tracking and control of the polarization state of the optical field in the optical fiber. Through the present invention, the problem of rapid drift of the polarization state of the light field caused by extreme external environment can be effectively solved, and the change of polarization state can be tracked and controlled in real time. Compared with the existing genetic algorithm, simulated annealing algorithm and particle swarm algorithm, the number of iterations is greatly reduced, and the program runs faster. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a logic schematic diagram of an extended Kalman filter algorithm incorporating Bayesian prior distribution in a method for controlling the polarization state of a light field in a continuous variable quantum key distribution system provided by the present invention;

[0046] Figure 2 In a method for controlling the polarization state of a light field in a continuous variable quantum key distribution system provided by the present invention, different polarization state change rates correspond to optimal Q values;

[0047] Figure 3 In a method for controlling the polarization state of a light field in a continuous variable quantum key distribution system provided by the present invention, a polarization state change rate of 500 rad / s is tracked and controlled by a polarization control unit to perform a Z-axis data distribution diagram;

[0048] Figure 4 In a method for controlling the polarization state of a light field in a continuous variable quantum key distribution system provided by the present invention, a polarization state change rate of 2.4 krad / s is tracked and controlled by a polarization control unit to perform a Z-axis data distribution diagram. DETAILED DESCRIPTION

[0049] In order to gain a deeper understanding of the present invention, we will provide a comprehensive and detailed description of the present invention. However, the present invention has multiple implementations and is not limited to the specific examples listed herein. The presentation of these examples is intended to deepen the comprehensive understanding of the disclosure of the present invention.

[0050] A polarization control method in a continuous variable quantum key distribution system, the method comprising the following steps:

[0051] Step 1: Time division multiplexing and polarization multiplexing are used at the transmitter in the continuous variable quantum key distribution system to transmit the local light field and the signal light field in the same optical fiber and reach the receiver through the optical fiber channel in a long-distance and complex environment;

[0052] Step 2: The single-mode optical fiber at the receiving end is separated by a beam splitter to separate a portion of the optical field for measuring the polarization state change of the optical field in the current optical fiber, and the polarization control unit is used to perform polarization demultiplexing on the signal optical field and the local optical field;

[0053] Furthermore, in step 2, the polarization state change in the single-mode optical fiber is a continuous change, and a portion of the single-mode optical fiber light field is connected to a polarization analyzer to measure the real-time polarization state. It should be emphasized that the polarization analyzer can only measure the polarization state of the light field at the beam splitter in real time.

[0054] Furthermore, the integrating optical detector is used to convert the optical field signal received at the receiving end into the peak voltage of the output pulse in real time, and at the same time, the peak voltage is input into the polarization control unit as a feedback signal.

[0055] Furthermore, the polarization state change is represented by a point on the Poincare sphere, that is, the transformation relationship between the azimuth angle χ and the ellipticity ψ of the polarization state and the three-dimensional Cartesian coordinates (x, y, z):

[0056]

[0057] Furthermore, the polarization controller in step 3 is a polarization controller using four squeezers in cascade, and the azimuth angles are arranged in sequence of 0°, 45°, 0°, and 45°. The squeezing force is controlled by the driving voltage to change the polarization state of the optical fiber, and the extinction ratio of 30dB is used as the output objective function value.

[0058] Step 3: The polarization control unit uses the real-time polarization state measured by the polarization analyzer as a feedback signal, and uses the fast computing speed and reprogrammable characteristics of FPGA hardware, combined with an integral light detector and a polarization controller, to quickly run the extended Kalman filter algorithm incorporating the Bayesian parameter method on the FPGA hardware to track and control the polarization state;

[0059] Furthermore, the specific operations of step 3 are:

[0060] Step 3.1: Track the three-dimensional Cartesian coordinate (x, y, z) data in real time using the time update equation:

[0061] x k =Ax k-1 +Bu k

[0062] Among them, x k represents the k-th estimated polarization state, xk-1 represents the k-1th estimated value, u k is the k-th control noise term; A=(100) represents the state transfer matrix, and B represents the control matrix;

[0063] The prior and posterior covariance matrices incorporating the Bayesian parameter method during the k-th cycle are:

[0064] P k =AP k-1 A T +Q

[0065] Among them, P k represents the posterior error covariance matrix of the kth process; P k-1 represents the kth process prior error covariance matrix; Q represents the covariance matrix of the noise in the measurement process; A T represents the transpose of the state transfer matrix;

[0066] Step 3.2: Construct the state measurement equation in the x direction:

[0067] x k =x k-1 +K K (z k -Hx k-1 )

[0068] Among them, x k Represents the current polarization state estimate, x k-1 represents the k-1th estimated value, H = (1 0 0) represents the initial state transfer matrix, z k Represents the current polarization state coordinate measurement value, Kalman gain K K :

[0069] K K =P k H T (HP k-1 H T +R) -1

[0070] Where R is the measurement noise covariance matrix, which is updated in real time during the experiment;

[0071] Step 3.3: Based on the extended Kalman filter algorithm, perform first-order Taylor expansion on the three-dimensional Cartesian coordinate representation (x, y, z) of the polarization state; specifically:

[0072] Based on the relationship between the polarization state in the Poincare sphere coordinate system and the key parameters (χ, ψ) in the three-dimensional Cartesian coordinate system, the measurement state prediction equation is expressed as:

[0073]

[0074] Among them, d t Rsinχcosψ represents the control term at time t, a and b represent the predicted quantities, and w represents the real-time noise tracking term; x t represents the real-time estimated value of the polarization state in the x direction at time t after the extended Kalman filter method with Bayesian prior estimation is incorporated; t represents the estimated value of the azimuth in the spherical coordinate system at time t; ψ t It represents the estimated value of ellipticity in the spherical coordinate system at time t; the first-order Taylor expansion is obtained:

[0075] Rsinacosb+(aRcosχcosψ-bRsinχsinψ)

[0076] The gain parameter B obtained in the x direction x :

[0077]

[0078] Step 3.4: Get the gain parameter B in the y direction y :

[0079]

[0080] Get the gain parameter B in the z direction z :

[0081]

[0082] Step 3.5: At the initial moment, input the time update equation according to the measured value to generate a predicted value, substitute the predicted value into the state measurement equation, generate a voltage value to track and control the optical fiber polarization state in real time, and the target polarization state extinction ratio reaches 30dB.

[0083] Furthermore, in the measurement process of step 3.1, the Bayesian prior distribution adopts Gaussian distribution, which has the best effect on polarization state tracking.

[0084] Furthermore, during the measurement process of step 3.1, the noise Q value adopts different optimal Q values ​​for different polarization change rates.

[0085] Furthermore, the specific process of the extended Kalman filter algorithm incorporating the Bayesian parameter method is as follows: according to the polarization analyzer data, the current polarization state is analyzed in real time to determine the Q value, the compensation voltage is calculated in real time by establishing the time update equation and the measurement state update equation, and whether the current voltage value needs to be updated is determined in real time based on whether the feedback signal light field extinction ratio reaches 30dB. The polarization control unit program starts or ends synchronously with the overall operation of the system. Figure 1 As shown, Figure 1The invention discloses an extended Kalman filter algorithm flow chart incorporating a Bayesian prior estimation method into a polarization control unit of a continuous variable quantum key distribution system.

[0086] Among them, the Bayesian prior distribution adopts Gaussian distribution, the time update equation and the state measurement equation are executed alternately, and the measurement value at the initial moment is input into the time update equation to generate a preliminary prediction value. The prediction value enters the measurement equation for update, and then the update result is fed back to the time update equation, so that the system can reach a stable tracking state in this cycle. The polarization control unit uses the real-time coordinates of the optical fiber polarization state initially obtained by the polarization analyzer and the extended Kalman filter as the core algorithm. According to different polarization change rates, the optimal Q value is adjusted in real time to track and control the polarization state in the optical fiber in real time.

[0087] Figure 2 As shown in the figure, after simulation and experimental measurement, Gaussian distribution is used as the prior distribution, and different optimal Q values ​​in the polarization control unit correspond to different incident polarization state change rates. The experimental results show that when Gaussian distribution is the prior distribution, the polarization control unit with 4 cascade degrees of freedom has the best effect on the target polarization state control.

[0088] Figure 3 When the polarization state change rate is 500 rad / s, the polarization control unit tracks and controls the polarization state. In the figure, the polarization control unit can track and control the polarization state in real time, and the target polarization state can reach 94%.

[0089] Figure 4 When the polarization state change rate increases to 2.4 krad / s, the polarization control unit tracks and controls the polarization state. The polarization control unit in the figure can effectively track and control the polarization state in real time, achieving a target polarization state reduced to 85%.

[0090] The present invention proposes a polarization control method in a continuous variable quantum key distribution system, which can track and compensate the polarization state of the light field in real time, and can effectively overcome the influence of the complex external field environment on the polarization state of the light field in the optical fiber channel. The present invention can effectively promote the practical progress of the continuous variable quantum key distribution system.

[0091] The contents not described in detail in the specification of the present invention belong to the prior art known to the professional and technical personnel in the field. Although the illustrative specific embodiments of the present invention are described above to facilitate the understanding of the present invention by the technical personnel in the field, it should be clear that the present invention is not limited to the scope of the specific embodiments. For the ordinary technical personnel in the field, as long as various changes are within the spirit and scope of the present invention defined and determined by the attached claims, these changes are obvious, and all inventions and creations using the concept of the present invention are protected.

Claims

1. A polarization control method in a continuous variable quantum key distribution system, characterized in that: The method comprises the following steps: Step 1: Time division multiplexing and polarization multiplexing are used at the transmitter in the continuous variable quantum key distribution system to transmit the local light field and the signal light field in the same optical fiber and reach the receiver through the optical fiber channel in a long-distance and complex environment; Step 2: The single-mode optical fiber at the receiving end is separated by a beam splitter to separate a portion of the optical field for measuring the polarization state change of the optical field in the current optical fiber, and the polarization control unit is used to perform polarization demultiplexing on the signal optical field and the local optical field; Step 3: The polarization control unit uses the real-time polarization state measured by the polarization analyzer as a feedback signal, and combines the integral light detector and polarization controller to quickly run the extended Kalman filter algorithm incorporating the Bayesian parameter method on the FPGA hardware to track and control the polarization state.

2. The polarization control method in a continuous variable quantum key distribution system according to claim 1, characterized in that: In step 2, the polarization state in the single-mode optical fiber changes continuously, and a portion of the single-mode optical fiber light field is connected to a polarization analyzer to measure the real-time polarization state. The polarization analyzer can only measure the polarization state of the light field at the beam splitter in real time.

3. The polarization control method in a continuous variable quantum key distribution system according to claim 2, characterized in that: By using the integral optical detector, the optical field signal received at the receiving end is converted into the peak voltage of the output pulse in real time, and is input into the polarization control unit as a feedback signal.

4. The polarization control method in a continuous variable quantum key distribution system according to claim 3, characterized in that: The change in polarization state is represented by a point on the Poincare sphere, which represents the transformation relationship between the azimuth angle χ and the ellipticity ψ of the polarization state and the three-dimensional Cartesian coordinates (x, y, z):

5. The polarization control method in a continuous variable quantum key distribution system according to claim 4, characterized in that: The polarization controller in step 3 is a polarization controller using four squeezers in cascade, and the azimuth angles are arranged in sequence as 0°, 45°, 0°, and 45°. The squeezing force is controlled by the driving voltage to change the polarization state of the optical fiber, and the extinction ratio of 30dB is used as the output objective function value.

6. The polarization control method in a continuous variable quantum key distribution system according to claim 5, characterized in that: Step 3: The polarization control unit uses the real-time polarization state measured by the polarization analyzer as a feedback signal, combines the integral light detector and the polarization controller, and quickly runs the extended Kalman filter algorithm incorporating the Bayesian parameter method on the FPGA hardware to track and control the polarization state; the specific operations are: Step 3.1: Track the three-dimensional Cartesian coordinate (x, y, z) data in real time using the time update equation: x k =Ax k-1 +Bu k Among them, x k represents the k-th estimated polarization state, x k-1 represents the k-1th estimated value, u k is the k-th control noise term; A=(100) represents the state transfer matrix, and B represents the control matrix; The prior and posterior covariance matrices incorporating the Bayesian parameter method during the k-th cycle are: P k =AP k-1 From T +Q Among them, P k represents the posterior error covariance matrix of the kth process; P k-1 represents the kth process prior error covariance matrix; Q represents the covariance matrix of the noise in the measurement process; A T represents the transpose of the state transfer matrix; Step 3.2: Construct the state measurement equation in the x direction: x k =x k-1 +K K (z k -Hx k-1 ) Among them, x k Represents the current polarization state estimate, x k-1 represents the k-1th estimated value, H = (100) represents the initial state transfer matrix, z k Represents the current polarization state coordinate measurement value, Kalman gain K K : K K =P k H T (HP k-1 H T +R) -1 Where R is the measurement noise covariance matrix, which is updated in real time during the experiment; Step 3.3: Based on the extended Kalman filter algorithm, perform first-order Taylor expansion on the three-dimensional Cartesian coordinate representation (x, y, z) of the polarization state; specifically: Based on the relationship between the polarization state in the Poincare sphere coordinate system and the key parameters (χ, ψ) in the three-dimensional Cartesian coordinate system, the measurement state prediction equation is expressed as: Among them, d t Rsinχcosψ represents the control term at time t, a and b represent the predicted quantities, and w represents the real-time noise tracking term; x t represents the real-time estimated value of the polarization state in the x direction at time t after the extended Kalman filter method with Bayesian prior estimation is incorporated; t represents the estimated value of the azimuth in the spherical coordinate system at time t; ψ t It represents the estimated value of ellipticity in the spherical coordinate system at time t; the first-order Taylor expansion is obtained: Rsinacosb+(aRcosχcosψ-bRsinχsinψ) The gain parameter B obtained in the x direction x : Step 3.4: Get the gain parameter B in the y direction y : Get the gain parameter B in the z direction z : Step 3.5: At the initial moment, input the time update equation according to the measured value to generate a predicted value, substitute the predicted value into the state measurement equation, generate a voltage value to track and control the optical fiber polarization state in real time, and the target polarization state extinction ratio reaches 30dB.

7. The polarization control method in a continuous variable quantum key distribution system according to claim 6, characterized in that: In the measurement process of step 3.1, the Bayesian prior distribution adopts Gaussian distribution.

8. The polarization control method in a continuous variable quantum key distribution system according to claim 7, characterized in that: During the measurement process of step 3.1, the noise Q value adopts different optimal Q values ​​for different polarization change rates.

9. According to the polarization control method in a continuous variable quantum key distribution system according to claim 7, it is characterized in that the specific process of the extended Kalman filter algorithm incorporating the Bayesian parameter method is: according to the polarization analyzer data, the current polarization state is analyzed in real time to determine the Q value, the compensation voltage is calculated in real time by establishing a time update equation and a measurement state update equation, and whether the current voltage value needs to be updated is determined in real time based on whether the light field extinction ratio of the feedback signal reaches 30dB as the standard, and the polarization control unit program starts or ends synchronously with the overall operation of the system.

Images