Quantum key distribution network authentication management system

Abstract

The invention relates to the technical field of quantum key distribution authentication, and discloses a quantum key distribution network authentication management system. The system comprises an authentication information flow capture unit, a data processing unit, a quality evaluation unit, a drift correction unit and a system iteration control unit. The authentication information flow capturing unit acquires a network original authentication data flow in real time; the authentication data processing unit adopts a quantum resonance model to dynamically decompose original data to generate a quantum characteristic mode; the authentication quality evaluation unit quantifies the purity index according to each mode, removes the mode with the lowest purity, and forms a preliminary purification authentication flow; the authentication drift correction unit scans the preliminary purification authentication flow point by point and corrects a drift point to generate an optimized authentication flow; and the system iteration controller determines whether to start a new round of processing according to the optimized authentication flow quality until the new round of processing meets the stable operation standard of the system. According to the system, accurate processing and optimization of authentication data are realized through multi-unit cooperation.

Description

Technical Field

[0001] This invention relates to the field of quantum key distribution authentication technology, specifically to a quantum key distribution network authentication management system. Background Technology

[0002] With the rapid development of quantum communication technology, quantum key distribution networks (QKD) are increasingly being used to ensure the security of information transmission. Based on the principles of quantum mechanics, QKD technology can theoretically achieve unconditionally secure key transmission. Authentication management, as a crucial component of QKD network operation, directly affects the legitimacy and reliability of key transmission within the network. However, current QKD network authentication management systems still face numerous technical bottlenecks in practical applications.

[0003] In the acquisition of authentication data streams, existing systems mostly employ periodic acquisition methods, making it difficult to capture the dynamically changing raw authentication data streams in real time. Data transmission in quantum key distribution networks is characterized by high real-time requirements and high fluctuation frequency; periodic acquisition can easily lead to the omission or lag of some critical data, resulting in subsequent processing based on incomplete or outdated data, affecting the accuracy of authentication results. In fields such as finance and government, where the security requirements of quantum communication are extremely high, even minor data omissions can trigger serious authentication security vulnerabilities, threatening the security of key transmission.

[0004] In authentication data processing, traditional systems often employ static decomposition models, which are ill-suited to the dynamic nature of data in quantum key distribution networks. Static models pre-determine fixed decomposition parameters; when network load and transmission environment change, the decomposed feature components fail to accurately reflect the true characteristics of the data, resulting in insufficient usability of the generated feature modes and interfering with subsequent quality assessments. For example, when network load suddenly increases, static models cannot adjust their decomposition strategies in a timely manner, and the decomposed feature modes may contain a large amount of redundant information, failing to effectively support subsequent quality judgments.

[0005] The certification quality assessment process lacks scientific and quantitative indicators. Existing assessment methods rely heavily on empirical judgment or simple statistical analysis, failing to establish a dedicated purity assessment system for the characteristics of quantum feature modes and thus unable to accurately identify invalid modes with low purity. This results in a significant amount of interference remaining in the purified certification stream, affecting the efficiency and effectiveness of subsequent correction work. Empirical judgment is susceptible to human factors, with different operators using different criteria, leading to inconsistent assessment results. Simple statistical analysis cannot delve into the intrinsic properties of quantum feature modes and struggles to accurately distinguish between valid and invalid modes.

[0006] In handling authentication drift points, existing systems mostly use sampling inspection to identify drift points rather than point-by-point scanning. Sampling inspection has a certain degree of randomness and is prone to missing some well-hidden drift points. These uncorrected drift points accumulate in the authentication flow, causing a continuous decline in authentication flow quality and making it difficult to meet the stability requirements of system operation. Hidden drift points may have a small impact on the authentication results in the early stages, but their impact will gradually amplify over time, eventually potentially causing the entire authentication system to fail.

[0007] Existing systems lack effective iterative control mechanisms. After completing one round of data processing, they cannot determine whether to initiate a new round of processing based on the actual quality assessment results of the optimized authentication stream. They often employ fixed processing cycles or single-processing modes. When the authentication stream quality fails to meet standards, timely secondary optimization is impossible, leading to a prolonged period of system instability and impacting the overall security of the quantum key distribution network. Fixed processing cycles cannot flexibly handle authentication streams of varying quality, while single-processing modes cannot continuously optimize substandard authentication streams, making the system ill-suited to complex and ever-changing network environments. Summary of the Invention

[0008] The purpose of this invention is to provide a quantum key distribution network authentication management system to solve the problems mentioned in the background art.

[0009] To achieve the above objectives, the present invention provides a quantum key distribution network authentication management system, the system comprising:

[0010] The authentication information flow capture unit acquires the original authentication data flow from the network in real time.

[0011] The authentication data processing unit uses a quantum resonance model to dynamically decompose the original authentication data stream and generate a set of quantum characteristic modes.

[0012] The certification quality assessment unit removes the quantum characteristic modes with the lowest purity based on the quantitative purity index of each quantum characteristic mode, forming a preliminary purification certification stream;

[0013] The authentication drift correction unit scans the initial purification authentication stream point by point, identifies and corrects authentication drift points, and generates an optimized authentication stream.

[0014] The system iteration controller, based on the quality assessment results of the optimized authentication flow, decides whether to initiate a new round of authentication information flow capture and processing until the authentication flow quality meets the system's stable operation standards.

[0015] Preferably, the steps for the authentication data processing unit to perform dynamic decomposition using a quantum resonance model include: presetting the initial decomposition scale and initial constraint weights of the quantum resonance model; inputting the original authentication data stream into the quantum resonance model to obtain an initial set of quantum feature modes; calculating the mode fidelity index between the initial set of quantum feature modes and the original authentication data stream; optimizing and iterating the initial decomposition scale and initial constraint weights based on the mode fidelity index through an adaptive parameter adjustment mechanism to obtain the optimal decomposition scale and optimal constraint weights; and finally decomposing the original authentication data stream using the quantum resonance model configured with the optimal decomposition scale and optimal constraint weights to output the target set of quantum feature modes.

[0016] Preferably, the specific steps for calculating the modal fidelity index include: dividing the original authentication data stream into multiple authentication segments according to a time window; synchronously dividing each initial quantum feature mode into modal components corresponding to the time window; calculating the quantum disorder and quantum irregularity for each authentication segment and its corresponding modal component; normalizing and weighting the quantum disorder and quantum irregularity of each authentication segment to obtain the comprehensive uncertainty measure of the authentication segment; normalizing and weighting the quantum disorder and quantum irregularity of each modal component to obtain the comprehensive uncertainty measure of the modal component; calculating the correlation strength coefficient between the original authentication data stream and each initial quantum feature mode by comparing the comprehensive uncertainty measures of the authentication segments and their corresponding modal components; and constructing the modal fidelity index based on all correlation strength coefficients.

[0017] Preferably, the steps of calculating quantum disorder and quantum irregularity include: extracting the quantum state probability distribution matrix for the certified fragment or modal component; calculating the Shannon entropy value as the quantum disorder based on the quantum state probability distribution matrix; simultaneously calculating the singular value standard deviation of the quantum state probability distribution matrix as the quantum irregularity; and using a dynamic weight allocation algorithm to perform a weighted combination of the Shannon entropy value and the singular value standard deviation.

[0018] Preferably, the execution steps of the adaptive parameter adjustment mechanism include: establishing a two-dimensional search space for the decomposition scale and constraint weights; randomly generating multiple sets of parameter configurations within the search space; calculating the modal fidelity index corresponding to each set of parameter configurations; performing multi-generation evolution operations on the parameter configurations using a quantum genetic algorithm, including selecting and retaining high-fidelity parameter configurations, performing quantum crossover mutations on the parameter configurations, and generating a new generation of parameter configuration populations; and outputting the current optimal decomposition scale and optimal constraint weights when the rate of change of the modal fidelity index is lower than the convergence threshold.

[0019] Preferably, the operation steps of the authentication quality assessment unit include: calculating the global purity score for each quantum feature mode, the global purity score being based on the average of the comprehensive uncertainty measure of the mode over all time windows; comparing the global purity scores of all quantum feature modes and identifying the quantum feature mode with the lowest score; removing the quantum feature mode with the lowest score from the set of quantum feature modes; and reconstructing all quantum feature modes before and after the removal of the mode in their original order to form a preliminary purified authentication stream.

[0020] Preferably, the execution steps of the certification drift correction unit include: setting a dynamic tolerance threshold coefficient; calculating the mean and variance of the preliminary purification certification stream; determining the effective certification range based on the mean, variance, and tolerance threshold coefficient; determining whether the value of each certification data point in the preliminary purification certification stream exceeds the effective certification range for each data point; marking data points exceeding the effective certification range as certification drift points; replacing each certification drift point with the weighted average of its preceding and following data points; and generating an optimized certification stream after replacing all certification drift points.

[0021] Preferably, the specific steps of the weighted average calculation method include: taking data points within a symmetrical window centered on the authentication drift point; calculating the distance weighting coefficient between each data point in the window and the center point; performing a weighted average calculation on the valid data points in the window based on the distance weighting coefficient; and using the weighted average to replace the original value of the authentication drift point.

[0022] Preferably, the decision-making steps of the system iterative controller include: monitoring the stability indicators of the optimized authentication stream, the stability indicators including the modal purity mean and authentication drift point density of the authentication stream; when the modal purity mean is lower than the purity threshold or the authentication drift point density is higher than the density threshold, sending a re-acquisition command to the authentication information stream capture unit; when the stability indicators detected three times consecutively all meet the preset standard, determining that the authentication stream quality meets the system stable operation standard.

[0023] Preferably, the determination steps for the system's stable operation standard include: calculating the minimum purity score of all quantum feature modes in the optimized authentication stream; detecting whether there are any unprocessed authentication drift points in the optimized authentication stream; when the minimum purity score is greater than or equal to the purity threshold and there are no unprocessed authentication drift points, the system is deemed to have met the stable operation standard; otherwise, the iterative processing flow continues.

[0024] Compared with the prior art, the beneficial effects of the present invention are:

[0025] The authentication information stream capture unit processes the raw authentication data stream in real time, closely following the dynamic changes in data transmission within the quantum key distribution network. It promptly captures each critical authentication data node, preventing processing errors caused by data acquisition delays or omissions. In quantum key distribution networks, the real-time nature of data transmission directly impacts the timeliness of authentication. Real-time captured raw data streams provide the latest and most complete data foundation for all subsequent processing stages, ensuring that each step is based on the current real-time authentication state of the network and preventing authentication misjudgments due to outdated data. Whether the network load is stable or fluctuating dramatically, real-time capture ensures that the acquired raw data is highly synchronized with the actual network conditions, providing a prerequisite for the accuracy of subsequent processing stages.

[0026] The authentication data processing unit introduces a quantum resonance model to dynamically decompose the original authentication data stream. This model can adaptively adjust the decomposition parameters according to the real-time changes in the original data, breaking the fixed parameter limitations of traditional static decomposition models. The dynamic decomposition characteristics of the quantum resonance model are highly compatible with the quantum properties of data in quantum key distribution networks, enabling more accurate mining of internal feature correlations. The generated quantum feature modes better reflect the essential attributes of the data, reducing the mixing of irrelevant or interfering features. This allows subsequent quality assessment to be based on more targeted feature modes, reducing interference factors in the assessment process and enhancing the practical value of feature modes. Even if the network transmission environment undergoes sudden changes, such as the appearance of external interference signals or transmission delay fluctuations, the quantum resonance model can still ensure that the decomposed quantum feature modes retain high accuracy and effectiveness by dynamically adjusting the decomposition parameters, providing a reliable analytical object for quality assessment.

[0027] The certification quality assessment unit uses the quantitative purity index of each quantum characteristic mode as the evaluation basis. Through clear quantitative standards, it distinguishes the purity differences between different modes, accurately identifying and removing invalid or interfering modes with the lowest purity. This purity-based quantitative assessment method is more scientific and objective than traditional assessment methods that rely on experience or simple statistics. It can eliminate characteristic components that are meaningless to the certification results to the greatest extent, significantly increasing the proportion of effective components in the initial purification certification stream. This reduces the processing burden of the subsequent certification drift correction unit, allowing the correction stage to focus more on correcting drift points in the effective certification data, thus improving overall processing efficiency. The introduction of the quantitative purity index makes the evaluation results of certification data from different batches and time periods comparable, avoiding subjective differences caused by human experience judgment, ensuring that each purification process meets a consistent standard, and improving the overall quality of the initial purification certification stream.

[0028] The authentication drift correction unit performs a point-by-point scan of the initial purified authentication stream, comprehensively covering every data point in the stream and leaving no potential authentication drift point unchecked. In quantum key distribution networks, authentication drift points may exist individually or in small, consecutive clusters, and some drift points have relatively hidden characteristics. Point-by-point scanning can meticulously examine each data node, accurately identify these hidden drift points, and perform targeted corrections. After point-by-point scanning and correction, the data consistency and stability in the optimized authentication stream are significantly improved, reducing the degradation in authentication stream quality caused by the accumulation of drift points, making the authentication stream more in line with the system's operational quality requirements. Compared to sampling inspection, point-by-point scanning eliminates the risk of omissions; even drift points with inconspicuous characteristics can be accurately identified and corrected, ensuring that the optimized authentication stream achieves a high level of data continuity and accuracy, providing a guarantee for stable system operation.

[0029] The system iterative controller can flexibly determine the processing flow based on the quality assessment results of the optimized authentication stream, forming a complete closed-loop processing mechanism. When the quality of the optimized authentication stream meets the system's stable operation standards, subsequent iterations can be stopped; if the quality has not yet met the standards, a new round of authentication information stream capture and processing is immediately initiated, ensuring that each processing step is optimized based on the previous result. This dynamic iterative control method avoids the limitations of traditional fixed-cycle or single-processing modes, allowing the system to autonomously adjust its processing strategy according to the actual authentication stream quality, continuously driving the improvement of authentication stream quality until the standards required for stable system operation are reached. During the long-term operation of the quantum key distribution network, this iterative optimization capability helps the system adapt to changes in the network environment, such as network load fluctuations and increased external interference, maintaining high authentication management quality and ensuring the authentication reliability of the quantum key distribution network under different operating states, meeting the application requirements of various scenarios with high demands for quantum communication security. Whether in daily stable operation or dealing with sudden network conditions, the iterative control mechanism ensures that the system always processes authentication data in the optimal state, maintaining high quality and stability of authentication management. Attached Figure Description

[0030] Figure 1 This is a timing diagram of the quantum key distribution network authentication management system described in this invention;

[0031] Figure 2 A flowchart illustrating the dynamic decomposition of the authentication data processing unit;

[0032] Figure 3 This is a flowchart for calculating quantum disorder and quantum irregularity. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] Please see Figure 1 The present invention provides a quantum key distribution network authentication management system, the system comprising: an authentication information flow capture unit, an authentication data processing unit, an authentication quality assessment unit, an authentication drift correction unit, and a system iteration controller.

[0035] The authentication information stream capture unit is responsible for acquiring the original authentication data stream from the network in real time. This original data stream originates from the authentication interaction process within the quantum key distribution network and contains authentication information in time-series format. The authentication data processing unit receives the original authentication data stream output by the authentication information stream capture unit and dynamically decomposes it using a quantum resonance model. The quantum resonance model is a decomposition tool based on quantum mechanics principles, capable of decomposing complex data streams into a set of quantum feature modes, each representing an inherent mode in the authentication data. The authentication quality assessment unit evaluates the set of quantum feature modes based on the quantization purity index of each quantum feature mode. The quantization purity index reflects the clarity and reliability of the mode. The authentication quality assessment unit compares the quantization purity indices of all quantum feature modes, removing the quantum feature mode with the lowest purity to form a preliminary purified authentication stream. The authentication drift correction unit scans the preliminary purified authentication stream point by point, identifying and correcting authentication drift points. Authentication drift points refer to abnormal data points that deviate from the normal range. The authentication drift correction unit generates an optimized authentication stream through calculation and replacement operations. The system iteration controller monitors the quality assessment results of the optimized authentication flow. The quality assessment results include stability indicators. The system iteration controller decides whether to start a new round of authentication information flow capture and processing based on preset standards until the authentication flow quality meets the system's stable operation standards.

[0036] Example 1: See Figure 2The authentication data processing unit of the quantum key distribution network authentication management system uses a quantum resonance model for dynamic decomposition. The implementation steps of the quantum resonance model include presetting the initial decomposition scale and initial constraint weights. The initial decomposition scale is a positive integer parameter that defines the basic time window length used by the quantum resonance model to analyze the data stream. The initial constraint weights are real-valued parameters between zero and one, used to adjust the smoothing constraint strength during the model decomposition process. The original authentication data stream is input into the quantum resonance model to obtain the initial set of quantum characteristic modes. The original authentication data stream is a binary sequence or quantum state measurement value sequence captured in real time from the quantum key distribution network interface. The quantum resonance model performs a mathematical transformation on the input sequence based on the principle of quantum resonance. The transformation process involves operator operations in Hilbert space, generating a set of mutually independent mode components in the frequency domain or feature space. These mode components constitute the initial set of quantum characteristic modes. Calculating the mode fidelity index between the initial set of quantum characteristic modes and the original authentication data stream requires dividing the original authentication data stream into multiple authentication segments according to the time window. The time window is divided in accordance with the initial decomposition scale parameter. Each authentication segment contains authentication data values ​​at consecutive time points, and the number of authentication segments depends on the ratio of the total length of the original authentication data stream to the time window size. Each initial quantum feature mode is synchronously divided into mode components corresponding to the time window. The division operation ensures that each mode component is perfectly aligned with the authentication segment in the time dimension. The mode component reflects the projection or representation of the original data on a specific feature mode within that time window.

[0037] For each certified segment and its corresponding modal component, its quantum disorder and quantum irregularity need to be calculated separately. For a certified segment, its quantum state probability distribution matrix is ​​extracted. This matrix is ​​constructed by statistically analyzing the frequency or amplitude of different quantum states within the certified segment. The rows of the matrix correspond to time points, and the columns correspond to possible quantum state basis vectors. The Shannon entropy is calculated as the quantum disorder based on the quantum state probability distribution matrix. The Shannon entropy is obtained by summing the negative probability multiplied by the logarithm of the probability, characterizing the uncertainty or information content of the quantum states within the certified segment. Simultaneously, the singular value standard deviation of the quantum state probability distribution matrix is ​​calculated as the quantum irregularity. This standard deviation is obtained by calculating the standard deviation of the singular values ​​after singular value decomposition of the matrix, reflecting the dispersion or structural perturbation of the certified segment data in the feature space. For modal components, the same quantum disorder and quantum irregularity calculation process is performed. The quantum state probability distribution matrix of each modal component is derived from its own numerical sequence. The quantum disorder and quantum irregularity of each certified segment are normalized and weighted to obtain a comprehensive uncertainty measure for the certified segment. Normalization scales both quantum disorder and quantum irregularity to the range of zero to one, eliminating the influence of dimensions. Weighted fusion uses fixed or dynamically calculated weight coefficients based on data characteristics to combine the two normalized measures into a single scalar value. A higher overall uncertainty measure value for the certified segment indicates greater instability or noise in the data segment. The overall uncertainty measure of each modal component is obtained by normalizing and weighting the quantum disorder and quantum irregularity, using the same fusion method as for the certified segment. The overall uncertainty measure of a modal component characterizes the purity or reliability of that mode within a specific time window.

[0038] The correlation strength coefficient between the original certified data stream and each initial quantum characteristic mode is calculated by comparing the comprehensive uncertainty measure of the certified fragment and its corresponding modal component. For each time window, the absolute difference or correlation coefficient between the comprehensive uncertainty measure of the certified fragment and the comprehensive uncertainty measure of the modal component is calculated. The correlation strength coefficient is defined as a function of this difference or the correlation coefficient value is used directly. The larger the correlation strength coefficient value, the more similar the certified fragment and the modal component are in terms of uncertainty characteristics, that is, the stronger the interpretability of the modal component for the original data. A modal fidelity index is constructed based on all correlation strength coefficients. The modal fidelity index is usually the average or weighted average of the correlation strength coefficients over all time windows. The modal fidelity index is a global scalar used to evaluate the overall quality of the decomposition results under the current parameter configuration of the quantum resonance model. A higher modal fidelity index means that the initial quantum characteristic mode set can better preserve the essential characteristics of the original certified data stream.

[0039] Based on the modal fidelity index, an adaptive parameter adjustment mechanism is used to iteratively optimize the initial decomposition scale and initial constraint weights. The adaptive parameter adjustment mechanism searches within a two-dimensional parameter space comprised of the decomposition scale and constraint weights, aiming to find the parameter combination that maximizes the modal fidelity index. The optimization iteration process starts with the initial parameters and continuously generates new parameter combinations and evaluates their modal fidelity index using algorithms such as gradient descent, random search, or evolutionary computation. Iteration continues until the change in the modal fidelity index is less than a preset convergence threshold or the maximum number of iterations is reached. The parameter combination obtained at this point is recorded as the optimal decomposition scale and optimal constraint weights. A quantum resonance model configured with the optimal decomposition scale and optimal constraint weights is used to perform the final decomposition of the original authentication data stream. The quantum resonance model is reinitialized using the optimized parameters. The optimal decomposition scale determines the temporal granularity of the model's data analysis, and the optimal constraint weights control the strictness of the model's fit to the data. The original authentication data stream is then fed back into the reconfigured quantum resonance model. The model performs the same mathematical transformation process as the initial decomposition, but due to parameter optimization, the modal components generated by this transformation exhibit better feature separation and noise suppression. The output is called the target quantum feature mode set. This target quantum feature mode set, as the final product of the authentication data processing unit, is passed to the authentication quality assessment unit in the system for further processing. The entire dynamic decomposition process improves the accuracy of feature extraction through parameter optimization loops, and the adaptability of the quantum resonance model ensures its effectiveness in processing authentication data streams with different characteristics.

[0040] Example 2: See Figure 3 The authentication data processing unit of the quantum key distribution network authentication management system needs to calculate quantum disorder and quantum irregularity during the dynamic decomposition process. The steps for calculating quantum disorder and quantum irregularity include extracting the quantum state probability distribution matrix for each authentication fragment or mode component. The quantum state probability distribution matrix is ​​an m-row, n-column real matrix, where m represents the number of time points contained in the authentication fragment or mode component, and n represents the number of quantum state basis vectors defined by the quantum key distribution system. Each element in the matrix represents a probability estimate of observing a specific quantum state at a specific time point. This probability estimate is obtained by statistically projecting the values ​​of the authentication fragment or mode component onto the corresponding quantum state basis vectors or by calculating the normalized frequency. The Shannon entropy value is calculated as the quantum disorder degree based on the quantum state probability distribution matrix. The calculation process of the Shannon entropy value is to calculate the probability distribution of each row vector of the matrix, that is, each time point, and calculate the Shannon entropy of the probability distribution vector at each time point. Then, the arithmetic mean of the Shannon entropy values ​​at all time points is taken. This average value is the quantum disorder degree value of the authentication segment or mode component. The larger the quantum disorder degree value, the higher the uncertainty of the data in the time dimension and the richer the information content.

[0041] Simultaneously, the standard deviation of the singular values ​​of the quantum state probability distribution matrix is ​​calculated as the quantum irregularity. The calculation of the standard deviation of the singular values ​​requires performing compact singular value decomposition on the entire quantum state probability distribution matrix. The singular value decomposition produces a set of non-negative singular values, which are arranged in descending order. The standard deviation of this set of singular values ​​is calculated, and the standard deviation value is defined as the quantum irregularity. The magnitude of the quantum irregularity value reflects the structural characteristics of the quantum state probability distribution matrix. A larger value indicates that the singular value distribution of the matrix is ​​dispersed, which means that the data contains multiple pattern components with large differences in intensity. A smaller value indicates that the singular value distribution is concentrated, which means that the data pattern is relatively simple. A dynamic weighting algorithm is used to weight and combine Shannon entropy and singular value standard deviation. The algorithm adjusts the weight coefficients of the two metrics in real time according to the data characteristics of the current certified fragment or modal component. For example, when the quantum disorder value is significantly higher than the historical average, the algorithm will give the quantum disorder a lower weight to suppress the influence of instantaneous fluctuations. When the quantum irregularity value shows continuous stability, the algorithm will give the quantum irregularity a higher weight to emphasize the structural characteristics of the data. The weighted combination results in a scalar value between zero and one. This scalar value is used as a comprehensive expression of quantum disorder and quantum irregularity for subsequent modal fidelity calculation.

[0042] The adaptive parameter adjustment mechanism involves establishing a two-dimensional search space for decomposition scales and constraint weights. The search range for the decomposition scale is defined as all positive integers from the minimum to the maximum decomposition scale. The minimum decomposition scale is determined by the minimum authentication time unit of the quantum key distribution network, and the maximum decomposition scale is determined by the system's real-time processing capability. The search range for the constraint weights is defined as the real number interval between zero and one. The search space contains all possible combinations of decomposition scales and constraint weights. Multiple sets of parameter configurations are randomly generated within the search space. Each set of parameter configurations contains a specific decomposition scale value and a specific constraint weight value. The parameter configurations are generated using a uniform random sampling method, and the initial number of parameter configurations is set to a fixed value based on computing resources. For each set of parameter configurations, the corresponding modal fidelity index is calculated. The calculation process involves decomposing the original authentication data stream using the quantum resonance model of this set of parameter configurations to obtain a set of quantum characteristic modes. Then, the modal fidelity index between the original authentication data stream and the quantum characteristic modes is calculated according to the steps described in Example 1. This calculation process requires a complete decomposition and fidelity evaluation process to be performed independently for each set of parameter configurations. The quantum genetic algorithm is used to perform multi-generation evolution operations on the parameter configuration. The quantum genetic algorithm treats each set of parameter configuration as an individual. The individual's genes are encoded by the decomposition scale value and constraint weight value. The initial population consists of multiple randomly generated sets of parameter configurations.

[0043] The quantum genetic algorithm's selection operation preserves high-fidelity parameter configurations. Based on a roulette wheel selection strategy, the probability of each parameter configuration being selected for the next generation is proportional to its modality fidelity index; the higher the modality fidelity index, the greater the probability of retention. Quantum crossover mutation is then performed on the parameter configurations. This operation randomly selects two parameter configurations from the population as parents, exchanging a portion of their genetic code to generate new offspring parameter configurations. Quantum mutation randomly changes the genetic code values ​​of the parameter configurations with a small probability. Quantum crossover mutation introduces the concept of qubit superposition, allowing parameters to explore a wider range of the search space. Finally, a new generation of parameter configurations is generated, consisting of the selected and quantum crossover mutation configurations, with the same size as the initial population. When the rate of change of the modal fidelity index is lower than the convergence threshold, the current optimal decomposition scale and optimal constraint weights are output. The convergence threshold is a preset small positive number used to judge the stability of the optimization process. The rate of change of the modal fidelity index is calculated by comparing the average modal fidelity index of several consecutive generations of population. If the absolute value of the rate of change is less than the convergence threshold, the quantum genetic algorithm is considered to have converged. At this time, the parameter configuration with the highest modal fidelity index is selected from the current population as the optimal solution. This optimal solution includes the optimal decomposition scale and optimal constraint weights. The adaptive parameter adjustment mechanism enables the parameter configuration of the quantum resonance model to adapt to different authentication data stream characteristics through iterative optimization of the quantum genetic algorithm. The global search capability of the quantum genetic algorithm reduces the risk of getting trapped in local optima. The parameter optimization process improves the accuracy and robustness of the quantum resonance model in decomposing authentication data streams in dynamic network environments. The authentication data processing unit uses the optimal decomposition scale and optimal constraint weights for final decomposition, ensuring that the generated quantum feature mode set has high quality and high fidelity.

[0044] Example 3: The authentication quality assessment unit of the quantum key distribution network authentication management system is responsible for processing the target quantum feature mode set from the authentication data processing unit. The operation steps of the authentication quality assessment unit involve calculating the global purity score for each quantum feature mode. The global purity score is calculated based on the average of the comprehensive uncertainty measure of the quantum feature mode across all time windows. The comprehensive uncertainty measure comes from the evaluation results of the mode components under each time window by the authentication data processing unit. The calculation process requires dividing the quantum feature mode into continuous segments in the time dimension, with each segment corresponding to a time window. The segmentation method is completely consistent with the time window division in the authentication data processing unit. For each quantum feature mode, all its time window segments are traversed and the comprehensive uncertainty measure value of each segment is obtained. These values ​​are then arithmetically averaged, and the average value is the global purity score of the quantum feature mode. The level of the global purity score directly reflects the stability and reliability of the quantum feature mode over the entire time span. A higher score indicates that the mode is less affected by noise and has clear features.

[0045] Comparing the global purity scores of all quantum feature modes requires juxtaposing the scores of each mode and comparing their numerical values ​​to identify the quantum feature mode with the lowest score. The scoring comparison process employs a linear scanning algorithm. The algorithm initializes a temporary variable to store the currently known minimum score and its corresponding mode identifier. The algorithm iterates through the entire target quantum feature mode set, comparing the global purity score of each mode with the value in the temporary variable. If a smaller score is found, the temporary variable is updated. After the iteration, the mode identifier stored in the temporary variable represents the quantum feature mode with the lowest global purity score. Removing the quantum feature mode with the lowest score from the set is a physical removal operation. Before the removal operation, the target quantum feature mode set contains N modes; after the removal operation, it becomes a new set containing N-1 modes. The removed quantum feature mode is no longer involved in subsequent processing.

[0046] The initial purified authentication stream is formed by reconstructing all quantum characteristic modes before and after mode removal in their original order. This reconstruction process requires superimposing and fusing the remaining quantum characteristic modes in the time dimension. The reconstruction operation can be expressed as the following formula:

[0047]

[0048] in: Indicates at a point in time The generated preliminary purification and authentication stream values; This represents the total number of quantum characteristic modes remaining after the elimination operation; Indicates the first The weight coefficients of each quantum feature mode can be set to a fixed value of 1 or adaptively assigned based on the global purity score of the mode; Indicates the first The retained quantum characteristic modes at time point The amplitude of the data. The reconstruction formula ensures a linear combination of all retained modes, maintaining the temporal continuity of the authentication flow. Both sides of the formula have the same dimensions, representing the amplitude of the authentication data.

[0049] The execution steps of the authentication drift correction unit include setting a dynamic tolerance threshold coefficient, which is a real number parameter greater than zero, and its value can be dynamically adjusted according to the network environment noise level. The mean and variance of the initial cleaned authentication stream are calculated. The mean is obtained by taking the arithmetic mean of the values ​​at all time points in the initial cleaned authentication stream, and the variance is calculated using the standard sample variance formula based on the same dataset. The effective authentication range is determined based on the mean, variance, and tolerance threshold coefficient. The effective authentication range is usually defined as a closed interval [mean - dynamic tolerance threshold coefficient × variance, mean + dynamic tolerance threshold coefficient × variance], which represents the normal fluctuation range of authentication data points. Each authentication data point in the initial cleaned authentication stream is judged to see if its value exceeds the effective authentication range. The judgment process involves comparing the value of each data point with the upper and lower bounds of the effective authentication range. Data points exceeding the effective authentication range are marked as authentication drift points. The marking operation typically creates a Boolean marker array in memory with the same length as the initial cleaned authentication stream. Each element in the array corresponds to a time point, and a true value indicates that the point is an authentication drift point. For each authentication drift point, a weighted average of its preceding and following data points is used for replacement. This replacement operation avoids using data marked as authentication drift points for calculations to prevent error propagation. After replacing all authentication drift points, an optimized authentication stream is generated. This optimized stream maintains the basic form of the initially purified authentication stream in the time series but eliminates significant abnormal fluctuations. The sequential operation of the authentication quality assessment unit and the authentication drift correction unit constitutes the core of the authentication stream purification process. The authentication quality assessment unit improves overall data quality through modality-level filtering, while the authentication drift correction unit repairs local anomalies through data point-level correction. The collaboration of these two units ensures the high reliability and stability of the authentication stream output by the quantum key distribution network authentication management system.

[0050] Example 4: The authentication drift correction unit of the quantum key distribution network authentication management system uses a weighted average calculation method when performing authentication drift point replacement operations. The specific steps of the weighted average calculation method include taking data points within a symmetrical window centered on the authentication drift point. The window size is set to a fixed odd value. For example, a window size of five indicates taking the two data points before the authentication drift point, the authentication drift point itself, and the two data points after the authentication drift point. The authentication drift point itself, since it has been marked as an outlier, does not participate in the calculation but is only used to locate the center of the window. The distance weight coefficient between each data point in the window and the center point is calculated. The distance weight coefficient is calculated based on the absolute value of the time index difference between the data point and the center point. The smaller the time index difference, the larger the corresponding distance weight coefficient. The distance weight coefficient is usually assigned using an inverse proportional function or a Gaussian decay function. A weighted average calculation is performed on the valid data points in the window according to the distance weight coefficient. Valid data points refer to data points in the window that have not been marked as authentication drift points. The weighted average calculation multiplies the value of each valid data point by its corresponding distance weight coefficient, sums the results, and then divides by the sum of all weight coefficients. The weighted average value is used to replace the original value of the authentication drift point. The replacement operation directly modifies the data value of the initial purification authentication stream at the authentication drift point position. The replaced value reflects the local trend of the effective data points within the window.

[0051] The decision-making steps of the system iterative controller include monitoring the stability metrics of the optimized authentication stream. These metrics include the mean modal purity and the authentication drift point density. The mean modal purity is derived from the average value calculated by the authentication quality assessment unit after quantum feature mode extraction and global purity scoring of the optimized authentication stream. The mean modal purity reflects the clarity of the overall features of the authentication stream. The authentication drift point density is the number of authentication drift points per unit data length. Calculating the authentication drift point density requires counting the number of remaining or newly discovered authentication drift points in the optimized authentication stream after processing by the authentication drift correction unit and dividing this number by the total number of data points in the authentication stream. When the mean modal purity is lower than the purity threshold or the authentication drift point density is higher than the density threshold (which are preset constant boundary values), a re-acquisition command is sent to the authentication information stream capture unit. This re-acquisition command is a digital control signal that triggers the authentication information stream capture unit to restart capturing the original authentication data from the network interface. When the stability indicators of three consecutive tests all meet the preset standards, the preset standards are defined as the average modal purity not being lower than the purity threshold and the authentication drift point density not being higher than the density threshold. The authentication flow quality is then judged to meet the system's stable operation standards.

[0052] The choice of window size and weighting function in the weighted average calculation process directly affects the correction effect. Different parameter configurations are suitable for different authentication data characteristics. Refer to Table 1, which shows the data point distance weighting coefficient allocation scheme within the window. This scheme adopts a symmetrical window structure with a window size of five.

[0053] Table 1: Data point distance weighting coefficients when the window size is 5

[0054] The position offset of the data point relative to the center point absolute value of time index difference Distance weight coefficient assignment The second point 2 0.1 The first point 1 0.3 Center point (certification drift point) 0 0.0 (not included in the calculation) The first point after 1 0.3 The second point 2 0.1

[0055] The distance weighting coefficients in the table are assigned based on a linear decay rule; the larger the absolute value of the time index difference, the smaller the weighting coefficient. In practical applications, the distance weighting coefficients can be generated using nonlinear functions such as Gaussian functions, based on the data noise characteristics. Gaussian functions can assign higher weights to data points closer to the center point while attenuating the influence of distant points more quickly. When calculating the weighted average, only four valid data points—the second-to-last, first-to-last, first-to-last, and second-to-last points—are used. The contribution of each point is its value multiplied by the corresponding distance weighting coefficient. The weighted average is equal to (second-to-last point value × 0.1 + first-to-last point value × 0.3 + first-to-last point value × 0.3 + second-to-last point value × 0.1) divided by the sum of the weighting coefficients (0.1 + 0.3 + 0.3 + 0.1 = 0.8). The monitoring behavior of the system iterative controller is executed periodically at fixed time intervals. During each monitoring cycle, the system iterative controller obtains the latest modal purity mean and certification drift point density values ​​from the certification quality assessment unit and the certification drift correction unit. The system iteration controller maintains a status register to record the number of times a preset standard is met consecutively. When a monitoring result meets the preset standard, the register count is incremented; when the result does not meet the standard, the count is reset to zero. When the register count reaches three, the system iteration controller generates a system stability signal and stops sending re-acquisition commands. After the authentication stream quality meets the system's stable operation standard, the quantum key distribution network authentication management system enters a steady-state operating mode. If the register count does not reach three and the monitoring result triggers the re-acquisition condition, the system iteration controller immediately sends a re-acquisition command to the authentication information stream capture unit and resets the register. The entire system then enters a new round of authentication stream purification iteration. The coordinated operation of the weighted average calculation and the system iteration control logic ensures the local accuracy of authentication drift correction and the global stability of system operation. The authentication drift correction unit repairs anomalies through a data-driven approach, and the system iteration controller controls the termination conditions of the purification process through a feedback mechanism.

[0056] Example 5: The steps for determining the stable operation standard of the quantum key distribution network authentication management system include calculating the minimum purity score of all quantum feature modes in the optimized authentication stream. The optimized authentication stream is the final output data stream processed by the authentication quality assessment unit and the authentication drift correction unit. Calculating the minimum purity score requires obtaining the global purity score of each quantum feature mode from the authentication quality assessment unit. The global purity score is the average of the comprehensive uncertainty measure of the mode over all time windows. The system iterative controller traverses the global purity score records of all quantum feature modes and finds the minimum value. The minimum purity score reflects the quality level of the weakest mode in the optimized authentication stream. Detecting whether there are unprocessed authentication drift points in the optimized authentication stream requires querying the execution log of the authentication drift correction unit. After completing the correction, the authentication drift correction unit generates an authentication drift point processing report, which clearly lists all identified and corrected authentication drift points as well as any potentially missed anomalies. The system iterative controller analyzes this report to confirm whether all data points exceeding the effective authentication range have been correctly marked and replaced. When the minimum purity score is greater than or equal to the purity threshold and there are no unprocessed authentication drift points, the purity threshold is a system-preset constant lower limit, indicating that the system has reached the standard for stable operation. The purity threshold is usually set to a specific value between zero and one, such as 0.85, which represents the lowest allowable modal quality boundary of the system. The absence of unprocessed authentication drift points means that the authentication drift correction unit has successfully processed all detected abnormal data points, and the optimized authentication stream is clean at the point level. Otherwise, the iterative processing flow continues, and the system iteration controller sends a control signal to the authentication information stream capture unit to start a new round of authentication information stream capture and processing.

[0057] Consider a concrete example: Suppose the optimized authentication stream currently being processed by the quantum key distribution network authentication management system contains three quantum feature modes with global purity scores of 0.92, 0.88, and 0.91, respectively. The system iterative controller calculates the minimum purity score to be 0.88, while the system's preset purity threshold is 0.85. The processing report provided by the authentication drift correction unit shows that five authentication drift points were detected in the optimized authentication stream of 100 data points, and all were successfully corrected; there are no unprocessed authentication drift points. In this case, the minimum purity score of 0.88 is greater than the purity threshold of 0.85, and the authentication drift points have been processed. The system iterative controller determines that the authentication stream quality meets the system's stable operation standard. In another example scenario, the optimized authentication stream contains four quantum feature modes with global purity scores of 0.90, 0.83, 0.94, and 0.89, respectively. The minimum purity score is 0.83, which is lower than the system's preset purity threshold of 0.85. Although the certification drift correction unit reports that all certification drift points have been processed, the system iteration controller still determines that the system stability standard has not been met. The system iteration controller triggers a re-acquisition command, the certification information flow capture unit begins a new round of data acquisition, and the certification data processing unit, certification quality assessment unit, and certification drift correction unit execute processing tasks sequentially. The third example involves a case where certification drift points are not fully processed. After optimization, the minimum purity score of the quantum feature modes in the certification flow is 0.87, which is higher than the purity threshold of 0.85. However, the certification drift correction unit notes in its report that one certification drift point is marked as an unprocessed certification drift point because it is located at the data flow boundary and a complete neighborhood window cannot be obtained. The system iteration controller detects the existence of this unprocessed certification drift point, determines that the system stability standard is not met, and the system continues the iterative processing flow. In subsequent iterations, the certification drift correction unit may use boundary processing strategies such as mirror expansion to correct this special point.

[0058] The determination of the system's stable operation standard is a rigorous binary decision-making process, requiring both the minimum purity score and the detection results of unprocessed authentication drift points to simultaneously meet certain conditions. During the decision-making process, the system's iterative controller accesses multiple internal status registers, which store the latest quality indicators collected from each processing unit. The decision logic is implemented through hardware comparators or software conditional statements. The comparator compares the real-time calculated minimum purity score with a purity threshold, while the conditional statements check the status flags of the authentication drift point processing reports. The output of the decision operation is a Boolean signal that controls the switching of the entire quantum key distribution network's authentication management system's operating mode. When the signal is true, the system enters a stable operating state, and the quantum key distribution process can proceed based on a high-quality authentication stream; when the signal is false, the system remains in an iterative optimization loop, continuously improving the authentication stream quality. This decision mechanism ensures that the authentication data output by the quantum key distribution network's authentication management system has high reliability and consistency.

[0059] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A quantum key distribution network authentication management system, characterized by, The system comprises: An authentication information flow capturing unit that acquires network original authentication data flow in real time; An authentication data processing unit that adopts a quantum resonance model to dynamically decompose the original authentication data flow to generate a set of quantum characteristic modes; An authentication quality evaluation unit that removes the quantum characteristic mode with the lowest purity based on the quantized purity index of each quantum characteristic mode to form a preliminary purified authentication flow; An authentication drift correction unit that performs point-by-point scanning on the preliminary purified authentication flow, identifies and corrects authentication drift points to generate an optimized authentication flow; A system iteration controller that determines whether to start a new round of authentication information flow capturing and processing process according to the quality evaluation result of the optimized authentication flow until the authentication flow quality meets the system stable operation standard.

2. The quantum key distribution network authentication management system according to claim 1, characterized by, The implementation steps of the authentication data processing unit adopting the quantum resonance model for dynamic decomposition include: presetting the initial decomposition scale and initial constraint weight of the quantum resonance model; inputting the original authentication data flow into the quantum resonance model to obtain an initial quantum characteristic mode set; calculating the mode fidelity index between the initial quantum characteristic mode set and the original authentication data flow; optimizing and iterating the initial decomposition scale and initial constraint weight based on the mode fidelity index through an adaptive parameter adjustment mechanism to obtain the optimal decomposition scale and optimal constraint weight; and finally decomposing the original authentication data flow by using the quantum resonance model configured with the optimal decomposition scale and optimal constraint weight to output a target quantum characteristic mode set.

3. The quantum key distribution network authentication management system of claim 1, wherein, The specific steps of calculating the mode fidelity index include: dividing the original authentication data flow into multiple authentication segments according to the time window; synchronously dividing each initial quantum characteristic mode into a mode component corresponding to the time window; calculating the quantum chaos degree and quantum irregularity degree of each authentication segment and corresponding mode component, respectively; performing normalized weighted fusion on the quantum chaos degree and quantum irregularity degree of each authentication segment to obtain the comprehensive uncertainty measure of the authentication segment; performing normalized weighted fusion on the quantum chaos degree and quantum irregularity degree of each mode component to obtain the comprehensive uncertainty measure of the mode component; comparing the comprehensive uncertainty measures of the authentication segment and the corresponding mode component to calculate the correlation strength coefficient of the original authentication data flow and each initial quantum characteristic mode; and constructing the mode fidelity index based on all correlation strength coefficients.

4. The quantum key distribution network authentication management system according to claim 3, wherein, The steps of calculating the quantum chaos degree and quantum irregularity degree include: for the authentication segment or mode component, extracting its quantum state probability distribution matrix; calculating the Shannon entropy value as the quantum chaos degree according to the quantum state probability distribution matrix; simultaneously calculating the singular value standard deviation of the quantum state probability distribution matrix as the quantum irregularity degree; and performing weighted combination on the Shannon entropy value and singular value standard deviation by using a dynamic weight distribution algorithm.

5. The quantum key distribution network authentication management system of claim 2, wherein, The execution steps of the adaptive parameter adjustment mechanism include: establishing a two-dimensional search space of decomposition scale and constraint weight; randomly generating multiple sets of parameter configurations in the search space; calculating the modal fidelity index corresponding to each set of parameter configurations; performing multi-generation evolution operation on the parameter configurations using quantum genetic algorithm, including selecting and retaining high-fidelity parameter configurations, quantum crossover and mutation of parameter configurations, and generating a new generation of parameter configuration population; when the change rate of the modal fidelity index is lower than the convergence threshold, output the current optimal decomposition scale and optimal constraint weight.

6. The quantum key distribution network authentication management system of claim 1, wherein, The operation steps of the authentication quality evaluation unit include: calculating the global purity score of each quantum feature mode, which is based on the average value of the comprehensive uncertainty metric of the mode in all time windows; comparing the global purity scores of all quantum feature modes to identify the quantum feature mode with the lowest score; removing the quantum feature mode with the lowest score from the quantum feature mode set; reconstructing all quantum feature modes before and after the removed mode in the original order to form a preliminary purified authentication stream.

7. The quantum key distribution network authentication management system of claim 1, wherein, The execution steps of the authentication drift correction unit include: setting a dynamic tolerance threshold coefficient; calculating the mean and variance of the preliminary purified authentication stream; determining the effective authentication range based on the mean, variance and tolerance threshold coefficient; for each authentication data point in the preliminary purified authentication stream, judging whether its value is outside the effective authentication range; marking the data points outside the effective authentication range as authentication drift points; replacing each authentication drift point with the weighted average of its adjacent data points; generating an optimized authentication stream after replacing all authentication drift points.

8. The quantum key distribution network authentication management system of claim 7, wherein, The specific steps of the weighted average calculation method include: taking the data points in the symmetric window centered on the authentication drift point; calculating the distance weight coefficient of each data point in the window and the center point; calculating the weighted average of the effective data points in the window according to the distance weight coefficient; using the weighted average value to replace the original value of the authentication drift point.

9. The quantum key distribution network authentication management system of claim 1, wherein, The decision steps of the system iterative controller include: monitoring the stability indicators of the optimized authentication stream, including the mean of the modal purity of the authentication stream and the density of the authentication drift points; sending a reacquisition instruction to the authentication information stream capture unit when the mean of the modal purity is lower than the purity threshold or the density of the authentication drift points is higher than the density threshold; determining that the authentication stream quality meets the system stable operation standard when the stability indicators detected for three consecutive times all meet the preset standard.

10. The quantum key distribution network authentication management system of claim 9, wherein, The determination steps of the system stable operation standard include: calculating the minimum purity score of all quantum feature modes of the optimized authentication stream; detecting whether there are any unprocessed authentication drift points in the optimized authentication stream; determining that the system stable operation standard is met when the minimum purity score is greater than or equal to the purity threshold and there are no unprocessed authentication drift points; otherwise, continue the iterative processing process.

Images