CN-122027126-A - Data dynamic encryption transmission method and system based on quantum key distribution

CN122027126ACN 122027126 ACN122027126 ACN 122027126ACN-122027126-A

Abstract

The application discloses a data dynamic encryption transmission method and system based on quantum key distribution, relates to the technical field of information security, and solves the technical problems that the prior art often adopts a fixed algorithm and a fixed key length, so that the security is insufficient in a high-risk scene and the performance is wasted in a low-risk scene, and the efficiency of the data dynamic encryption transmission method is low; the quantum key pool is constructed through the quantum key distribution link, the encryption strategy is dynamically generated by combining the intelligent decision model of environment perception, and data is encrypted and transmitted according to the encryption strategy, so that the bottleneck that static encryption is difficult to adapt to a complex communication environment is effectively broken through, the self-adaptive optimization of encryption strength, algorithm and key source is realized, the resource utilization efficiency, emergency response capacity and overall transmission efficiency are improved on the premise of ensuring high safety, and key technical support is provided for constructing an intelligent and elastic safety communication system.

Inventors

CUI JUAN
LIANG CHAO
Request for anonymity
Request for anonymity
Request for anonymity
Request for anonymity
DENG XIAOJUN
CHU CHENG
YANG SHAOJIE
Shen Tianmei
Wang Guohang
Request for anonymity
LIU YUE
WANG JUN
WANG LEI
ZHANG WEI
GU XIAOPENG
GUO XIAOLEI

Assignees

安徽华典大数据科技有限公司
合肥市公安局
中国电子科技集团有限公司电子科学研究院
中科智慧(苏州)科技有限公司
合肥中科智慧社区科技有限公司
合肥中科网安量子科技有限公司
合肥中科数商数据科技有限公司
中科网安(苏州)量子科技有限公司
中科数安(常州)量子科技有限公司

Dates

Publication Date: 20260512
Application Date: 20251224

Claims (10)

1. The data dynamic encryption transmission method based on quantum key distribution is characterized by comprising the following steps: Obtaining IDs of two communication parties, and constructing a quantum key distribution link according to the IDs, wherein the IDs comprise a sender and a receiver; determining a quantum key pool based on the quantum key distribution link; Acquiring data to be transmitted and environment parameters between IDs of two communication parties, wherein the environment parameters comprise link performance parameters, service characteristic parameters and security evaluation parameters; Performing data processing on the environmental parameters to obtain feature vectors, wherein the data processing comprises data cleaning and standardization operation; The method comprises the steps of inputting a feature vector into an intelligent decision model to obtain an encryption strategy instruction, wherein the encryption strategy instruction comprises an encryption algorithm type, a key length and a key source identifier, and constructing the intelligent decision model through an artificial intelligent model for evaluating the feature vector so as to determine the encryption strategy instruction with optimal data to be transmitted currently; and carrying out encryption transmission on the data to be transmitted according to the encryption strategy instruction.
2. The method for dynamically encrypting and transmitting data based on quantum key distribution according to claim 1, wherein the determining a quantum key pool based on a quantum key distribution link comprises: continuously generating a quantum key through a quantum key distribution link, wherein the quantum key is shared by IDs of two communication parties; Storing a plurality of quantum keys into a quantum key pool to obtain an initial key pool; The method comprises the steps of dynamically dividing a plurality of quantum keys in an initial key pool according to key generation time points to obtain a partition key pool, wherein the partition key pool comprises a hot area key pool, a warm area key pool and a cold area key pool, the key generation time points refer to time points when quantum keys are generated, the hot area key pool refers to quantum key compositions within N hours from a current time point in a time window, the warm area key pool refers to quantum key compositions within M hours and N hours from the history in the time window, the cold area key pool refers to all quantum key compositions before the history within M hours in the time window, M and N are integers, M > N, and M is smaller than the size of the time window; a final quantum key pool is determined based on the number of partitioned key pools.
3. The data dynamic encryption transmission method based on quantum key distribution according to claim 1, wherein the intelligent decision model is composed of an improved neural network model; the improved neural network model comprises an input layer, a full connection layer, an attention weighting layer and an output layer; The input layer is used for receiving the feature vector after data processing; The full-connection layer comprises a full-connection layer I, a full-connection layer II and a full-connection layer III; the full-connection layer I is used for combining and converting the input feature vectors to obtain initial advanced features, and the full-connection layer II is further used for combining and converting the initial advanced features extracted by the full-connection layer I to obtain refined advanced features; The attention weighting layer is used for carrying out weight distribution and feature fusion on the refined advanced features so as to obtain context features; The output layer consists of an algorithm output layer, a key length output layer and a key area output layer, wherein the input data of the output layer is a decision advanced feature, and the output result is an encryption algorithm type, a key length and a key source identifier; The input layer is connected with the full-connection layer I and the attention weighting layer respectively, the full-connection layer I is connected with the full-connection layer II, the full-connection layer II is connected with the attention weighting layer, the attention weighting layer is connected with the full-connection layer III, and the full-connection layer III is connected with the output layer.
4. The method for dynamically encrypting and transmitting data based on quantum key distribution according to claim 1, wherein the step of inputting the feature vector into the intelligent decision model to obtain the encryption policy instruction comprises the steps of: Step one, the feature vector passes through an output layer to obtain an input feature vector, wherein the input feature vector is consistent with the feature vector; step two, the input features are subjected to full connection layer one to obtain initial advanced features; step three, the initial advanced features are subjected to full connection layer two to obtain refined advanced features; step three, carrying out weighted fusion operation on the input feature vector and the refined advanced feature through an attention weighting layer to obtain a context feature; step four, obtaining a decision advanced feature through the context feature through a full connection layer III; And fifthly, parallelly outputting the encryption algorithm type, the key length and the key source identification through the algorithm output layer, the key length output layer and the key area output layer of the output layer by the decision advanced features.
5. The method for dynamically encrypting and transmitting data based on quantum key distribution according to claim 4, wherein the step of performing weighted fusion operation on the input feature vector and the refined advanced features through the attention weighting layer to obtain the context features comprises the following steps: an input feature vector STX and a refined advanced feature JGT are obtained, wherein, ; N and m are both represented as feature dimensions, and m < n; represented as a real set; The refined advanced features are converted into query vectors Q through an activation function, wherein the activation function adopts a ReLU activation function, and the following conditions are satisfied: Wherein, the method comprises the steps of, Represented as an activation function; And Are denoted as trainable parameters; , ; calculating feature attention weights by formula The formula satisfies: Wherein, i and j are respectively represented as numbers corresponding to a plurality of features in the input feature vector; Expressed as Kronecker product operator; The feature attention weight and the input feature vector are subjected to weighted fusion operation, and then feature splicing operation is carried out on the feature attention weight and the refined advanced feature to obtain fusion features The fusion characteristics satisfy: ; represented as a feature stitching operation; Represented as a trainable parameter that is to be used, ; Processing the fusion features by activating functions to obtain context features The context feature satisfies: Wherein, the method comprises the steps of, Represented as an activation function; And Are all represented as trainable parameters, an , 。
6. A data dynamic encryption transmission method based on quantum key distribution according to claim 3, wherein the intelligent decision model is constructed by: acquiring a plurality of historical feature vectors and corresponding historical encryption strategy instructions; Dividing a plurality of historical feature vectors and corresponding historical encryption strategy instructions into training data, verification data and test data, and carrying out data preprocessing on the training data, the verification data and the test data to obtain a training set, a verification set and a test set; Selecting an improved neural network model as a base model; training a basic model through a training set, and adjusting learning rate and super parameters on a verification set to obtain a pre-training model; And verifying the pre-training model on the test set to finally obtain the intelligent decision model with the input as the feature vector and the output as the encryption strategy instruction.
7. The data dynamic encryption transmission method based on quantum key distribution according to claim 6, wherein the loss function of the intelligent decision model satisfies: Wherein, the method comprises the steps of, Represented as the total loss of the intelligent decision model; denoted as a loss of classification, Represented as a regularized loss of attention, Expressed as constraint penalty term loss; And Respectively expressed as a regular term weight and a penalty term weight; And ∈(0,1); The classification loss satisfies: wherein k represents the numbers corresponding to the algorithm output layer, the key length output layer and the key area output layer in the output layer; Denoted as the sub-loss weight corresponding to the kth output layer, ∈(0,1); Denoted as the cross entropy loss for the kth output layer.
8. The quantum key distribution-based data dynamic encryption transmission method of claim 7, wherein the attention regularization loss satisfies: Wherein t is denoted as a time point number, Expressed as attention weights, and 1 and 2 are expressed as L1 regularization and L2 regularization, respectively; Items are expressed as loss of concentration sparsity; the term is expressed as a loss of attention stability.
9. The quantum key distribution-based data dynamic encryption transmission method of claim 8, wherein the constraint penalty term loss satisfies: Wherein yb is represented as a number corresponding to the training samples, YBS is represented as a total number of training samples, CF is represented as a violation constraint penalty value, I () is represented as an indicator function, C () is represented as a boolean function, Denoted as the yb-th training sample, A predictive encryption policy instruction corresponding to the yb training sample; For judging that the training sample is Whether the lower predictive encryption strategy instruction violates the core security rule or not, and the output result is True or False; the output result of (2) is 1 or 0.
10. The data dynamic encryption transmission system based on quantum key distribution is characterized by comprising a data acquisition module and a data analysis module, wherein the data acquisition module is connected with the data analysis module; The data acquisition module acquires the IDs of the two communication parties through the data acquisition equipment and constructs a quantum key distribution link according to the IDs of the two communication parties; the data analysis module comprises a key generation unit, an encryption selection unit and a data transmission unit; the key generation unit is used for determining a quantum key pool based on the quantum key distribution link; The encryption selection unit is used for carrying out data processing on the environmental parameters to obtain feature vectors, and inputting the feature vectors into the intelligent decision model to obtain encryption strategy instructions; And the data transmission unit is used for carrying out encryption transmission on the data to be transmitted according to the encryption strategy instruction.

Description

Data dynamic encryption transmission method and system based on quantum key distribution Technical Field The application belongs to the technical field of information security, and particularly relates to a data dynamic encryption transmission method and system based on quantum key distribution. Background Under the current background of rapid development of digitization and intelligence, the demand of information security systems for data security is increasingly urgent. The data comprise highly sensitive contents such as citizen identity information, case investigation data, social security dynamics and the like, and once revealed or tampered, the national security and public order are seriously threatened. Traditional encryption techniques rely primarily on mathematical complexity assumptions, with fundamental risks in the face of the potential cracking capabilities of future quantum computing. The quantum key distribution QKD is used as a novel cryptographic technology based on the quantum physical principle, can realize unconditional secure key negotiation in information theory, ensures the security by the Hessenberg uncertainty principle and the quantum unclonable theorem, and radically eliminates the possibility of eavesdropping without being perceived. The QKD is applied to dynamic encryption transmission of the security data, so that a high-security communication channel resistant to quantum attack can be constructed, real-time updating and end-to-end encryption of a secret key can be realized, active defense capacity and data integrity guarantee level of an information system in a complex network environment are effectively improved, and the QKD has great strategic significance in constructing a new-generation security and credibility intelligent police system. The traditional encryption method usually adopts a fixed algorithm and a fixed key length, and can not dynamically adjust the encryption strength according to the link quality and the service emergency degree, so that the security is insufficient in a high-risk scene and the performance is wasted in a low-risk scene, the efficiency of the data dynamic encryption transmission method is lower, and therefore, the data dynamic encryption transmission method for quantum key distribution still needs to be further improved. Disclosure of Invention The application aims to at least solve one of the technical problems in the prior art, and provides a data dynamic encryption transmission method and system based on quantum key distribution, which are used for solving the technical problems that the prior art usually adopts a fixed algorithm and a fixed key length, encryption strength cannot be dynamically adjusted according to link quality and business emergency degree, so that safety is insufficient in a high-risk scene and performance is wasted in a low-risk scene, and the efficiency of the data dynamic encryption transmission method is low. To achieve the above object, a first aspect of the present application provides a data dynamic encryption transmission method based on quantum key distribution, including: Obtaining IDs of two communication parties, and constructing a quantum key distribution link according to the IDs, wherein the IDs comprise a sender and a receiver; determining a quantum key pool based on the quantum key distribution link; Acquiring data to be transmitted and environment parameters between IDs of two communication parties, wherein the environment parameters comprise link performance parameters, service characteristic parameters and security evaluation parameters; Performing data processing on the environmental parameters to obtain feature vectors, wherein the data processing comprises data cleaning and standardization operation; The method comprises the steps of inputting a feature vector into an intelligent decision model to obtain an encryption strategy instruction, wherein the encryption strategy instruction comprises an encryption algorithm type, a key length and a key source identifier, and constructing the intelligent decision model through an artificial intelligent model for evaluating the feature vector so as to determine the encryption strategy instruction with optimal data to be transmitted currently; and carrying out encryption transmission on the data to be transmitted according to the encryption strategy instruction. According to the method, the quantum key distribution and the intelligent encryption decision mechanism based on environment perception are deeply fused, the bottleneck that the traditional static encryption strategy is difficult to adapt to complex and changeable communication environments is effectively broken through, the encryption strength, algorithm selection and dynamic collaborative optimization of key sources are realized, the resource utilization efficiency and emergency response capability are improved on the basis of ensuring high safety, the overall efficiency of dynamic encryption transmi