Credit and credential data transmission secrecy method based on self-adaptive dynamic authority control

Abstract

The invention discloses a credential data transmission secrecy method based on self-adaptive dynamic authority control. The method comprises the following steps: S1, constructing a topological structure of a data transmission network, and outputting a preliminary scheme of a dependency relationship and an encryption strategy; s2, generating a dynamically adjusted encryption and authority control strategy combination through a causal inference module; s3, the encryption and authority control strategy combination is input into a perturbation analysis module, small-range strategy perturbation is carried out, and an optimized strategy result is output; s4, inputting an optimized strategy result into a space-time adaptive prediction model based on meta-learning, and obtaining a foresight optimization strategy; s5, inputting the prospective optimization strategy into an adaptive feedback control module, and outputting a dynamic optimization real-time strategy; s6, inputting the dynamic optimization real-time strategy to a network flow state real-time monitoring module, and optimizing an encryption and authority control strategy; and S7, forming closed-loop adaptive regulation and control. According to the method, dynamic encryption and authority control are realized by adopting network topology, causal inference and meta-learning.

Description

Technical Field

[0001] This invention relates to the field of network security technology, and in particular to a method for ensuring the confidentiality of data transmission in the context of domestically developed information technology based on adaptive dynamic access control. Background Technology

[0002] With the rapid development of information technology, the amount of data transmitted over networks is growing exponentially, making the security and confidentiality of data transmission increasingly prominent. In the context of information technology application innovation, information security is particularly important, especially in sensitive sectors such as government, finance, and healthcare, where the demands for data transmission security and privacy protection are even more stringent. However, traditional encryption and access control methods are gradually revealing their limitations in dealing with modern, complex network environments. Existing network security mechanisms typically employ static encryption and access control strategies, relying on pre-defined encryption algorithms and access configurations. While this provides basic security to a certain extent, it lacks flexibility and dynamic adjustment capabilities in the face of constantly changing network conditions and diverse security threats.

[0003] Existing encryption methods primarily rely on symmetric and asymmetric encryption techniques, such as AES and RSA. These techniques encrypt data using fixed keys, ensuring data confidentiality. However, these techniques exhibit significant performance bottlenecks when faced with large-scale data transmission or frequent network state changes. On one hand, fixed-key encryption mechanisms are vulnerable to malicious attacks during transmission; once the key is cracked, the entire transmitted data is at risk of leakage. On the other hand, existing encryption algorithms typically require substantial computational resources, making it difficult to guarantee data transmission efficiency in network environments with high real-time requirements. Furthermore, these static encryption techniques fail to effectively address the complex dynamic changes in modern networks, such as network congestion, changes in inter-node communication, packet loss, and latency fluctuations, resulting in a lack of flexibility in adapting encryption strategies to these changes.

[0004] In terms of access control, traditional access management mechanisms are typically based on fixed user roles and access control policies. This static access configuration model is prone to problems such as permission abuse or over-allocation in practical applications. For example, in a distributed network environment, permission abuse may grant low-level users unnecessary privileges, leading to the risk of data leakage. Furthermore, access management lacks real-time linkage with network conditions and fails to dynamically adjust access configurations according to changes in the network environment. Existing access control mechanisms rely on predefined rules and policies and cannot intelligently adjust based on node status, traffic changes, latency fluctuations, and other factors in the network, making them susceptible to security vulnerabilities.

[0005] On the other hand, existing encryption and access control technologies suffer from low resource utilization efficiency. In complex network environments, static encryption and access control strategies typically consume significant amounts of computing and storage resources, especially during large-scale data transmission, where this resource waste is even more pronounced. Furthermore, static encryption strategies cannot be dynamically optimized based on actual transmission demands and resource availability, making it difficult for the system to effectively allocate resources when handling large-scale concurrent requests, leading to a decline in system performance.

[0006] To address the shortcomings of existing technologies, researchers have proposed security strategies based on adaptive control and dynamic adjustment in recent years. For example, some systems attempt to introduce encryption mechanisms based on behavior monitoring, dynamically adjusting encryption strategies by analyzing user behavior patterns. However, this approach remains relatively inflexible and slow to respond to complex and ever-changing network conditions and potential threats. Meanwhile, some systems attempt to optimize encryption and access control strategies using machine learning techniques, but these solutions often face problems such as insufficient training data and poor generalization ability, making them difficult to widely implement in practical applications.

[0007] Furthermore, the multi-layered and multi-node characteristics of modern networks make the dependencies, communication paths, and traffic patterns between nodes more complex during data transmission. Traditional encryption and access control methods fail to fully consider the correlations and dependencies between nodes in the network, resulting in insufficient coverage of security policies and an inability to accurately address potential security risks. At the same time, existing encryption and access control technologies typically operate independently, lacking organic integration with network topology and transmission characteristics, making it difficult to provide comprehensive and dynamic protection.

[0008] Therefore, how to provide a secure data transmission method for domestically developed information technology based on adaptive dynamic access control is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0009] One objective of this invention is to propose a secure data transmission method for domestically developed information technology applications based on adaptive dynamic access control. This invention employs complex network topology analysis, causal inference, perturbation analysis, and meta-learning techniques to dynamically adjust encryption and access control strategies, adaptively optimizing them in response to real-time changes in the transmission network. By analyzing the dependencies and transmission characteristics between network nodes, it predicts future transmission demands and optimizes strategies in advance, ensuring efficient and secure data transmission even in complex environments. This method boasts advantages such as high security, high resource utilization, and strong adaptability, and can monitor and respond quickly to potential threats in real time, ensuring the stability and security of data transmission.

[0010] A method for ensuring the confidentiality of data transmission in the context of domestic IT innovation based on adaptive dynamic access control, according to an embodiment of the present invention, includes the following steps: S1. Through the complex network topology analysis module, construct the topology of the data transmission network, analyze the association and dependency relationships between network nodes, generate a topology diagram of the relationship between encryption strategies and access control, and identify and output the potential dependency relationships between network nodes and the preliminary scheme of encryption strategies. S2. Based on the obtained dependencies and preliminary scheme, the causal relationship between different encryption strategies and access control strategies is analyzed through the causal inference module. The effect of different strategy combinations in the current transmission environment is inferred, and a dynamically adjusted combination of encryption and access control strategies is generated. S3. Input the encryption and access control policies into the perturbation analysis module to perform small-scale policy perturbation, evaluate the impact of policy perturbation on system security, encryption strength, access granularity and resource utilization, adjust policy parameters based on system feedback, and output optimized policy results. S4. Input the optimized strategy results into the spatiotemporal adaptive prediction model based on meta-learning, analyze the spatiotemporal characteristics and environmental state changes in historical data transmission, quickly adjust the model parameters through the adaptive training mechanism, predict the future data transmission needs and potential threats, optimize encryption and access control strategies in advance, and obtain a forward-looking optimization strategy. S5. Input the forward-looking optimization strategy into the adaptive feedback control module, monitor the changes in the data flow state during data transmission in real time, adjust the encryption and access control strategies in combination with real-time feedback information, and output the dynamically optimized real-time strategy. S6. Input the dynamic optimization real-time strategy generated by the adaptive feedback control module into the network flow real-time monitoring module to continuously monitor and analyze the data flow, detect potential threats and anomalies in real time, and respond quickly and further optimize encryption and access control strategies based on traffic characteristics. S7. The analysis results and feedback control information of the real-time network flow monitoring module are re-inputted into the complex network topology analysis module and the causal inference module to form a closed-loop adaptive control.

[0011] Optionally, S1 specifically includes: S11. By acquiring the basic information of each network node in the data transmission network, including the unique identifier of the network node, geographical location, connection strength, data transmission rate, communication latency and historical communication records, a network node information matrix is ​​constructed. S12. Based on the node information matrix, construct the topology of the data transmission network to form an undirected weighted graph. ,in Represents the set of nodes in the network. Indicates the connection between network nodes. Represent the weight matrix and calculate the weights. : ; in, Represents network nodes and network nodes Real-time data transmission rate between them Represents network nodes and network nodes Communication delay between them Represents network nodes and network nodes Historical data transmission volume between them Represents network nodes and network nodes The change in angle between them , and Indicates the weighting coefficient. , , and Indicates dynamically adjusted indices and weighting coefficients; S13. By analyzing the undirected weighted graph To analyze the relationships between network nodes, a centrality analysis algorithm is used to calculate the degree centrality of each network node. Betweenness centrality and proximity centrality : Degree centrality This reflects the number of network nodes and directly connected network nodes: ; in, Indicates the total number of network nodes. Represents network nodes and network nodes Does a connection exist? Betweenness centrality Used to evaluate the mediating role of network nodes in the shortest path: ; in, Represents network nodes To network node The number of shortest paths between them. Indicates passing through network nodes The number of shortest paths; The proximity centrality This represents the average shortest path distance from a network node to other network nodes. ; in, Represents network nodes To network node The shortest path distance between them; S14. Based on the centrality analysis results, calculate the dependencies between network nodes and construct a dependency matrix. : ; in, , and This represents the weight coefficients used to calculate dependencies. and This represents the centrality adjustment factor. Indicates the attenuation coefficient. Represents network nodes betweenness centrality, Represents network nodes Proximity centrality Represents network nodes Degree centrality, Represents network nodes Degree centrality, Represents an exponential function; S15. Based on the dependency matrix Generate a preliminary encryption strategy allocation scheme, form a mapping diagram between encryption strategies and dependencies, use the preliminary scheme to identify the encryption level required for each network node, and output the preliminary encryption strategy allocation results; S16. The final output includes a preliminary scheme for the potential dependencies between network nodes and encryption strategies.

[0012] Optionally, S2 specifically includes: S21. Based on the potential dependencies and preliminary encryption strategy schemes output in step S16, construct the encryption strategy matrix. Access control policy matrix ,in Indicates the number of encryption policies. Indicates the number of access control policies; S22, Define the strategy combination matrix ,in Indicate encryption strategy and access control policies Based on real-time and historical data from the transmission environment, a causal inference model is established to describe the multiple impacts of each strategy combination on system performance. ; in, This indicates the impact of the current strategy combination on the system's target performance. Indicates the bias term. , and Indicate encryption strategy The influence coefficients of the linear, square, and logarithmic terms of the system. , and Indicates access control policy The influence coefficients on the linear, cubic, and square root terms of the system. This represents the interaction coefficient between the encryption policy and the access control policy. Indicates the noise term; S23. Train the causal inference model, optimize the model parameters using maximum likelihood estimation and gradient descent, and calculate the strategy combination based on the training results. Causal effect: ; in, Indicates strategy combination The causal effect; S24, Based on causal effects Rank all strategy combinations and select the combination with the highest causal effect score; S25. Based on the causal effect analysis results, dynamically adjust the strategy combination, and generate a dynamically adjusted encryption and access control strategy combination by combining changes in the transmission environment and real-time system monitoring data.

[0013] Optionally, S3 specifically includes: S31, Based on strategy combination Generate perturbation matrix ,in Indicates the combination of strategies The slight perturbation; S32. For the perturbation matrix Define the strategy combination after perturbation. : ; in, Indicates the disturbance attenuation coefficient. The norm of the strategy combination This represents the weights used to control perturbation noise. This represents the relative angular change between the security policy and the access control policy; S33, By combining strategies Perform analysis and calculate the disturbance effect:

[0014] ; in, This represents the second-order change in system performance after a disturbance. This indicates the impact of the current strategy combination on the system's target performance. Indicates the number of encryption policies. Indicates the number of access control policies. The adjustment coefficient representing the interaction effect between strategies. Indicates the encryption strategy. Indicates the access control policy; S34. Based on the calculation results of the perturbation effect, rank and optimize the strategy combination, giving priority to the perturbation effect. The minimum combination of strategies; S35. Generate a new strategy combination matrix based on the optimization results. And combine the encryption and access control strategies optimized by perturbation analysis into a matrix. Output.

[0015] Optionally, S4 specifically includes: S41. Input the optimized strategy results into the spatiotemporal adaptive prediction model based on meta-learning. By conducting in-depth analysis of the spatiotemporal characteristics in historical data transmission, including data traffic, network latency, node movement and environmental state change factors, construct a spatiotemporal feature matrix and extract the long-term and short-term dependencies between nodes. S42. Based on the meta-learning model, it is quickly trained using a small amount of historical data, extracts transmission patterns from different network environments, quickly adjusts parameters, and generalizes in different transmission scenarios. S43. The meta-learning model combines the dynamic changes of spatiotemporal features to analyze the changing trends of future transmission demands. By adaptively adjusting model parameters, it predicts possible future network traffic fluctuations and potential security threats. ; in, express Model parameters at time 10:00 express Model parameters at time 10:00 Indicates the learning rate. Represents model parameters gradient, The attenuation factor representing the spatiotemporal characteristics, Represents an exponential function. Indicates spatiotemporal characteristics, This indicates the impact of the current strategy combination on the system's target performance. Indicates the number of encryption policies. Indicates the number of access control policies; S44. The meta-learning model monitors the real-time environment and combines transmission history and current characteristics to make forward-looking predictions of future data transmission needs, and plans optimized encryption and access control strategies in advance. S45. Through an adaptive training mechanism, the meta-learning model is continuously updated to output a forward-looking and optimized combination of encryption and access control strategies. S46. The final prospective optimization strategy is obtained and continuously optimized in real-time transmission based on the results of the meta-learning prediction model.

[0016] Optionally, S5 specifically includes: S51. Input the forward-looking optimization strategy into the adaptive feedback control module to monitor the flow state changes during data transmission in real time. The flow state changes include sudden increases in traffic, packet loss, network latency fluctuations, and abnormal communication patterns. Establish a high-dimensional flow state matrix. ,in Indicates time At that time, network nodes With network nodes Between Dynamic changes in class transmission characteristics; S52. Combining Bayesian inference and spatiotemporal dependency models, construct an anomaly prediction model and generate an anomaly prediction matrix. ,in Indicates time At that time, network nodes With network nodes The probability of anomalies that may occur between them is used to identify abnormal behavior of network nodes in the early stage through Bayesian inference and spatiotemporal dependency model, and to provide early warning of potential security threats. S53. Define a feedback function based on adaptive weighting. By combining the flow matrix and anomaly prediction matrix The information is used to perform a performance evaluation of the current encryption and access control policies. ; in, Represents network nodes With network nodes The transmission status between them This represents the feedback adjustment coefficient. Indicates the first Weighting coefficients for flow-like characteristics, A dynamic weighting index representing flow regime characteristics. Indicates the number of encryption policies. Indicates the number of access control policies; S54. Based on the results calculated by the feedback function, the combination of encryption strategy and access control strategy is adjusted in real time to generate a dynamically optimized strategy matrix. : S55, Based on the adjusted strategy matrix Continuous monitoring of flow regime changes, combined with the output of anomaly prediction models, allows for continuous optimization of strategy combinations.

[0017] The beneficial effects of this invention are: First, the data transmission network structure constructed through complex network topology analysis in this invention not only identifies the relationships and dependencies between network nodes, but also generates personalized encryption strategies and access control schemes based on the different weights and centrality information of the nodes. This method can automatically identify key nodes requiring enhanced encryption and access control during data transmission, avoiding the resource waste and performance bottlenecks caused by uniform encryption or access control configuration across the entire network in traditional technologies. Through this dynamic adjustment based on topology, this invention can maximize transmission efficiency and resource utilization while ensuring data security.

[0018] Secondly, this invention introduces a causal inference module, enabling it to not only rely on data in the current transmission environment when adjusting encryption and access control policies, but also predict the impact of different policy combinations on system performance by analyzing the causal relationship between encryption and access control policies. This causal inference mechanism allows the system to accurately adjust policy combinations when dealing with complex transmission requirements and potential network threats, ensuring the system always remains in an optimal state. Through this inference and optimization, this invention can effectively cope with various complex factors in different transmission scenarios, giving the system high adaptability and foresight.

[0019] Furthermore, the perturbation analysis module further enhances the accuracy of policy adjustment. By performing fine-grained optimization of the current policy combination through small-scale perturbations, this invention not only dynamically adjusts the policy but also ensures that the policy adjustment process does not cause fluctuations or instability in system performance. This meticulous adjustment method, combined with the real-time monitoring function of the adaptive feedback control module, can respond immediately when any anomalies are detected during transmission. Especially when dealing with sudden network security threats or transmission anomalies, the adaptive feedback mechanism of this invention can quickly adjust the policy to ensure that system security is not affected.

[0020] Finally, by introducing a meta-learning model, this invention can quickly learn and adapt to the current network state through training with a small amount of historical data when facing complex and ever-changing network transmission environments. This meta-learning mechanism enables the system to quickly adjust model parameters and predict potential threats and demands in future transmissions when the transmission environment changes. The application of meta-learning combined with a spatiotemporal adaptive prediction model allows the system to optimize encryption and access control strategies in advance in both short-term and long-term dimensions, ensuring consistently high protection effectiveness in complex network environments. Attached Figure Description

[0021] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0022] Figure 1 This is a flowchart of a data transmission confidentiality method based on adaptive dynamic access control proposed in this invention; Figure 2 This is a schematic diagram of the complex network topology analysis module structure of a domestically developed data transmission confidentiality method based on adaptive dynamic access control proposed in this invention. Figure 3 This is a flowchart of the causal inference module of a domestically developed data transmission confidentiality method based on adaptive dynamic access control proposed in this invention. Detailed Implementation

[0023] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0024] refer to Figure 1-3 A method for ensuring the confidentiality of data transmission in the context of domestic IT innovation based on adaptive dynamic access control includes the following steps: S1. Through the complex network topology analysis module, construct the topology of the data transmission network, analyze the association and dependency relationships between network nodes, generate a topology diagram of the relationship between encryption strategies and access control, and identify and output the potential dependency relationships between network nodes and the preliminary scheme of encryption strategies. S2. Based on the obtained dependencies and preliminary scheme, the causal relationship between different encryption strategies and access control strategies is analyzed through the causal inference module. The effect of different strategy combinations in the current transmission environment is inferred, and a dynamically adjusted combination of encryption and access control strategies is generated. S3. Input the encryption and access control policies into the perturbation analysis module to perform small-scale policy perturbation, evaluate the impact of policy perturbation on system security, encryption strength, access granularity and resource utilization, adjust policy parameters based on system feedback, and output optimized policy results. S4. Input the optimized strategy results into the spatiotemporal adaptive prediction model based on meta-learning, analyze the spatiotemporal characteristics and environmental state changes in historical data transmission, quickly adjust the model parameters through the adaptive training mechanism, predict the future data transmission needs and potential threats, optimize encryption and access control strategies in advance, and obtain a forward-looking optimization strategy. S5. Input the forward-looking optimization strategy into the adaptive feedback control module, monitor the changes in the data flow state during data transmission in real time, adjust the encryption and access control strategies in combination with real-time feedback information, and output the dynamically optimized real-time strategy. S6. Input the dynamic optimization real-time strategy generated by the adaptive feedback control module into the network flow real-time monitoring module to continuously monitor and analyze the data flow, detect potential threats and anomalies in real time, and respond quickly and further optimize encryption and access control strategies based on traffic characteristics. S7. The analysis results and feedback control information of the real-time network flow monitoring module are re-inputted into the complex network topology analysis module and the causal inference module to form a closed-loop adaptive control.

[0025] In this embodiment, S1 specifically includes: S11. By acquiring the basic information of each network node in the data transmission network, including the unique identifier of the network node, geographical location, connection strength, data transmission rate, communication latency and historical communication records, a network node information matrix is ​​constructed. S12. Based on the node information matrix, construct the topology of the data transmission network to form an undirected weighted graph. ,in Represents the set of nodes in the network. Indicates the connection between network nodes. Represent the weight matrix and calculate the weights. : ; in, Represents network nodes and network nodes Real-time data transmission rate between them Represents network nodes and network nodes Communication delay between them Represents network nodes and network nodes Historical data transmission volume between them Represents network nodes and network nodes The change in angle between them , and Indicates the weighting coefficient. , , and Indicates dynamically adjusted indices and weighting coefficients; S13. By analyzing the undirected weighted graph To analyze the relationships between network nodes, a centrality analysis algorithm is used to calculate the degree centrality of each network node. Betweenness centrality and proximity centrality : Degree centrality This reflects the number of network nodes and directly connected network nodes: ; in, Indicates the total number of network nodes. Represents network nodes and network nodes Does a connection exist? Betweenness centrality Used to evaluate the mediating role of network nodes in the shortest path: ; in, Represents network nodes To network node The number of shortest paths between them. Indicates passing through network nodes The number of shortest paths; The proximity centrality This represents the average shortest path distance from a network node to other network nodes. ; in, Represents network nodes To network node The shortest path distance between them; S14. Based on the centrality analysis results, calculate the dependencies between network nodes and construct a dependency matrix. : ; in, , and This represents the weight coefficients used to calculate dependencies. and This represents the centrality adjustment factor. Indicates the attenuation coefficient. Represents network nodes betweenness centrality, Represents network nodes Proximity centrality Represents network nodes Degree centrality, Represents network nodes Degree centrality, Represents an exponential function; S15. Based on the dependency matrix Generate a preliminary encryption strategy allocation scheme, form a mapping diagram between encryption strategies and dependencies, use the preliminary scheme to identify the encryption level required for each network node, and output the preliminary encryption strategy allocation results; S16. The final output includes a preliminary scheme for the potential dependencies between network nodes and encryption strategies.

[0026] In this embodiment, S2 specifically includes: S21. Based on the potential dependencies and preliminary encryption strategy schemes output in step S16, construct the encryption strategy matrix. Access control policy matrix ,in Indicates the number of encryption policies. Indicates the number of access control policies; S22, Define the strategy combination matrix ,in Indicate encryption strategy and access control policies Based on real-time and historical data from the transmission environment, a causal inference model is established to describe the multiple impacts of each strategy combination on system performance. ; in, This indicates the impact of the current strategy combination on the system's target performance. Indicates the bias term. , and Indicate encryption strategy The influence coefficients of the linear, square, and logarithmic terms of the system. , and Indicates access control policy The influence coefficients on the linear, cubic, and square root terms of the system. This represents the interaction coefficient between the encryption policy and the access control policy. Indicates the noise term; S23. Train the causal inference model, optimize the model parameters using maximum likelihood estimation and gradient descent, and calculate the strategy combination based on the training results. Causal effect: ; in, Indicates strategy combination The causal effect; S24, Based on causal effects Rank all strategy combinations and select the combination with the highest causal effect score; S25. Based on the causal effect analysis results, dynamically adjust the strategy combination, and generate a dynamically adjusted encryption and access control strategy combination by combining changes in the transmission environment and real-time system monitoring data.

[0027] In this embodiment, S3 specifically includes: S31, Based on strategy combination Generate perturbation matrix ,in Indicates the combination of strategies The slight perturbation; S32. For the perturbation matrix Define the strategy combination after perturbation. : ; in, Indicates the disturbance attenuation coefficient. The norm of the strategy combination This represents the weights used to control perturbation noise. This represents the relative angular change between the security policy and the access control policy; S33, By combining strategies Perform analysis and calculate the disturbance effect:

[0028] ; in, This represents the second-order change in system performance after a disturbance. This indicates the impact of the current strategy combination on the system's target performance. Indicates the number of encryption policies. Indicates the number of access control policies. The adjustment coefficient representing the interaction effect between strategies. Indicates the encryption strategy. Indicates the access control policy; S34. Based on the calculation results of the perturbation effect, rank and optimize the strategy combination, giving priority to the perturbation effect. The minimum combination of strategies; S35. Generate a new strategy combination matrix based on the optimization results. And combine the encryption and access control strategies optimized by perturbation analysis into a matrix. Output.

[0029] In this embodiment, S4 specifically includes: S41. Input the optimized strategy results into the spatiotemporal adaptive prediction model based on meta-learning. By conducting in-depth analysis of the spatiotemporal characteristics in historical data transmission, including data traffic, network latency, node movement and environmental state change factors, construct a spatiotemporal feature matrix and extract the long-term and short-term dependencies between nodes. S42. Based on the meta-learning model, it is quickly trained using a small amount of historical data, extracts transmission patterns from different network environments, quickly adjusts parameters, and generalizes in different transmission scenarios. S43. The meta-learning model combines the dynamic changes of spatiotemporal features to analyze the changing trends of future transmission demands. By adaptively adjusting model parameters, it predicts possible future network traffic fluctuations and potential security threats. ; in, express Model parameters at time 10:00 express Model parameters at time 10:00 Indicates the learning rate. Represents model parameters gradient, The attenuation factor representing the spatiotemporal characteristics, Represents an exponential function. Indicates spatiotemporal characteristics, This indicates the impact of the current strategy combination on the system's target performance. Indicates the number of encryption policies. Indicates the number of access control policies; S44. The meta-learning model monitors the real-time environment and combines transmission history and current characteristics to make forward-looking predictions of future data transmission needs, and plans optimized encryption and access control strategies in advance. S45. Through an adaptive training mechanism, the meta-learning model is continuously updated to output a forward-looking and optimized combination of encryption and access control strategies. S46. The final prospective optimization strategy is obtained and continuously optimized in real-time transmission based on the results of the meta-learning prediction model.

[0030] In this embodiment, S5 specifically includes: S51. Input the forward-looking optimization strategy into the adaptive feedback control module to monitor the flow state changes during data transmission in real time. The flow state changes include sudden increases in traffic, packet loss, network latency fluctuations, and abnormal communication patterns. Establish a high-dimensional flow state matrix. ,in Indicates time At that time, network nodes With network nodes Between Dynamic changes in class transmission characteristics; S52. Combining Bayesian inference and spatiotemporal dependency models, construct an anomaly prediction model and generate an anomaly prediction matrix. ,in Indicates time At that time, network nodes With network nodes The probability of anomalies that may occur between them is used to identify abnormal behavior of network nodes in the early stage through Bayesian inference and spatiotemporal dependency model, and to provide early warning of potential security threats. S53. Define a feedback function based on adaptive weighting. By combining the flow matrix and anomaly prediction matrix The information is used to perform a performance evaluation of the current encryption and access control policies. ; in, Represents network nodes With network nodes The transmission status between them This represents the feedback adjustment coefficient. Indicates the first Weighting coefficients for flow-like characteristics, A dynamic weighting index representing flow regime characteristics. Indicates the number of encryption policies. Indicates the number of access control policies; S54. Based on the results calculated by the feedback function, the combination of encryption strategy and access control strategy is adjusted in real time to generate a dynamically optimized strategy matrix. : S55, Based on the adjusted strategy matrix Continuous monitoring of flow regime changes, combined with the output of anomaly prediction models, allows for continuous optimization of strategy combinations.

[0031] Example 1: To verify the feasibility of this invention in practice, it was applied to a cross-departmental data sharing system within a government department. This department needs to transfer large amounts of confidential information, including personnel information, financial data, and internal decision-making documents, between multiple subordinate units. The transmission of this data must ensure a high level of security while also maintaining high transmission efficiency to meet daily work needs and emergencies. However, traditional data transmission methods rely on fixed encryption and access control mechanisms, which are unable to cope with frequently changing network environments. Especially when the network topology between different units is complex and there are numerous communication nodes, existing static security strategies become inadequate, leading to low data transmission efficiency and a gradual increase in security risks.

[0032] We introduce a data transmission security method based on adaptive dynamic access control to address the aforementioned issues. This method dynamically adjusts encryption and access control strategies through complex network topology analysis, causal inference, perturbation analysis, meta-learning, and adaptive feedback control. This ensures that the system can optimize encryption strategies in real time under complex and ever-changing network environments and effectively prevent access abuse.

[0033] First, through the complex network topology analysis module, the system automatically acquires basic information about each network node (such as servers, end-user devices, routers, etc.), including each node's unique identifier, geographical location, connection strength, data transmission rate, and communication latency. This information is used to construct the network topology, revealing the dependencies between nodes. For example, the transmission latency between a server node located at the Beijing headquarters and branch network nodes located in Shanghai, Guangzhou, etc., is relatively short, while the latency with certain overseas nodes is longer. Based on this information, the system can identify the main transmission paths and assign appropriate encryption and access control policies to each node.

[0034] Through the causal inference module, the system evaluated and trained on combinations of different encryption and access control strategies, deriving the impact of each strategy combination on data transmission security and efficiency. For example, while using a stronger encryption algorithm improves security, the data transmission rate decreases due to increased computation. Therefore, the system uses a causal inference model to derive the optimal strategy combination, ensuring that stronger encryption strategies are used on critical nodes, while lighter encryption strategies are used on low-risk nodes, thus balancing security and efficiency.

[0035] In practice, the headquarters in City A needed to simultaneously send a highly confidential urgent decision-making document to Cities B, C, and D. Traditional transmission methods, due to rigid access control and fixed encryption strategies, resulted in excessively long data transmission times, severely impacting the timeliness of decision-making. However, by adopting this invention, the system made minor adjustments to various encryption and access control strategy combinations through perturbation analysis, evaluated the impact of different strategy combinations on network performance and security, and quickly generated the optimal encryption and access control scheme.

[0036] The meta-learning module plays a particularly prominent role in this scenario. By analyzing past emergency data transmission records from City A to City B, the system extracts the spatiotemporal characteristics of network transmission and predicts potential future transmission demands and threats. Through optimization of the prediction results, the system pre-allocates resources and encryption strategies for key nodes and transmission paths. For example, if a significant increase in transmission volume during peak hours from City A to City B is predicted, the system pre-allocates stronger encryption strategies for this path and dynamically adjusts resource allocation to ensure smoother data transmission.

[0037] Table 1. Comparison of experimental data between traditional methods and adaptive dynamic access control methods.

[0038] Based on the experimental data in Table 1 above, it is clear that the adaptive dynamic access control method of this invention is significantly superior to traditional static encryption and access control methods in several aspects. Firstly, regarding data transmission success rate, the success rate of traditional methods fluctuates between 83% and 87%, with an average of 85%. In contrast, the success rate of the method of this invention is significantly improved, reaching a minimum of 97%, a maximum of 99%, and an average of 98%. This demonstrates that the method of this invention can ensure more stable and reliable data transmission in complex and ever-changing network environments.

[0039] Regarding transmission latency, the method of this invention also demonstrates significant advantages. In traditional methods, the average transmission latency from city A to cities B, C, and D is 400ms, 350ms, and 500ms, respectively, which is relatively high and may further increase during peak hours. In contrast, the transmission latency of this invention can be reduced to as low as 240ms, 220ms, and 300ms, and even at its highest, it is only 280ms, 250ms, and 340ms, significantly improving transmission efficiency. This indicates that by dynamically adjusting encryption and access control strategies, this invention can better optimize network resource allocation and transmission paths, thereby reducing data transmission latency.

[0040] Regarding data security, traditional methods typically have long response times when dealing with cyberattacks or emergencies, often exceeding 10 minutes, and in some cases, taking more than 15 minutes to take effective measures. In contrast, the method of this invention, through real-time monitoring and adaptive feedback control, can rapidly detect anomalies and react within 1-2 minutes, and in some cases, even reduce the response time to less than 1 minute. This significant improvement in security demonstrates the flexibility and efficiency of this invention in responding to cyber threats.

[0041] Furthermore, resource utilization is another important performance indicator. Traditional static encryption and access control methods are relatively rigid in resource management, with resource utilization typically remaining between 75% and 85%, often resulting in resource waste under high loads. In contrast, this invention dynamically optimizes resource allocation, achieving a balanced improvement in resource utilization from low to high loads, with a maximum improvement of 40% and an average improvement of approximately 35%. This means that after using the method of this invention, the system can utilize CPU and bandwidth resources more efficiently, further improving overall performance.

[0042] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for ensuring the confidentiality of data transmission in the context of domestic IT innovation based on adaptive dynamic access control, characterized in that, Includes the following steps: S1. Through the complex network topology analysis module, construct the topology of the data transmission network, analyze the association and dependency relationships between network nodes, generate a topology diagram of the relationship between encryption strategies and access control, and identify and output the potential dependency relationships between network nodes and the preliminary scheme of encryption strategies. S2. Based on the obtained dependencies and preliminary scheme, the causal relationship between different encryption strategies and access control strategies is analyzed through the causal inference module. The effect of different strategy combinations in the current transmission environment is inferred, and dynamically adjusted encryption and access control strategy combinations are generated. S3. Input the encryption and access control policies into the perturbation analysis module to perform small-scale policy perturbation, evaluate the impact of policy perturbation on system security, encryption strength, access granularity and resource utilization, adjust policy parameters based on system feedback, and output the optimized policy results. S4. Input the optimized strategy results into the spatiotemporal adaptive prediction model based on meta-learning, analyze the spatiotemporal characteristics and environmental state changes in historical data transmission, quickly adjust the model parameters through the adaptive training mechanism, predict the future data transmission needs and potential threats, optimize encryption and access control strategies in advance, and obtain a forward-looking optimization strategy. S5. Input the forward-looking optimization strategy into the adaptive feedback control module, monitor the changes in the data flow state during data transmission in real time, adjust the encryption and access control strategies in combination with real-time feedback information, and output the dynamically optimized real-time strategy. S6. Input the dynamic optimization real-time strategy generated by the adaptive feedback control module into the network flow real-time monitoring module to continuously monitor and analyze the data flow, detect potential threats and anomalies in real time, and respond quickly and further optimize encryption and access control strategies based on traffic characteristics. S7. The analysis results and feedback control information of the real-time network flow monitoring module are re-inputted into the complex network topology analysis module and the causal inference module to form a closed-loop adaptive control.

2. The method for ensuring secure data transmission in information technology innovation based on adaptive dynamic access control according to claim 1, characterized in that, S1 specifically includes: S11. By acquiring the basic information of each network node in the data transmission network, including the unique identifier, geographical location, connection strength, data transmission rate, communication latency and historical communication records of the network node, a network node information matrix is ​​constructed. S12. Based on the node information matrix, construct the topology of the data transmission network to form an undirected weighted graph. ,in Represents the set of nodes in the network. Indicates the connection between network nodes. Represent the weight matrix and calculate the weights. : ； in, Represents network nodes and network nodes Real-time data transmission rate between them Represents network nodes and network nodes Communication delay between them Represents network nodes and network nodes Historical data transmission volume between them Represents network nodes and network nodes The change in angle between them , and Indicates the weighting coefficient. , , and Indicates dynamically adjusted indices and weighting coefficients; S13. By analyzing the undirected weighted graph To analyze the relationships between network nodes, a centrality analysis algorithm is used to calculate the degree centrality of each network node. Betweenness centrality and proximity centrality : Degree centrality This reflects the number of network nodes and directly connected network nodes: ； in, Indicates the total number of network nodes. Represents network nodes and network nodes Does a connection exist? Betweenness centrality Used to evaluate the mediating role of network nodes in the shortest path: ； in, Represents network nodes To network node The number of shortest paths between them. Indicates passing through network nodes The number of shortest paths; The proximity centrality This represents the average shortest path distance from a network node to other network nodes. ； in, Represents network nodes To network node The shortest path distance between them; S14. Based on the centrality analysis results, calculate the dependencies between network nodes and construct a dependency matrix. : ； in, , and This represents the weight coefficients used to calculate dependencies. and This represents the centrality adjustment factor. Indicates the attenuation coefficient. Represents network nodes betweenness centrality, Represents network nodes Proximity centrality Represents network nodes Degree centrality, Represents network nodes Degree centrality, Represents an exponential function; S15. Based on the dependency matrix Generate a preliminary encryption strategy allocation scheme, form a mapping diagram between encryption strategies and dependencies, use the preliminary scheme to identify the encryption level required for each network node, and output the preliminary encryption strategy allocation results; S16. The final output includes a preliminary scheme for the potential dependencies between network nodes and encryption strategies.

3. The method for ensuring secure data transmission in information technology innovation based on adaptive dynamic access control according to claim 1, characterized in that, S2 specifically includes: S21. Based on the potential dependencies and preliminary encryption strategy schemes output in step S16, construct the encryption strategy matrix. Access control policy matrix ,in Indicates the number of encryption policies. Indicates the number of access control policies; S22, Define the strategy combination matrix ,in Indicate encryption strategy and access control policies Based on real-time and historical data from the transmission environment, a causal inference model is established to describe the multiple impacts of each strategy combination on system performance. ； in, This indicates the impact of the current strategy combination on the system's target performance. Indicates the bias term. , and Indicate encryption strategy The influence coefficients of the linear, squared, and logarithmic terms of the system. , and Indicates access control policy The influence coefficients on the linear, cubic, and square root terms of the system. This represents the interaction coefficient between the encryption policy and the access control policy. Indicates the noise term; S23. Train the causal inference model, optimize the model parameters using maximum likelihood estimation and gradient descent, and calculate the strategy combination based on the training results. Causal effect: ； in, Indicates strategy combination The causal effect; S24, Based on causal effects Rank all strategy combinations and select the combination with the highest causal effect score; S25. Based on the causal effect analysis results, dynamically adjust the strategy combination, and generate a dynamically adjusted encryption and access control strategy combination by combining changes in the transmission environment and real-time system monitoring data.

4. The method for ensuring secure data transmission in information technology innovation based on adaptive dynamic access control according to claim 1, characterized in that, S3 specifically includes: S31, Based on strategy combination Generate perturbation matrix ,in Indicates the combination of strategies The slight perturbation; S32. For the perturbation matrix Define the strategy combination after perturbation. : ； in, Indicates the disturbance attenuation coefficient. The norm of the strategy combination This represents the weights used to control perturbation noise. This represents the relative angular change between the security policy and the access control policy; S33, By combining strategies Perform analysis and calculate the disturbance effect:

5. 。 in, This represents the second-order change in system performance after a disturbance. This indicates the impact of the current strategy combination on the system's target performance. Indicates the number of encryption policies. Indicates the number of access control policies. The adjustment coefficient representing the interaction effect between strategies. Indicates the encryption strategy. Indicates the access control policy; S34. Based on the calculation results of the perturbation effect, rank and optimize the strategy combination, giving priority to the perturbation effect. The minimum combination of strategies; S35. Generate a new strategy combination matrix based on the optimization results. And combine the encryption and access control strategies optimized by perturbation analysis into a matrix. Output.

6. The method for ensuring secure data transmission in information technology innovation based on adaptive dynamic access control according to claim 1, characterized in that, S4 specifically includes: S41. Input the optimized strategy results into the spatiotemporal adaptive prediction model based on meta-learning. By conducting in-depth analysis of the spatiotemporal characteristics in historical data transmission, including data traffic, network latency, node movement and environmental state change factors, construct a spatiotemporal feature matrix and extract the long-term and short-term dependencies between nodes. S42. Based on the meta-learning model, it is quickly trained using a small amount of historical data, extracts transmission patterns from different network environments, quickly adjusts parameters, and generalizes in different transmission scenarios. S43. The meta-learning model combines the dynamic changes of spatiotemporal features to analyze the changing trends of future transmission demands. By adaptively adjusting model parameters, it predicts possible future network traffic fluctuations and potential security threats. ； in, express Model parameters at time 10:00 express Model parameters at time 10:00 Indicates the learning rate. Represents model parameters gradient, The attenuation factor representing the spatiotemporal characteristics, Represents an exponential function. Indicates spatiotemporal characteristics, This indicates the impact of the current strategy combination on the system's target performance. Indicates the number of encryption policies. Indicates the number of access control policies; S44. The meta-learning model monitors the real-time environment and combines transmission history and current characteristics to make forward-looking predictions of future data transmission needs, and plans optimized encryption and access control strategies in advance. S45. Through an adaptive training mechanism, the meta-learning model is continuously updated to output a forward-looking and optimized combination of encryption and access control strategies. S46. The final prospective optimization strategy is obtained and continuously optimized in real-time transmission based on the results of the meta-learning prediction model.

7. The method for secure data transmission based on adaptive dynamic access control according to claim 1, characterized in that, S5 specifically includes: S51. Input the forward-looking optimization strategy into the adaptive feedback control module to monitor the flow state changes during data transmission in real time. The flow state changes include sudden increases in traffic, packet loss, network latency fluctuations, and abnormal communication patterns. Establish a high-dimensional flow state matrix. ,in Indicates time At that time, network nodes With network nodes Between Dynamic changes in class transmission characteristics; S52. Combining Bayesian inference and spatiotemporal dependency models, construct an anomaly prediction model and generate an anomaly prediction matrix. ,in Indicates time At that time, network nodes With network nodes The probability of anomalies that may occur between them is used to identify abnormal behavior of network nodes in the early stage through Bayesian inference and spatiotemporal dependency model, and to provide early warning of potential security threats. S53. Define a feedback function based on adaptive weighting. By combining the flow matrix and anomaly prediction matrix The information is used to perform a performance evaluation of the current encryption and access control policies. ； in, Represents network nodes With network nodes The transmission status between them This represents the feedback adjustment coefficient. Indicates the first Weighting coefficients for flow-like characteristics, A dynamic weighting index representing flow regime characteristics. Indicates the number of encryption policies. Indicates the number of access control policies; S54. Based on the results calculated by the feedback function, the combination of encryption strategy and access control strategy is adjusted in real time to generate a dynamically optimized strategy matrix. : S55, Based on the adjusted strategy matrix Continuous monitoring of flow regime changes, combined with the output of anomaly prediction models, allows for continuous optimization of strategy combinations.

Images