Data transmission method and device, equipment and medium

By performing abnormal detection and digital asset portrait processing on data transmission traffic, calculating traffic sensitivity and performing precise desensitization, the problem of insufficient data transmission security in the existing technology is solved, the security and efficiency of data transmission are improved, and the security and efficiency of data transmission are reduced. Data breach risk.

CN118890185APending Publication Date: 2024-11-01EVERSEC BEIJING TECH
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410962054.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-17
Publication Date
2024-11-01

AI Technical Summary

Technical Problem

The existing technology is difficult to effectively improve the security of data transmission, especially when facing data leakage and sensitive information theft, resulting in financial, business and reputation losses in enterprises.

Method used

By obtaining the data traffic to be transmitted, performing basic abnormal data detection and digital asset portrait processing, determining the abnormal state of the data packet, and then calculating the traffic sensitivity, and selecting appropriate desensitization strategies for data transmission based on the sensitivity to improve the security of data transmission sex.

Benefits of technology

It improves the security and efficiency of data transmission, ensures accurate desensitization of data during transmission, and reduces economic losses and data leakage risks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118890185A_ABST
    Figure CN118890185A_ABST
Patent Text Reader

Abstract

The embodiment of the invention discloses a data transmission method and device, equipment and a medium. The method comprises the following steps: acquiring to-be-transmitted data traffic, and acquiring corresponding content data according to the to-be-transmitted data traffic; performing abnormal data basic detection on the content data according to a preset basic detection rule base, and determining a basic abnormal state of the content data; performing digital asset portrait processing on the content data to obtain a to-be-transmitted data packet including the content data, and performing abnormal data packet detection on the to-be-transmitted data packet to obtain a data packet abnormal state of the to-be-transmitted data packet; determining the traffic sensitivity of the data traffic to be transmitted according to the basic abnormal state and the data packet abnormal state; and selecting a corresponding target desensitization strategy from pre-constructed candidate desensitization strategies according to the traffic sensitivity, and performing desensitization processing on the data packet to be transmitted according to the target desensitization strategy so as to transmit the desensitized content data. According to the scheme, the security of data transmission is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of data transmission, and in particular, to a data transmission method, apparatus, device and medium. Background Art

[0002] At present, while data is playing an increasingly important role, data security issues are becoming increasingly prominent. The definition of data security refers to taking necessary measures to ensure that data is in a state of effective protection and legal use, as well as the ability to ensure continuous security. Therefore, how to improve the security of data transmission is crucial. Summary of the invention

[0003] The present invention provides a data transmission method, device, equipment and medium to improve the security of data transmission.

[0004] According to one aspect of the present invention, there is provided a data transmission method, comprising:

[0005] Acquire the data flow to be transmitted, and acquire corresponding content data according to the data flow to be transmitted;

[0006] According to a preset basic detection rule library, performing basic abnormal data detection on the content data to determine a basic abnormal state of the content data;

[0007] Performing digital asset profiling processing on the content data to obtain a data packet to be transmitted including the content data, and performing abnormal data packet detection on the data packet to be transmitted to obtain a data packet abnormal state of the data packet to be transmitted;

[0008] Determining the flow sensitivity of the data flow to be transmitted according to the basic abnormal state and the data packet abnormal state;

[0009] According to the traffic sensitivity, a corresponding target desensitization strategy is selected from pre-constructed candidate desensitization strategies, and the data packet to be transmitted is desensitized according to the target desensitization strategy to transmit the desensitized content data.

[0010] According to another aspect of the present invention, there is provided a data transmission device, comprising:

[0011] A content data acquisition module, used to acquire the data flow to be transmitted, and acquire corresponding content data according to the data flow to be transmitted;

[0012] A basic abnormal state determination module, configured to perform a basic abnormal data detection on the content data according to a preset basic detection rule base, and determine a basic abnormal state of the content data;

[0013] A data packet abnormal state determination module is used to perform digital asset profiling processing on the content data to obtain a data packet to be transmitted including the content data, and perform abnormal data packet detection on the data packet to be transmitted to obtain a data packet abnormal state of the data packet to be transmitted;

[0014] A flow sensitivity determination module, used to determine the flow sensitivity of the data flow to be transmitted according to the basic abnormal state and the data packet abnormal state;

[0015] The data transmission module is used to select a corresponding target desensitization strategy from pre-built candidate desensitization strategies according to the traffic sensitivity, and desensitize the data packet to be transmitted according to the target desensitization strategy to transmit the desensitized content data.

[0016] According to another aspect of the present invention, there is provided an electronic device, comprising:

[0017] one or more processors;

[0018] A memory for storing one or more programs;

[0019] When one or more programs are executed by one or more processors, the one or more processors can execute any one of the data transmission methods provided by the embodiments of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement any one of the data transmission methods provided in the embodiments of the present invention when executed.

[0021] The embodiment of the present invention provides a data transmission scheme, which obtains the data flow to be transmitted, and obtains the corresponding content data according to the data flow to be transmitted; performs basic abnormal data detection on the content data according to a preset basic detection rule library to determine the basic abnormal state of the content data; performs digital asset portrait processing on the content data to obtain a data packet to be transmitted including the content data, and performs abnormal data packet detection on the data packet to be transmitted to obtain the data packet abnormal state of the data packet to be transmitted; determines the flow sensitivity of the data flow to be transmitted according to the basic abnormal state and the data packet abnormal state; selects the corresponding target desensitization strategy from the pre-constructed candidate desensitization strategy according to the flow sensitivity, and performs desensitization processing on the data packet to be transmitted according to the target desensitization strategy, so as to transmit the desensitized content data. The above scheme, by jointly determining the flow sensitivity of the data flow to be transmitted according to the abnormal data basic detection and the abnormal data packet detection, improves the accuracy of the determined flow sensitivity, thereby improving the accuracy of the target desensitization strategy determined according to the flow sensitivity, and desensitizing the data packet to be transmitted, and improves the security of data transmission; at the same time, in the embodiment of the present invention, the required data is obtained from the data flow to be transmitted, without obtaining it from the database, thereby improving the efficiency.

[0022] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 is a flow chart of a data transmission method provided by Embodiment 1 of the present invention;

[0025] Figure 2 is a flow chart of a data transmission method provided by Embodiment 2 of the present invention;

[0026] Figure 3A This is a general architecture diagram of a pipeline-side sensitive data transmission feature detection system provided in Embodiment 3 of the present invention;

[0027] Figure 3B This is a schematic diagram of the structure of a high-speed regular matching engine provided by Embodiment 3 of the present invention;

[0028] Figure 3CThis is an architecture diagram of a desensitization process provided by Embodiment 3 of the present invention;

[0029] Figure 4 is a structural diagram of a data transmission device provided by Embodiment 4 of the present invention;

[0030] Figure 5 It is a structural schematic diagram of an electronic device for implementing a data transmission method provided in Embodiment 5 of the present invention. DETAILED DESCRIPTION

[0031] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0032] In the digital economy era, the circulation and transaction of data can maximize its value, and compliance assurance is the bottom line requirement of organizational data security governance, and risk management is an important issue that needs to be addressed in data security governance. Therefore, the goal of data security governance is to achieve the development and utilization of data under the premise of compliance assurance and risk management, to ensure the sustainable and healthy development of the business, and to ensure the two-way promotion of data security and business development.

[0033] Data security issues can cause immeasurable losses to enterprises through the theft and leakage of sensitive information, and may lead to multiple losses in finance, business and reputation, which may even be irreparable in some scenarios. Therefore, helping industrial enterprises avoid losses caused by data security leaks and protecting important data throughout its life cycle can invisibly reduce economic losses and thus improve economic benefits.

[0034] The definition of data security refers to taking necessary measures to ensure that data is in a state of effective protection and legal use, as well as the ability to ensure continuous security. To ensure the security of the entire data processing process, data processing includes data collection, storage, use, processing, transmission, provision, and disclosure. Information security or data security has two opposing meanings: one is the security of the data itself, which mainly refers to the active protection of data using modern cryptographic algorithms, such as data confidentiality, data integrity, and two-way strong identity authentication. The second is the security of data protection, which mainly refers to the active protection of data using modern information storage methods, such as disk arrays, data backup, and off-site disaster recovery. Data security is an active protection measure. The security of the data itself must be based on reliable encryption algorithms and security systems, mainly symmetric algorithms and public key cryptographic systems.

[0035] Data processing security refers to how to effectively prevent database damage or data loss caused by hardware failure, power failure, crash, human error, program defects, viruses or hackers during data entry, processing, statistics or printing. Some sensitive or confidential data may be read by unqualified personnel or operators, resulting in data leakage and other consequences. Data storage security refers to the readability of the database outside the system operation. Once the database is stolen, even if there is no original system program, another program can still be written to view or modify the stolen database. From this perspective, an unencrypted database is unsafe and prone to commercial leakage, so the concept of data leakage prevention is derived, which involves issues such as confidentiality, security and software protection of computer network communications.

[0036] In the era of digital economy, the amount of global data generated by the widespread application of technologies such as big data, cloud computing, artificial intelligence, and the Internet of Things is growing exponentially. Data has become an important production factor and a basic strategic resource, and its value is becoming increasingly prominent. At the same time, global data security risks are increasing day by day. Issues such as data security and privacy protection, data storage, use, and cross-border flow risks have posed new challenges to data security governance capabilities. With the continuous progress of networking, digitalization, and informatization, the endless stream of data leaks and network security incidents have also brought new challenges to personal information and privacy protection. How to prevent the theft of important data, large-scale data abuse, protect personal data privacy, and safeguard the legitimate rights and interests of end consumers have become important issues in the data application service industry.

[0037] Data, as a new means of production, is considered a production factor and the new gold. Its importance is self-evident. Flow and sharing have become new features of data. Data security work has gradually changed from a "system"-centered approach to a "data"-centered approach. The higher the value contained in the data, the more vulnerable it is to attack, and the attack methods are more complex and diverse. The ever-changing data security threats have caused great harm to the digital economic order, and strong security protection measures and continuous security operations are needed to effectively resist them.

[0038] As we fully enter the 5G era, the collection, circulation, integration and sharing of massive amounts of data will inevitably occur. The resulting data security issues will also bring new risks and challenges to operation and maintenance management. Therefore, it is urgent to strengthen data security governance based on the big data environment, reduce the occurrence of personal information leakage or modification, and protect personal information security.

[0039] Embodiment 1

[0040] Figure 1This is a flowchart of a data transmission method provided in Example 1 of the present invention. This embodiment is applicable to situations where data is transmitted. The method can be executed by a data transmission device, which can be implemented in software and / or hardware and can be configured in an electronic device that carries the data transmission function.

[0041] See also Figure 1 The data transmission method shown includes:

[0042] S110: Acquire data flow to be transmitted, and acquire corresponding content data according to the data flow to be transmitted.

[0043] The data flow to be transmitted refers to the flow of data to be transmitted. Exemplarily, the data flow to be transmitted may include content data. Content data refers to the specific data content to be transmitted. The embodiment of the present invention does not impose any limitation on the format of the content data, which can be set by technicians based on experience. Exemplarily, the content data can be a file.

[0044] Specifically, the data flow to be transmitted is obtained, and the content data is obtained from the data flow to be transmitted.

[0045] S120: Perform basic abnormal data detection on the content data according to a preset basic detection rule library to determine a basic abnormal state of the content data.

[0046] Among them, the basic detection rule base refers to a database that stores relevant rules for basic detection. The embodiment of the present invention does not impose any limitation on the setting of the basic detection rule base, which can be set by a technician based on experience. Exemplarily, at least one initial detection rule can be stored in the basic detection rule base. The initial detection rule refers to a pre-set rule for performing basic detection of abnormal data. The embodiment of the present invention does not impose any limitation on the setting of the initial detection rule, which can be set by a technician based on experience. Exemplarily, the initial detection rule may include at least one of a protocol parsing rule, a protocol decoding rule, a file format detection rule, and a file restoration rule.

[0047] Among them, the basic detection of abnormal data refers to the basic detection of content data, that is, the basic detection of abnormal data can be understood as a coarse-grained detection of content data. The basic abnormal state can characterize whether the content data is abnormal after the basic detection of abnormal data is performed on the content data. Exemplarily, the basic abnormal state may include a basic no-abnormal state and a basic abnormal state. The basic no-abnormal state is used to characterize that there is no abnormality after the basic detection of abnormal data is performed on the content data. The basic abnormal state is used to characterize that there is an abnormality after the basic detection of abnormal data is performed on the content data.

[0048] In an optional embodiment, basic abnormal data detection is performed on content data according to a preset basic detection rule library to determine a basic abnormal state of the content data, including: determining candidate detection rules for performing basic abnormal data detection on content data according to a rule content table in the preset basic detection rule library; determining a rule effective time of the candidate detection rule according to a rule state table in the preset basic detection rule library, and selecting a target detection rule from the candidate detection rules according to the rule effective time; performing basic abnormal data detection on content data according to the target detection rule to determine a basic abnormal state of the content data.

[0049] The basic detection rule library in the embodiment of the present invention includes a rule content table and a rule status table. Among them, the rule content table can be used to store the specific content of the initial detection rule, that is, the content characteristics of each initial detection rule can be stored in the rule content table. Candidate detection rules refer to initial detection rules that may be used for basic detection of abnormal data selected from the rule content table according to needs. The rule status table can be used to store the temporal characteristics of the initial detection rules, that is, the timeliness characteristics of each initial detection rule can be stored in the rule status table. The rule effective time refers to the effective time when the candidate detection rule takes effect. The target detection rule refers to the candidate detection rule for basic detection of abnormal data.

[0050] Specifically, candidate detection rules for basic detection of abnormal data are selected from the rule content table; the rule effective time corresponding to each candidate detection rule is determined from the rule status table, and a target detection rule is selected from the candidate detection rules according to the rule effective time; according to the target detection rule, basic detection of abnormal data is performed on the content data to determine the basic abnormal status of the content data.

[0051] It can be understood that by determining the target detection rules for basic abnormal data detection on the content data based on the rule content table and the rule status table, the accuracy of the determined target detection rules is improved, thereby improving the accuracy of the basic abnormal status determined based on the target detection rules.

[0052] S130, performing digital asset profiling processing on the content data to obtain a data packet to be transmitted including the content data, and performing abnormal data packet detection on the data packet to be transmitted to obtain a data packet abnormal status of the data packet to be transmitted.

[0053] Among them, the data packet to be transmitted refers to the data packet that needs to be transmitted. The data packet abnormal state can indicate whether the data packet to be transmitted has abnormalities after abnormal data packet detection is performed on the data packet to be transmitted. Exemplarily, the data packet abnormal state can include a data packet no abnormality state and a data packet abnormality state. The data packet no abnormality state is used to indicate that there is no abnormality after abnormal data packet detection is performed on the data packet to be transmitted. The data packet abnormality state is used to indicate that there is an abnormality after abnormal data packet detection is performed on the data packet to be transmitted.

[0054] Exemplarily, digital asset profiling is performed on the content data to determine the data receiver identifier and data originator identifier of the content data. Digital asset profiling of the content data can be understood as determining the application scenario of the content data, that is, the data asset tag. Among them, the data originator identifier is used to characterize the identity of the data originator. The data originator is used to send content data. The data receiver identifier is used to characterize the identity of the data receiver. The data receiver is used to receive content data.

[0055] Exemplarily, the data packet to be transmitted may include content data, a data receiver identifier, a data sender identifier, a data asset tag, and a data access time, etc.

[0056] S140: Determine the traffic sensitivity of the data traffic to be transmitted according to the basic abnormal state and the data packet abnormal state.

[0057] The traffic sensitivity can be used to quantify the sensitivity of the data traffic to be transmitted.

[0058] Exemplarily, determining the flow sensitivity of the data flow to be transmitted according to the basic abnormal state and the data packet abnormal state includes: determining the basic weight corresponding to the basic abnormal state, and determining the data packet weight corresponding to the data packet abnormal state; determining the flow sensitivity of the data flow to be transmitted according to the basic weight, the data packet weight, the basic abnormal state and the data packet abnormal state. Among them, the basic weight refers to the proportion of the basic abnormal state in determining the flow sensitivity. The data packet weight refers to the proportion of the data packet abnormal state in determining the flow sensitivity. The embodiment of the present invention does not impose any limitation on the size of the basic weight and the data packet weight, which can be set by the technician according to experience or needs.

[0059] S150. According to the traffic sensitivity, a corresponding target desensitization strategy is selected from pre-constructed candidate desensitization strategies, and the data packet to be transmitted is desensitized according to the target desensitization strategy to transmit the desensitized content data.

[0060] Among them, the candidate desensitization strategy refers to a pre-set strategy that can perform data traffic desensitization. The embodiment of the present invention does not impose any restrictions on the setting of the candidate desensitization strategy, which can be set by technical personnel based on experience or needs. The target desensitization strategy refers to a strategy for desensitizing data packets to be transmitted. It should be noted that the embodiment of the present invention does not impose any restrictions on the number of target desensitization strategies, which can be at least one.

[0061] Specifically, according to the traffic sensitivity, at least one target desensitization strategy is selected from the candidate sensitive desensitization strategies; the data packet to be transmitted is desensitized according to the target desensitization strategy; and the content data of the data packet to be transmitted after the desensitization is sent to the data receiver.

[0062] The embodiment of the present invention provides a data transmission scheme, which obtains the data flow to be transmitted, and obtains the corresponding content data according to the data flow to be transmitted; performs basic abnormal data detection on the content data according to a preset basic detection rule library to determine the basic abnormal state of the content data; performs digital asset portrait processing on the content data to obtain a data packet to be transmitted including the content data, and performs abnormal data packet detection on the data packet to be transmitted to obtain the data packet abnormal state of the data packet to be transmitted; determines the flow sensitivity of the data flow to be transmitted according to the basic abnormal state and the data packet abnormal state; selects the corresponding target desensitization strategy from the pre-constructed candidate desensitization strategy according to the flow sensitivity, and performs desensitization processing on the data packet to be transmitted according to the target desensitization strategy, so as to transmit the desensitized content data. The above scheme, by jointly determining the flow sensitivity of the data flow to be transmitted according to the abnormal data basic detection and the abnormal data packet detection, improves the accuracy of the determined flow sensitivity, thereby improving the accuracy of the target desensitization strategy determined according to the flow sensitivity, and desensitizing the data packet to be transmitted, and improves the security of data transmission; at the same time, in the embodiment of the present invention, the required data is obtained from the data flow to be transmitted, without obtaining it from the database, thereby improving the efficiency.

[0063] Embodiment 2

[0064] Figure 2It is a flow chart of a data transmission method provided in the second embodiment of the present invention. Based on the above embodiments, this embodiment further refines the operation of "performing digital asset profiling on the content data to obtain a data packet to be transmitted including the content data, and performing abnormal data packet detection on the data packet to be transmitted to obtain the data packet abnormal state of the data packet to be transmitted" into "performing digital asset profiling on the content data to obtain the form data corresponding to the content data, and generating a data packet to be transmitted including the content data and the form data; performing deep abnormal data detection on the content data to determine the deep abnormal state of the content data; performing formal transmission detection on the form data according to a preset candidate form detection strategy to determine the formal abnormal state of the form data; generating a data packet abnormal state including a deep abnormal state and a formal abnormal state" to improve the determination mechanism of the data packet abnormal state. It should be noted that for the part not described in detail in the embodiment of the present invention, reference may be made to the description of other embodiments.

[0065] See also Figure 2 The data transmission method shown includes:

[0066] S210: Acquire data traffic to be transmitted, and acquire corresponding content data according to the data traffic to be transmitted.

[0067] S220: Perform basic abnormal data detection on the content data according to a preset basic detection rule library to determine a basic abnormal state of the content data.

[0068] S230, perform digital asset profiling on the content data to obtain form data corresponding to the content data, and generate a data packet to be transmitted including the content data and the form data.

[0069] The form data may be used to determine the data transmission scenario that the content data is in. Exemplarily, the form data may include a data receiver identifier, a data sender identifier, a data asset tag, and a data access time.

[0070] S240: Perform a deep abnormal data detection on the content data to determine a deep abnormal state of the content data.

[0071] Among them, the abnormal data deep detection refers to the abnormal detection of specific content in the content data, that is, the abnormal data deep detection can be understood as fine-grained detection of the content data. The deep abnormal state can characterize whether the content data is abnormal after the abnormal data deep detection is performed on the content data. Exemplarily, the deep abnormal state can include a deep no abnormal state and a deep abnormal state. The deep no abnormal state can characterize that there is no abnormality after the abnormal data deep detection is performed on the content data. The deep abnormal state can characterize the presence of abnormalities after the abnormal data deep detection is performed on the content data.

[0072] S250: Perform a form transmission detection on the form data according to a preset candidate form detection strategy to determine a form abnormality state of the form data.

[0073] Among them, the candidate form detection strategy refers to a pre-set strategy that can detect the form data. The form transmission detection refers to the abnormality detection of the form data. The form abnormality state can represent whether the form data is abnormal after the form transmission detection is performed on the form data. Exemplarily, the form abnormality state can include the form no abnormality state and the form abnormality state. The form no abnormality state can represent that there is no abnormality after the form transmission detection is performed on the form data. The form abnormality state can represent that there is an abnormality after the form transmission detection is performed on the form data.

[0074] In an optional embodiment, according to a preset candidate form detection strategy, a form transmission detection is performed on the form data to determine the form abnormality state of the form data, including: determining the current form detection strategy from the candidate form detection strategies; according to the current form detection strategy, performing a form transmission detection on the form data to obtain the current form abnormality state of the form data; according to the current form abnormality state, determining the form abnormality state of the form data.

[0075] Among them, the current form detection strategy refers to the candidate form detection strategy used to detect the form data at the current moment. The current form abnormal state can represent the detection result of the form data according to the current form detection strategy. Exemplarily, the current form abnormal state can include the current form abnormal state and the current form non-abnormal state. The current form abnormal state can represent that there is an abnormality after the form data is detected according to the current form strategy. The current form non-abnormal strategy can represent that there is no abnormality after the form data is detected according to the current form strategy.

[0076] Optionally, determining the form abnormality state of the form data according to the current form abnormality state includes: determining the current form strategy weight corresponding to the current form detection strategy; and performing weighted summation of the current form abnormality state according to the current form strategy weight to determine the form abnormality state. The current form strategy weight can represent the importance of the current form abnormality state in the form abnormality state.

[0077] Or optionally, determining the form abnormality state of the form data according to the current form abnormality state includes: if there is at least one current form abnormality state that is a current form abnormality state, determining the form abnormality state of the form data as a form abnormality state; if the current form abnormality states are all current form non-abnormal states, determining the form abnormality state of the form data as a form non-abnormal state.

[0078] Alternatively, according to the current form abnormality state, determining the form abnormality state of the form data includes: if the current form abnormality state is the number of abnormal states of the current form, and reaches a preset abnormality number threshold, then determining the form abnormality state of the form data is a form abnormal state; if not, then determining the form abnormality state of the form data is a form non-abnormal state. The embodiment of the present invention does not impose any limitation on the size of the preset abnormality number threshold, which can be set by a technician based on experience, or determined repeatedly through a large number of experiments.

[0079] It can be understood that by introducing the current form detection strategy, determining the current form abnormal state corresponding to the current form detection strategy, and then determining the form abnormal state of the form data based on the current form abnormal state, the accuracy of the determined form abnormal state is improved.

[0080] In an optional embodiment, the candidate form detection strategy includes at least one of a transmission rule detection strategy, a transmission context detection strategy, a transmission behavior detection strategy, a transmission baseline detection strategy, and a transmission prediction detection strategy.

[0081] Among them, the transmission rule detection strategy can be used to perform rule-based association analysis on formal data. Exemplarily, rule-based association analysis refers to the process in which the system matches rules through the event association engine to identify data security violations of known patterns, which is a type of association analysis technology. Through multi-event association, the system can perform complex event rule matching on data security event streams that meet multiple rules (called combined rules).

[0082] Among them, the transmission context detection strategy can be used to detect the scenarios represented by the formal data, which can be understood as context-based correlation analysis. For example, context-based correlation analysis refers to associating data security events with the actual operating environment of the current network and business, and identifying security leakage risks through a wider range of information correlation analysis. This technology is also called "context-aware".

[0083] Among them, the transmission behavior detection strategy can be used to detect the transmission behavior represented by the formal data, which can be understood as behavior-based correlation analysis. For example, the positioning of behavior-based correlation analysis is to reverse the data security analysis to the active analysis model based on anomaly detection, so that the mainstream analysis method of data security monitoring and analysis is no longer strongly dependent on the correlation engine.

[0084] Among them, the transmission baseline detection strategy can be used to detect changes in formal data, that is, the transmission baseline detection strategy can adopt dynamic baseline technology. Exemplarily, a periodic baseline analysis method is adopted, and the periodic baseline is calculated based on historical behavior data, usually a single-cycle database contour line. This curve is composed of several data contour points, each contour point represents a sampling time point, and if a new actual measurement value does not exceed the baseline range, the old contour value is updated through a weighted average algorithm. If the new actual measurement value exceeds the baseline range, it is discarded and does not participate in the calculation of the new contour value. In this reciprocating cycle, the baseline is always in dynamic change.

[0085] Among them, the transmission prediction detection strategy can be used to predict and detect the form data, that is, the transmission prediction detection strategy can adopt the prediction analysis technology. Exemplarily, the detection model and method based on the time window confidence interval are adopted. It can continuously self-adjust and approximate in actual operation, automatically eliminate abnormal historical data in the historical time window, and achieve a high degree of consistency between the historical time window data and the actual normal traffic behavior characteristics of the network, thereby improving the accuracy of abnormal behavior alarm.

[0086] It can be understood that by introducing the transmission rule detection strategy, the transmission context detection strategy, the transmission behavior detection strategy, the transmission baseline detection strategy and the transmission prediction detection strategy, the comprehensiveness of the candidate form detection strategy is improved.

[0087] In an optional embodiment, if the current form detection strategy is a transmission behavior detection strategy, then according to the current form detection strategy, a form transmission detection is performed on the form data to obtain the current form abnormality state of the form data, including: determining the data transmission scenario corresponding to the form data according to the data initiator identifier and the data receiver identifier in the form data; obtaining historical behavior data under the data transmission scenario from a database, and determining the historical access time according to the historical behavior data; determining the data access time from the form data, and performing a behavior transmission detection on the form data according to the historical access time and the data access time to obtain the current form abnormality state of the form data.

[0088] Among them, the data transmission scenario refers to the data transmission scenario represented by the formal data. The data access time refers to the time of accessing the data recipient. Exemplarily, the data access time can be determined based on the data flow to be transmitted. Historical behavior data refers to past behavior data in the data transmission scenario. Behavior transmission detection refers to abnormal detection of the transmission behavior of the formal data. Historical access time refers to the time of previous accesses to the data transmission scenario. The database can be used to store historical data of data transmission, such as historical behavior data.

[0089] Exemplarily, if the data transmission scenario is electricity consumption query and the data access time is 6:00, the historical behavior data is the behavior data in the previous electricity consumption query scenario obtained from the database, and the historical access time is determined based on the historical behavior data; the historical access time is matched with the data access time; if the match is successful, the current form abnormal state of the form data is determined to be that the current form has no abnormality; if not, the current form abnormal state of the form data is determined to be that the current form has an abnormal state.

[0090] It can be understood that by introducing the historical access time and the data access time and performing behavior transmission detection on the form data, the accuracy of determining the current form abnormal state is improved.

[0091] S260: Generate a data packet abnormality state including a deep abnormality state and a formal abnormality state.

[0092] S270: Determine the traffic sensitivity of the data traffic to be transmitted according to the basic abnormal state and the data packet abnormal state.

[0093] S280. According to the traffic sensitivity, a corresponding target desensitization strategy is selected from pre-built candidate desensitization strategies, and the data packet to be transmitted is desensitized according to the target desensitization strategy to transmit the desensitized content data.

[0094] The embodiment of the present invention provides a data transmission scheme, which performs digital asset profiling on content data to obtain a data packet to be transmitted including content data, and performs abnormal data packet detection on the data packet to be transmitted to obtain a data packet abnormal state operation of the data packet to be transmitted, which is refined into performing digital asset profiling on content data to obtain form data corresponding to the content data, and generating a data packet to be transmitted including content data and form data; performing abnormal data depth detection on the content data to determine the deep abnormal state of the content data; performing form transmission detection on the form data according to a preset candidate form detection strategy to determine the formal abnormal state of the form data; generating a data packet abnormal state including a deep abnormal state and a formal abnormal state, thereby improving the determination mechanism of the data packet abnormal state. The above scheme improves the accuracy of the determined data packet abnormal state by performing abnormal detection on the form data and the content data respectively, generating a data packet abnormal state including a deep abnormal state and a formal abnormal state.

[0095] Embodiment 3

[0096] Based on the above technical solution, the embodiment of the present invention provides a pipeline side sensitive data transmission feature detection system. The pipeline side sensitive data transmission feature detection system adopts the data transmission method provided by the above embodiment.

[0097] Exemplarily, the industrial Internet is the product of the deep integration of the industrial system and the Internet system under the digital wave. The embodiment of the present invention relies on the deep analysis and active monitoring of data traffic, combined with technologies such as multi-dimensional data association fusion, to build a pipeline-side sensitive data transmission feature detection system, deploy operator pipeline flow measurement data access, and perform flow behavior analysis and deep data mining on the data, providing strong technical support for industrial Internet data security. Through the research on sensitive data transmission feature detection technology based on pipeline-side flow analysis, we can make full use of the resource advantages of the operator's flow side, actively explore the comprehensive solution for industrial Internet data security monitoring, and simultaneously precipitate China Unicom's own security value engine. Focusing on network traffic analysis technology (NTA), we build core capabilities such as data security monitoring capabilities, flow behavior analysis, and deep data mining, and build an industrial Internet pipeline-side sensitive data transmission feature detection system. Realize the transmission behavior feature identification, data leakage monitoring, data abnormal flow monitoring, data security risk notification and tracking and verification of important sensitive data of the industrial Internet platform, and carry out data security monitoring for key industrial Internet platforms through inlet and outlet flow data and Internet of Things inlet and outlet flow data, providing data support for the industrial Internet data security monitoring platform.

[0098] In the embodiments of the present invention, the prior art involved is first described. Data security: By taking necessary measures, ensure that the data is in a state of effective protection and legal use, and have the ability to ensure a continuous security state. NTA technology: Analyze the TCP / IP protocol through identification methods such as TCP / UDP ports, message load characteristics, correlation analysis detection, and behavior characteristics, and identify specific services running on the TCP / IP protocol. Parallel protocol stack restoration technology: Use multi-threading technology to restore the captured Ethernet data message into application layer data for efficient analysis and processing. Code stream matching technology: Based on HTTP protocol information, DNS protocol information, MMS protocol information, FTP protocol information, Email protocol information, Voip protocol information, instant messaging protocol information, P2P protocol information, etc., a code stream feature library is established, and the communication traffic of the industrial Internet is monitored, and then compared with the feature library to identify sensitive information in the data stream. UEBA: That is, user and entity behavior analysis, which can achieve extensive security analysis and provides user-centric analysis around user behavior, but also around other things such as endpoints, networks and applications. The correlation of analysis across different entities makes the analysis results more accurate and makes detection more effective.

[0099] See also Figure 3AThe overall architecture diagram of the pipeline-side sensitive data transmission feature detection system is shown. Among them, the pipeline-side sensitive data transmission feature detection system adopts a Web interface to facilitate management personnel to analyze, process and generate reports anytime and anywhere. The pipeline-side sensitive data transmission feature detection system consists of a data security monitoring subsystem, a traffic behavior analysis and deep data mining subsystem and a data desensitization subsystem. In an embodiment of the present invention, the data security monitoring subsystem can obtain the data flow to be transmitted from relevant institutions; the traffic behavior analysis and deep data mining subsystem can obtain the data flow or content data to be transmitted from the data security monitoring subsystem. The data security monitoring subsystem sends the basic abnormal state to the data desensitization subsystem, and the traffic behavior analysis and deep data mining subsystem sends the data packet abnormal state to the data desensitization subsystem. The data desensitization subsystem determines the traffic sensitivity based on the basic abnormal state and the data packet abnormal state, and determines the corresponding target desensitization strategy, and desensitizes the data flow to be transmitted according to the target desensitization strategy.

[0100] Among them, the data security monitoring subsystem is mainly based on built-in regular high-speed events, content structure detection, abnormal behavior identification, file restoration detection and other engines and NTA (Network Terminal Appliance) technology for protocol parsing, protocol decoding, file format detection, file restoration; high-performance network traffic collection and analysis are achieved through various methods such as bit stream matching and parallel protocol stack restoration. Specific functions include: data collection, policy management, event detection, etc. Specifically, the data security monitoring subsystem can perform basic abnormal data detection on content data according to the preset basic detection rule library to determine the basic abnormal status of content data.

[0101] Among them, the traffic behavior analysis and deep data mining subsystem can be understood as establishing multi-dimensional analysis behavior characteristics based on traditional keywords, fingerprints, regular expressions, dictionaries and other technologies, introducing artificial intelligence engine content recognition technology, and effectively recognizing data and its content to complete the corresponding identification and audit. The main traffic behaviors to be identified include data abnormal flow analysis, data abnormal cross-border analysis, abnormal data communication analysis, data suspected leakage analysis, abnormal email monitoring analysis, data abnormal access analysis, malicious code event analysis, attack intrusion event analysis, abnormal traffic event analysis, abnormal file outbound analysis, etc. Through deep data mining, it is used to define and monitor for a long time the behavior of various businesses and terminals of industrial Internet traffic in the large network. Specifically, the traffic behavior analysis and deep data mining subsystem can perform digital asset portrait processing on content data to obtain data packets to be transmitted including content data, and perform abnormal data packet detection on the data packets to be transmitted to obtain the abnormal data packet status of the data packets to be transmitted.

[0102] Among them, the data desensitization subsystem can be understood as deforming certain sensitive information through candidate desensitization strategies under the condition of retaining the original characteristics of the data, so as to achieve reliable protection of sensitive privacy data. Without violating the system rules, the real data is transformed and provided for testing. For example, personal key information needs to be desensitized. Only authorized administrators or users can access the true value of the data through specific applications and tools when they must know, thereby reducing the risk of important data when sharing and moving. Specifically, the data desensitization subsystem can be used to determine the traffic sensitivity, select the corresponding target desensitization strategy from the pre-built candidate desensitization strategy according to the traffic sensitivity, and desensitize the data packets to be transmitted according to the target desensitization strategy.

[0103] Exemplarily, the data security monitoring subsystem includes a data acquisition module, a policy management module, and an event monitoring module. Among them, the data acquisition module is mainly composed of dedicated acquisition equipment hardware and customized acquisition and analysis software built into the hardware. The data acquisition module serves as the cornerstone and source of the entire pipeline-side sensitive data transmission feature detection system. In the process of data acquisition design and implementation, the perfection of the data acquisition function is ensured through hardware design and software design. It mainly realizes protocol parsing, protocol decoding, file format detection and file restoration functions.

[0104] Specifically, the data acquisition equipment can be a full-flow acquisition / analysis / processing product with independent intellectual property rights. This equipment series adopts the X86 architecture (The X86 architecture, a computer language instruction set executed by a microprocessor), and has the characteristics of high throughput, good I / O performance, and strong analysis and processing capabilities. At the same time, it has excellent advantages in terms of equipment integration, energy consumption, and reliability. The data acquisition equipment is developed and designed based on NTA (Network Traffic Analysis) technology. Its main technical advantages are as follows: In terms of hardware, customized motherboards and network cards are used to improve the line-speed processing and forwarding capabilities, and reduce the performance loss caused by latency and program processing disturbances. At the same time, by optimizing the underlying embedded acquisition program, ultra-high-speed lock-free queue performance is achieved, thereby supporting single-queue high-frequency read and write operations. Security design has also been carried out at the hardware security level to achieve the construction of firmware protection and recovery, secure multi-threading, memory encryption and other capabilities.

[0105] The data collection service implementation logic may include: the data collection device NIC (Network Interface Card) module is mainly used to receive the original code stream sent; EAL is mainly used to initialize the operating environment of EVERZC, and provide a general interface for the upper-layer application EVERZC, hiding the relevant details of the interaction between the underlying library and the device, and supporting the automatic identification of the data of each logical interface in the system; EVERZC can save the copy of the data packet in the entire processing process, greatly improve the system processing performance, adopt the active packet receiving method, completely avoid the system call of hard interrupt and soft interrupt, and adopt the passive packet recovery method, which is conducive to the flexible design of multi-level processing framework; the memory pool performs preliminary decoding of the message, including decoding and distribution of TCP (Transmission Control Protocol) / UDP (User Datagram Protocol) / SCTP (Stream Control Transmission Protocol) / HTTP2 and other protocols, assembling the parsed data in the log table structure of the Linux kernel, and supporting reporting data to the platform through protocol interfaces such as SFTP; CPU (central processing unit) thread for 4 / 5G N11, S11, S1-U, and N3 interface messages are deeply parsed based on the corresponding protocols.

[0106] The data acquisition module in the embodiment of the present invention has a high-speed regular engine and protocol processing capabilities. Specifically, the high-speed regular matching engine is built based on the high-speed regular matching technology, which is a special matching technology designed for cross-message matching in network scenarios. Because in real network scenarios, features are scattered in multiple messages. If there are features in messages that have not yet arrived, the traditional matching mode will not be applicable. Therefore, for streaming messages, the high-speed regular matching technology can save the current data matching state and use it as the initial matching state when new data is received. Regardless of the form in which the data is scattered in the message blocks that arrive in time sequence, the high-speed regular matching technology ensures the consistency of the final matching results. The high-speed regular matching technology compresses the saved matching state to reduce the memory usage of streaming messages. The high-speed regular matching engine solves the data integrity problem and greatly simplifies the program processing process.

[0107] For example, see Figure 3BThe structural diagram of the high-speed regular matching engine is shown in FIG. Among them, the rule table is the rule content table, and the state table is the rule state table. The implementation principle of the high-speed regular matching engine is as follows: the rule set is compiled into a regular expression, and the compiled file is used as the input of the high-speed regular matching engine; the high-speed regular matching engine has a built-in rule content table and a rule state table, wherein the rule content table is used to temporarily save the single packet characteristics of the traffic; the rule state table is mainly used to save the temporal characteristics of the rules; the rule state table and the rule content table together constitute the rule base of the matching engine; directly associated with the rule content table is the rule matching module, which performs regular detection of streaming characteristics on multiple single packet characteristics with transmission data traffic in the rule content table based on high-speed regular matching technology; directly associated with the rule state table is the state table management module, which is responsible for maintaining and managing the status of the matching engine, including the state conversion and regular cleaning work in the rule state table; the decision module performs secondary feature verification on the data packets processed by the rule matching module to ensure the accuracy of the output data; the high-speed regular matching engine can flexibly adjust the matching behavior through the callback function, such as adjusting rule n to generate a match at displacement i.

[0108] The advantages of the high-speed regular matching engine in the embodiment of the present invention are as follows: it supports PCRE syntax, supports logical operations between different regular rules, has rich syntax keywords, supports AND / OR calculations between rules, supports logical operations between different regular rules, and supports weight calculation; supports custom syntax rules, and can form complex syntax logic; supports complex combinations of operations such as cross-packet detection, relative offset, and bit operations; integrates a high-performance regular processing engine, and the scanning rate can be linearly expanded according to the CPU; is compatible with the commonly used snort syntax rules and MTx syntax rules in the industry; supports matching of tens of thousands to hundreds of thousands of rules according to the complexity of the rules, and the single-core performance can achieve 3Gbps to 25Gbps; and supports synchronous matching of multiple rules; provides flow scanning mode and block scanning mode. In block mode, 20,000 rules are tested, and a single core can complete more than 3Gbps. The line-speed scanning of open source PCRE is only 300Mbps; in flow mode, 200 rules are tested, and a single core can complete a line-speed scan of more than 6G; the performance and the number of rules increase in a nonlinear relationship, the number of rules increases from 20,000 to 100,000, and the performance decreases little, and a single core can still reach 2.6Gbps; it has good scalability, and as the number of cores used increases, the matching performance is basically in a linear growth trend; it supports mutual verification of forward and reverse data scanning results; it supports parsing and field extraction of most common protocols; it supports flexible start and stop of rules, and can adaptively switch rules according to the actual environment to meet the needs of different scenarios; it supports loading a variety of regular processing engines, and the rules cover Internet protocol identification, Internet of Things protocol identification, industrial control protocol identification, Internet of Vehicles protocol identification, equipment asset fingerprint identification, APP application identification, network security event identification and many other aspects.

[0109] Exemplarily, the basic detection rule library of the data acquisition module may include protocol parsing rules, protocol decoding rules, file format detection rules and file restoration rules.

[0110] Among them, the protocol parsing rules can be used to identify the application protocol information of the data flow to be transmitted, and the application port rules, protocol feature rules, protocol regular feature rules and combination rules are used to match the content data in the data flow to be transmitted to determine the corresponding protocol of the data flow to be transmitted. According to the different requirements of the pipeline-side sensitive data transmission feature detection system for protocol accuracy, independent rules or combined rules can be used to improve the matching efficiency of the pipeline-side sensitive data transmission feature detection system as much as possible while meeting the requirements. After identifying the application protocol, the protocol information is recorded in the flow information space, and the data packet to be transmitted corresponding to the data flow to be transmitted directly identifies the protocol of the data packet to be transmitted according to the protocol in the flow information, and no repeated processing is required, thereby improving efficiency.

[0111] Among them, the protocol decoding rules can be used to parse the protocol variable information of the data flow to be transmitted. The call of the decoding work is executed according to the protocol of the data flow to be transmitted. There are two decoding methods. Based on the analysis of formatted data, such as the analysis of NETBIOS (Network Basic Input / Output System) protocol, the content of each area is extracted according to its fixed format; based on the data analysis of the identification language, such as the analysis of HTTP (HyperText Transfer Protocol) protocol, the corresponding key value is obtained by analyzing its keywords. The parsed data is placed in the decoding element array of the data packet, and other modules can easily obtain the corresponding data by directly accessing the decoding element array. For specific information, the data is directly saved in the flow information space, such as: URL (Uniform Resource Locator) address, email address, number of bytes of the transferred file, type of the transferred file, etc.

[0112] Among them, the file format detection rule can be used to use address information and file information at the same time to relatively accurately detect the file format. For uncoded plain text protocols, the file extension in the URL address and EMAIL (email) address information, as well as the file header feature structure of the transmitted file data, are used to identify the file format; for coded protocols, such as mail attachments, they are dynamically decoded, and then the file attachment content is identified. The identification method is mainly based on the type of file extension, the header bytes of the file, and the PE (Portable Executable) entry point of the file.

[0113] Among them, the file restoration rule can be used to directly cache network transmitted files according to user preset conditions. The user preset conditions include: the byte size of the file, the file format, the transmission direction, the type of application protocol transmitted, whether the content of the transmitted file is complete, etc. The file content that meets the conditions is saved to the local hard disk, and a reporting program is provided. According to the feature analysis of the data packet corresponding to the data flow to be transmitted, the features are matched, the file content in the data stream is cached, the file is restored, and the feature library is supplemented and improved.

[0114] Among them, the policy management module can perform local policy management. Local policy management can be used to receive policies issued by official agencies and synchronize them locally; support policy detail queries by policy type and policy name; display policy information that meets the query conditions, and view detailed information of a certain policy. The policy management module is a module for the pipeline-side sensitive data transmission feature detection system to interact with the policy instructions of the official agency platform. The policy management module can provide unified baseline policy rule management and behavioral policy rule management. The management functions include adding, executing, canceling and viewing details of policies, and clarifying the rule format for policy interaction with data collection, analysis and monitoring equipment. The main way for the pipeline-side sensitive data transmission feature detection system to generate data security events is to trigger event generation through policy matching after traffic analysis, and the policy management module is the unified management of the effectiveness and invalidation of these policies.

[0115] Among them, the event monitoring module performs routine data content detection and analysis based on the results of protocol decoding and file restoration, and generates monitoring events by matching monitoring strategies. The scope of the event monitoring module includes sensitive data monitoring and abnormal file monitoring. The main functions are as follows: Sensitive data monitoring is used to monitor whether there is sensitive data in the content data of the data flow to be transmitted; abnormal file monitoring refers to extracting the file name, MD5 value and fingerprint information, and configuring monitoring strategies to discover sensitive information and monitor abnormal files; event log generation refers to recording network access behaviors based on monitoring strategies and generating event logs.

[0116] Exemplarily, the event monitoring module can be used for feature library reception, feature rule processing and rule matching. Among them, feature library reception refers to the data acquisition device receiving the data security event monitoring feature library upgrade package through the feature library receiving module. The upgrade package can come from the equipment manufacturer, or it can be issued by the official agency's industrial Internet platform data security monitoring and service platform. Among them, feature rule matching data refers to the instruction receiving module of the acquisition device supporting compatibility with the MTX engine and the Snort engine. After the downloaded feature library upgrade package is decompressed, the standard MTX and Snort feature libraries can be directly loaded into the MTX and Snort monitoring engines. After the feature library is loaded, the feature library rules take effect directly. For single or combined instruction rules or instruction rules of some third-party regulatory platforms, the data acquisition device of the data security monitoring subsystem provides a powerful instruction rule conversion function to realize the parsing, conversion and recompilation of instruction rules, and then update the compiled rules to the data acquisition device rule engine. The rule engine supports the continuous operation of the original rules without interrupting the business, and realizes the hot loading of the full rule library function. Among them, rule matching means that after the data acquisition device loads the latest data security event monitoring feature rules, it uses pattern matching technology to match cross-message rules, detects the detected traffic, and executes the action corresponding to the feature when the feature rule is hit. Sensitive data monitoring mainly generates detailed records of data security event monitoring events in accordance with the data format requirements of technical specifications. Message monitoring mainly generates detailed records of data security event monitoring events in accordance with technical specifications, and returns them as log attachments based on the retained original traffic PCAP data packets. Abnormal file monitoring mainly generates abnormal files and detailed records of abnormal file monitoring events in accordance with technical specifications.

[0117] Exemplarily, the traffic behavior analysis and deep data mining subsystem may include a data asset portrait module, a traffic behavior analysis module, and a deep data mining module. Among them, the data asset portrait module can be used for data identification and data classification. Data identification may include data format identification, conventional data content detection, natural language processing, and machine learning automatic classification, as well as false positive and false negative comparison processing. Data classification may include classification and grading and data asset labeling.

[0118] Among them, data recognition can be understood as the effective recognition of data and its content through machine learning automatic classification, Chinese natural language processing, conventional content detection technology and false alarm vulnerability comparison classification advanced recognition detection technology to complete the corresponding recognition and audit. On the basis of traditional keywords, fingerprints, regular expressions, dictionaries and other technologies, the content recognition technology of artificial intelligence engines is introduced to improve recognition accuracy and recognition performance.

[0119] Among them, data format recognition can be used to analyze the data format to distinguish which data format or attribute the electronic data faced by the system is stored in. Based on file attribute recognition and data screening, after data format recognition, data that is not composed of text content will enter the system data classification recognition system using the relevant attributes of the data as a tag identification method. For data composed of text content, the system will perform data content analysis to obtain the text content.

[0120] Among them, conventional data content detection can be understood as the data after format recognition being divided into text content and non-text content, among which the text content data is further subjected to content analysis and is mainly divided into two main types: natural semantic type and non-natural semantic type.

[0121] Among them, natural language processing and machine learning automatic classification: Traditional conventional content recognition and detection technology has some limitations in terms of data security protection, especially the lack of effective detection of unstructured data. Based on this, the embodiment of the present invention innovatively introduces Chinese natural language processing technology to solve the problem of content recognition and classification. The embodiment of the present invention uses machine learning to have basic application capabilities such as text classification and clustering, information retrieval, etc. to identify content data. Exemplarily, the definition of important sensitive data includes: artificial intelligence content recognition based on machine learning and natural language processing; online help to comb sample data through automatic clustering of Chinese natural language; online learning sample semantic feature analysis; by analyzing a large number of unlabeled original document sets, automatically combing themes according to content, and manually intervening to flexibly adjust semantic similarity to obtain satisfactory clustering effects; using clustering results as labeled samples, extracting short sentences and compound words as semantic features, training classifiers, and automatically generating text classification rule bases, and manually intervening in feature selection, and using reverse samples to strengthen training.

[0122] Among them, false positive and false negative comparison processing: semantic features and regular expressions in conventional content detection methods have limitations in the accuracy of data content recognition. Based on this, the embodiment of the present invention introduces machine learning technology, and uses Chinese natural language processing technology and data mining technology to achieve the ability to quickly classify a large amount of data and extract features to establish a data classification model in a very short time, and realize the discovery and rule optimization of false positive and false negative samples through machine learning technology. The false positive and false negative comparison processing in the embodiment of the present invention can support the adjustment of recognition rules according to false positive samples to reduce the false positive rate; support the adjustment of recognition rules according to false negative samples to reduce the false negative rate.

[0123] Data classification can be understood as distinguishing different data values ​​through data identification. Relying on the identification of different data categories, the data in actual use is compared, and the classification (grading) affiliation of the detected data is determined based on the comparison results, and then the specific protection method of the data is determined based on the value difference of the classification (grading) categories.

[0124] Among them, classification and grading can be understood as automatically classifying and grading data assets identified by an intelligent sensitive data identification engine, supporting multiple sensitive data identification modes, including predefined modes, expert custom modes, and similar data discovery modes. Exemplarily, classification and grading may include data classification and data grading. Among them, data classification is the process of distinguishing and classifying data according to certain principles and methods based on the attributes or characteristics of the data, and establishing a certain classification system and arrangement order to better manage and use government data. The embodiment of the present invention provides different data classification rule methods for data discovery: the same file can belong to different categories at the same time according to different hit strategies; for network sensitive content data, in addition to business-related data, it can also be classified according to some general content judgment criteria. Among them, data classification can be understood as determining the data type by the sensitivity of the data, thereby providing support for the formulation of open and sharing strategies for different types of Internet data.

[0125] Among them, data asset tags can be understood as setting different tags for data assets to facilitate daily data asset retrieval and management, and administrators can update data asset tags. Specifically, the engine is built through the blood relationship between data tables and data fields, mainly to automatically analyze the data generation log and build the blood relationship between data assets. Through the data blood relationship, the source of the data (that is, the data sender) and the downstream user (that is, the data receiver) can be clearly identified.

[0126] For example, the traffic behavior analysis module is based on the deep analysis of data traffic, through data security rule collision analysis, data and file flow analysis, and realizes the analysis of various types of abnormal behaviors in data traffic. The traffic behavior analysis module has the functions of data abnormal flow analysis, data abnormal cross-border and joint analysis, data suspected leakage analysis, abnormal file outbound analysis, malicious code and intrusion event analysis, etc.

[0127] Among them, data abnormal flow analysis can be understood as monitoring data flow events, automatically restoring data access behavior, and recording in detail the access status of data interfaces, so as to achieve comprehensive monitoring of the details of important and sensitive data flowing through applications. Exemplarily, data abnormal flow analysis may include: parsing common types of data transmission formats and extracting and structuring core data; auditing transparent data access behavior; filtering data flow events, and extracting data flow events from all data flows. The pipeline-side sensitive data transmission feature detection system in the embodiment of the present invention uses UEBA (User and Entity Behavior Analytics) technology to establish a behavior baseline for the data access party, including a normal behavior baseline and a historical behavior baseline, and uses anomaly detection technology to identify abnormal data access behavior from multiple dimensions. With the ability to configure user-defined risk control rules, administrators can formulate customized risk control rules based on actual risk control needs to meet the needs of data flow risk monitoring and disposal in different scenarios; at the same time, the pipeline-side sensitive data transmission feature detection system has the function of user-defined key account and IP monitoring, providing a configuration entry for key accounts and IPs, and focusing on monitoring these accounts and IPs, such as real-time traffic statistics, traffic type, traffic direction, abnormal traffic, shared traffic and other functions, as well as the traffic usage of each IP or specific application, to achieve comprehensive perception and precise management of traffic changes.

[0128] Among them, data anomaly cross-border and communication analysis can be understood as the functions of realizing real-time abnormal data flow perception, cross-border unknown data identification, illegal cross-border data flow tracing and evidence collection, abnormal data extraction, etc. for the transmission data flow, discovering abnormal outbound data, and realizing early warning and real-time response. Through data flow perception, we analyze the multi-dimensional distribution information of cross-border data flow, such as type, quantity, interface, outbound purpose, geographical location of cross-border flow, and profile the cross-border data flow behavior. For the subjects who use data across borders, we conduct statistical summary analysis based on multiple dimensions such as the type of data accessed, the amount of data accessed, the time of data access, the method of data access, and the frequency of data access, to form a complete data access behavior portrait. The cross-border data and communication analysis in the embodiments of the present invention may include: through technologies such as sensitive data identification and sensitive data flow analysis, early detection, macro-warning and emergency response of cross-border transmission of sensitive data can be achieved; cross-border unknown data identification refers to monitoring whether the relevant data is transmitted across borders by comparing important sensitive data content fields (feature fingerprints) in the data flow to be transmitted; tracing the source of illegal cross-border data flow refers to active identification, evidence preservation, fusion analysis, etc. of target operation behaviors such as data system intrusion, destruction, attack, operation and maintenance services, external connections, etc.; comprehensive evidence collection of abnormal data.

[0129] Among them, the suspected data leakage analysis can be understood as the effective recognition of data and its content through machine learning automatic classification, Chinese natural language processing, conventional content detection technology and false alarm vulnerability comparison classification advanced recognition detection technology to complete the corresponding identification and audit. Cooperate with monitoring, early warning, auditing and other means to achieve leakage protection of designated data, and automatically respond based on policies and rules. In the embodiment of the present invention, when abnormal data flow such as data leakage and data abuse occurs, all collected clues are interactively and deeply analyzed and traced at the same time through the sensitive data transmission feature detection system on the pipeline side, and the data access link is restored to locate the source of the leakage.

[0130] Among them, the analysis of abnormal file outbound transmission can support rich file type recognition, multi-layer nested compressed file content, image content converted from documents, single large-capacity files without extensions or modified extensions, etc.; it can scan and monitor file operations, and issue alarms according to preset policies, and can scan and monitor files or content transmitted through terminal peripheral ports, and issue alarms according to preset policies; through user behavior modeling and abnormal behavior analysis, screenshot watermark protection, identification and monitoring of terminal device transmission and other functions, it supports diversified analysis and event analysis. The embodiment of the present invention conducts pre-control of outbound e-mails to prevent the leakage of confidential e-mail information; conducts post-audit and post-analysis of global e-mails to effectively reduce the occurrence of intranet leakage incidents; intuitive, detailed, and multi-level presentation of audit data reports effectively improves the overall manageability of the enterprise mail system.

[0131] Among them, malicious code and intrusion event analysis can be understood as monitoring, analyzing, alarming, logging, and network auditing of the transmitted data traffic, which can timely discover network security events such as virus infection, virus propagation, website backdoors, web page tampering, and malicious URL access within the network. In the embodiment of the present invention, malicious code and intrusion event analysis are used to confirm the effectiveness and accuracy of known threats, and the authenticity and certainty of unknown threats are obtained by analyzing the behavior trajectory. The pipeline-side sensitive data transmission feature detection system provides a statistical function, allowing security managers to directly know how many attack behaviors have occurred in the current period, how many targets and sources are associated with them, how many attack events have occurred, and other information; and the query uses IP-based, keyword-based and other technologies to quickly locate events that have occurred.

[0132] Among them, the deep data mining module can provide deep analysis functions such as data security incident association analysis, deep threat mining, risk assessment quantification, and security status management through big data analysis methods such as association analysis and deep learning, relying on dynamic baseline technology and predictive analysis technology. Data association analysis is the expansion of clues, which is a process of mining and exploring unknown data from limited data clues. In this process, data analysts need to repeatedly compare, query, and connect clues in various databases. The use of visual association analysis technology can make boring data analysis vivid in the form of graphical interface and smooth interactive operation, which can greatly improve the work efficiency of data analysts. In the selection of visualization technology, the characteristics of data association relationships can be highlighted, which is convenient for analysts to understand. At the same time, visualization auxiliary functions such as data statistics can help analysts understand the meaning of data and improve work efficiency. This deep data mining module designs an association analysis rule engine to realize the accelerated processing and logic optimization of CEP (complex event processing) technology and security business scenarios, realize the regular analysis of historical behavior in traffic, and improve the detection efficiency of industrial safety incidents. Massive data analysis requires a lot of resources and time. The relationship between single analysis can narrow the scope of analysis and make data security goals clearer.

[0133] Exemplarily, the deep data mining module may include rule-based association analysis, context-based association analysis, behavior-based association analysis, dynamic baseline technology, and predictive analysis technology, etc.

[0134] For example, if data is exported in the form of files, which has the characteristics of diversity and concealment, data files can be exported through online modes such as mailboxes, IM (Instant Messaging Technology), network cloud disks, or offline modes such as mobile storage media. Traditional DLP solutions can only cover data export management in some scenarios. Traditional DLP is powerless for cross-border transmission of network encrypted data and cross-border data of offline mobile storage media. The data file export risk perception module combines user active reporting with DLP perception. The former is mainly used to discover and reduce the unconscious risks of data exporters, and the latter is mainly used to detect and prevent some malicious behaviors. Since some important and sensitive data are stored in mobile computers or mobile storage media, it is necessary to perform a comprehensive automatic data asset scan for mobile computers and mobile storage media that are exported to discover sensitive and important data assets stored therein. In the embodiment of the present invention, important and sensitive data export events in the form of files are aggregated, and visualization technology is used to display multi-dimensional distribution information such as the type, quantity, interface, export purpose, and geographical location of cross-border data flow. It helps managers better understand the overall data flow and make corresponding management decisions.

[0135] In an embodiment of the present invention, a streaming computing framework is adopted for industrial Internet security data, and a streaming analysis engine based on memory-based complex event processing technology is used to perform real-time correlation analysis on the collected data. At the same time, a continuous aggregation engine is used to perform real-time analysis of real-time data based on machine learning. The algorithms used by the machine learning engine include learning algorithms based on behavioral profiles, continuous clustering analysis algorithms, continuous classification algorithms, indicator analysis methods, periodic and non-periodic abnormality analysis methods based on self-learning, etc.

[0136] The pipeline-side sensitive data transmission feature detection system in the embodiment of the present invention has the function of data flow perception, and uses visualization technology to display multi-dimensional distribution information such as the type, quantity, interface, outbound destination, and geographical location of cross-border data flow, so that managers can better understand the overall data flow situation and make corresponding management decisions.

[0137] Exemplarily, the data desensitizing subsystem includes a heterogeneous data source access module, a permission management module, a desensitizing algorithm management module, a template management module, a sensitive information identification module, a custom configuration module and a built-in desensitizing scenario module.

[0138] Among them, the heterogeneous data source access module is used to access the data source. For example, the data desensitization platform supports a wide range of data sources, not only supporting mainstream relational databases such as ORACLE, MYSQL, SQL SERVER, DB2, Informix, Sybase, but also supporting data warehouses such as Teradata, Greenplum, Gbase, formatted text files, HDP, StarRing TDH and other big data platforms, as well as special environments such as AS / 400.

[0139] Among them, the permission management module has a complete and unified permission management system, which can control the permissions of different users, different roles, and different data sources, and complete management functions such as user establishment, user allocation, and user identity authentication. It meets the requirements that all resource information of system users has the ability to be configurable and allocable at the smallest granularity, and ensures that it is allocated to specific users. Each user can only use the resources allocated to him.

[0140] Among them, the desensitizing algorithm management module has a wealth of desensitizing algorithms, including deterministic randomization, fuzzification, blanking, random arrangement, duplicate value masking, random replacement, specific rule replacement and other algorithms.

[0141] Among them, the template management module has a convenient template management function, which is convenient for maintaining sensitive information and corresponding desensitizing algorithms for different scenarios and different needs, and solving the system's requirements for unified configuration, flexibility, and simple operation.

[0142] Among them, the sensitive information identification module can automatically identify sensitive information distributed in different tables in the database system through automated technical means. It can identify sensitive content according to field names, field notes and other information, and can also identify sensitive information based on data feature matching.

[0143] The custom configuration module can be used to customize sensitive information. For example, in addition to automatically identifying sensitive information, managers or business department personnel can also manually customize a sensitive information list and import it into the desensitization system as a basis for desensitization.

[0144] Among them, common desensitization scenarios are built into the built-in desensitization scenario module to facilitate meeting the most common desensitization application scenarios. Exemplarily, the candidate desensitization strategies may include database-to-database desensitization strategies, database-to-file desensitization strategies, file-to-file desensitization strategies, file-to-database desensitization strategies, local desensitization and remote desensitization strategies. Among them, the database-to-database desensitization strategy can be used for the situation where the desensitized data source (i.e., the data sender) and the target source (data receiver) are both databases. The database-to-file desensitization strategy can be used for the situation where the desensitized data source is a database type, and the target environment must first be landed in a file format and then manually imported into the database. The file-to-file desensitization strategy can be used for the situation where the desensitized data source and the target source are both formatted file types. The file-to-database desensitization strategy can be used for the situation where the desensitized data source is formatted text and the target source is a database type. The local desensitization and remote desensitization strategies include local desensitization sub-strategies and remote desensitization sub-strategies. The local desensitization sub-strategy is suitable for the scenario where the desensitized data source and the target source are both implemented through a database; the remote desensitization sub-strategy is suitable for the situation where the desensitized results can be directly transmitted to the target database with a networked environment.

[0145] The pipeline-side sensitive data transmission feature detection system in the embodiment of the present invention collects and accesses front-end flow data, performs unified data parsing and analysis, performs sensitive data transmission feature detection and analysis, and performs data desensitizing flow behavior analysis, while providing a friendly front-end display interface to facilitate users and managers to analyze, process, and generate reports anytime and anywhere.

[0146] Exemplarily, the pipeline-side sensitive data transmission feature detection system consists of a data security monitoring subsystem, a traffic behavior analysis and deep data mining subsystem, and a data desensitizing subsystem, completing the overall process of data collection input-data analysis and mining-data desensitizing reporting, and conducting a full-scale analysis and mining of the transmission characteristics, traffic behavior, and sensitive rules of the Unicom pipeline-side data.

[0147] Among them, data collection input can be accessed through the collection of specified traffic data through the front-end, and high-performance collection, analysis, matching and monitoring of network traffic can be achieved by using the built-in data security monitoring engine, big data analysis and processing, and policy rule adaptation. Specific functional modules include: data collection module, event monitoring module, and policy management module.

[0148] Among them, data analysis and mining can be understood as establishing multi-dimensional analysis behavior characteristics based on traditional keywords, fingerprints, regular expressions, dictionaries and other technologies, introducing artificial intelligence engine content recognition technology, and effectively recognizing the data and its content to complete the corresponding identification and auditing.

[0149] The traffic behavior analysis and deep data mining subsystem in the embodiment of the present invention is composed of a traffic behavior analysis module and a deep data mining module. Among them, the traffic behavior analysis module may include at least one of abnormal data flow analysis, abnormal data cross-border analysis, abnormal data communication analysis, suspected data leakage analysis, abnormal email monitoring analysis, abnormal data access analysis, malicious code event analysis, attack intrusion event analysis, abnormal traffic event analysis and abnormal file outbound analysis. Through deep data mining, it is used to define and monitor for a long time the behavior of various businesses and terminals of industrial Internet traffic in the large network. Among them, the deep data mining module may include at least one of rule-based association analysis, context-based association analysis and behavior-based association analysis.

[0150] Among them, data desensitization reporting can be understood as the data desensitization subsystem deforming certain sensitive information through candidate desensitization strategies while retaining the original characteristics of the data, so as to achieve reliable protection of sensitive privacy data. Without violating system rules, real data is transformed and provided for testing. For example, personal key information needs to be desensitized. Only authorized administrators or users can access the true value of the data through specific applications and tools when they need to know, thereby reducing the risk of sharing and moving important data. For example, see Figure 3C The architecture diagram of the desensitization processing shown in the figure. Among them, the desensitization processing framework includes desensitization algorithms, desensitization strategies (i.e., candidate desensitization strategies) and data management. The desensitization algorithm may include replacement, disruption, encryption, blanking, numerical and date changes and other rules. Data management may include relational databases, data warehouses, big data platforms, files, preprocessing, data extraction, data synchronization, data conversion, data mapping and data monitoring. The desensitization execution engine may include a relational database execution engine, a big data platform execution engine, a file execution engine and a data warehouse data execution engine.

[0151] In the embodiment of the present invention, by collecting the industrial Internet flow data on the pipeline side, on the one hand, real-time flow monitoring and analysis are realized, providing real-time or near-real-time top-level flow analysis, security alarms, rule formulation, screening, and rapid attack warning and monitoring capabilities; on the other hand, based on the big data storage of business data streams, multi-dimensional behavior analysis features are established on this basis, and deep data mining of massive data is performed to define and monitor the behavior of various businesses and terminals of industrial Internet flow for a long time. Formulate industrial Internet data security strategies, and gradually form the capture of sensitive data and transmission behavior characteristics of the industry industrial Internet.

[0152] Embodiment 4

[0153] Figure 4 This is a schematic diagram of the structure of a data transmission device provided in Embodiment 4 of the present invention. This embodiment is applicable to the case of transmitting data, and the method can be executed by a data transmission device, which can be implemented in software and / or hardware and can be configured in an electronic device that carries a data transmission function.

[0154] like Figure 4 As shown, the device includes: a content data acquisition module 410, a basic abnormal state determination module 420, a data packet abnormal state determination module 430, a flow sensitivity determination module 440 and a data transmission module 450. Among them,

[0155] The content data acquisition module 410 is used to acquire the data flow to be transmitted, and acquire corresponding content data according to the data flow to be transmitted;

[0156] A basic abnormal state determination module 420, configured to perform a basic abnormal data detection on the content data according to a preset basic detection rule base, and determine a basic abnormal state of the content data;

[0157] The data packet abnormal state determination module 430 is used to perform digital asset profiling processing on the content data to obtain a data packet to be transmitted including the content data, and perform abnormal data packet detection on the data packet to be transmitted to obtain a data packet abnormal state of the data packet to be transmitted;

[0158] A flow sensitivity determination module 440, configured to determine the flow sensitivity of the data flow to be transmitted according to the basic abnormal state and the data packet abnormal state;

[0159] The data transmission module 450 is used to select a corresponding target desensitization strategy from pre-built candidate desensitization strategies according to the traffic sensitivity, and desensitize the data packet to be transmitted according to the target desensitization strategy to transmit the desensitized content data.

[0160] The embodiment of the present invention provides a data transmission scheme, which obtains the data flow to be transmitted, and obtains the corresponding content data according to the data flow to be transmitted; performs basic abnormal data detection on the content data according to a preset basic detection rule library to determine the basic abnormal state of the content data; performs digital asset portrait processing on the content data to obtain a data packet to be transmitted including the content data, and performs abnormal data packet detection on the data packet to be transmitted to obtain the data packet abnormal state of the data packet to be transmitted; determines the flow sensitivity of the data flow to be transmitted according to the basic abnormal state and the data packet abnormal state; selects the corresponding target desensitization strategy from the pre-constructed candidate desensitization strategy according to the flow sensitivity, and performs desensitization processing on the data packet to be transmitted according to the target desensitization strategy, so as to transmit the desensitized content data. The above scheme, by jointly determining the flow sensitivity of the data flow to be transmitted according to the abnormal data basic detection and the abnormal data packet detection, improves the accuracy of the determined flow sensitivity, thereby improving the accuracy of the target desensitization strategy determined according to the flow sensitivity, and desensitizing the data packet to be transmitted, and improves the security of data transmission; at the same time, in the embodiment of the present invention, the required data is obtained from the data flow to be transmitted, without obtaining it from the database, thereby improving the efficiency.

[0161] Optionally, the data packet abnormal state determination module 430 includes:

[0162] A transmission data packet generating unit, configured to perform digital asset profiling processing on the content data, obtain form data corresponding to the content data, and generate a data packet to be transmitted including the content data and the form data;

[0163] a deep abnormal state determining unit, configured to perform abnormal data depth detection on the content data to determine a deep abnormal state of the content data;

[0164] a form abnormality state determining unit, configured to perform form transmission detection on the form data according to a preset candidate form detection strategy, and determine a form abnormality state of the form data;

[0165] The data packet abnormal state determining unit is used to generate a data packet abnormal state including the deep abnormal state and the formal abnormal state.

[0166] Optionally, the abnormal form state determining unit includes:

[0167] a current form detection strategy determination subunit, configured to determine a current form detection strategy from the candidate form detection strategies;

[0168] a current form abnormal state determination subunit, configured to perform form transmission detection on the form data according to the current form detection strategy to obtain a current form abnormal state of the form data;

[0169] The form abnormality state determining subunit is used to determine the form abnormality state of the form data according to the current form abnormality state.

[0170] Optionally, the candidate form detection strategy includes at least one of a transmission rule detection strategy, a transmission context detection strategy, a transmission behavior detection strategy, a transmission baseline detection strategy and a transmission prediction detection strategy.

[0171] Optionally, if the current form detection strategy is a transmission behavior detection strategy, the current form abnormal state determination subunit is specifically used to:

[0172] Determining a data transmission context corresponding to the formal data according to a data initiator identifier and a data receiver identifier in the formal data;

[0173] Acquire historical behavior data in the data transmission scenario from a database, and determine historical access time according to the historical behavior data;

[0174] Determine the data access time from the form data, and perform behavior transmission detection on the form data according to the historical access time and the data access time to obtain the current form abnormal state of the form data

[0175] Optionally, the basic abnormal state determination module 420 includes:

[0176] A candidate detection rule determination unit, configured to determine a candidate detection rule for performing basic abnormal data detection on the content data according to a rule content table in a preset basic detection rule library;

[0177] A target detection rule determination unit, configured to determine the rule effective time of the candidate detection rule according to a rule status table in a preset basic detection rule library, and select a target detection rule from the candidate detection rules according to the rule effective time;

[0178] The basic abnormal state determination unit is used to perform a basic abnormal data detection on the content data according to the target detection rule to determine the basic abnormal state of the content data.

[0179] The data transmission device provided in the embodiment of the present invention can execute the data transmission method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing each data transmission method.

[0180] In the technical solution of the present invention, the collection, storage, use, processing, transmission, provision and disclosure of the data traffic to be transmitted, the basic detection rule library and the candidate form detection strategy, etc., all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0181] Embodiment 5

[0182] Figure 5 It is a structural diagram of a data transmission device provided by Embodiment 5 of the present invention. Electronic device 510 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0183] like Figure 5 As shown, the electronic device 510 includes at least one processor 511, and a memory connected to the at least one processor 511 in communication, such as a read-only memory (ROM) 512, a random access memory (RAM) 513, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 511 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 512 or the computer program loaded from the storage unit 518 to the random access memory (RAM) 513. In the RAM 513, various programs and data required for the operation of the electronic device 510 can also be stored. The processor 511, the ROM 512, and the RAM 513 are connected to each other via a bus 514. An input / output (I / O) interface 515 is also connected to the bus 514.

[0184] A number of components in the electronic device 510 are connected to the I / O interface 515, including: an input unit 516, such as a keyboard, a mouse, etc.; an output unit 517, such as various types of displays, speakers, etc.; a storage unit 518, such as a disk, an optical disk, etc.; and a communication unit 519, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 519 allows the electronic device 510 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0185] The processor 511 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 511 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 511 performs the various methods and processes described above, such as a data transmission method.

[0186] In some embodiments, the data transmission method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 518. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 510 via the ROM 512 and / or the communication unit 519. When the computer program is loaded into the RAM 513 and executed by the processor 511, one or more steps of the data transmission method described above may be performed. Alternatively, in other embodiments, the processor 511 may be configured to perform the data transmission method in any other appropriate manner (e.g., by means of firmware).

[0187] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0188] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0189] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0190] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0191] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0192] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0193] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0194] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A data transmission method, characterized in that: include: Acquire the data flow to be transmitted, and acquire corresponding content data according to the data flow to be transmitted; According to a preset basic detection rule library, performing basic abnormal data detection on the content data to determine a basic abnormal state of the content data; Performing digital asset profiling processing on the content data to obtain a data packet to be transmitted including the content data, and performing abnormal data packet detection on the data packet to be transmitted to obtain a data packet abnormal state of the data packet to be transmitted; Determining the flow sensitivity of the data flow to be transmitted according to the basic abnormal state and the data packet abnormal state; According to the traffic sensitivity, a corresponding target desensitization strategy is selected from pre-constructed candidate desensitization strategies, and the data packet to be transmitted is desensitized according to the target desensitization strategy to transmit the desensitized content data.

2. The method according to claim 1, characterized in that The performing digital asset portrait processing on the content data to obtain a data packet to be transmitted including the content data, and performing abnormal data detection on the data packet to be transmitted to obtain a data packet abnormal state of the data packet to be transmitted, includes: Performing digital asset profiling on the content data to obtain form data corresponding to the content data, and generating a data packet to be transmitted including the content data and the form data; Performing a deep abnormal data detection on the content data to determine a deep abnormal state of the content data; According to a preset candidate form detection strategy, performing form transmission detection on the form data to determine a form abnormality state of the form data; A packet anomaly status including the deep anomaly status and the formal anomaly status is generated.

3. The method according to claim 2, characterized in that The step of performing a form transmission detection on the form data according to a preset candidate form detection strategy to determine the form abnormality state of the form data includes: determining a current form detection strategy from the candidate form detection strategies; According to the current form detection strategy, performing form transmission detection on the form data to obtain a current form abnormal state of the form data; According to the current form abnormality state, the form abnormality state of the form data is determined.

4. The method according to claim 3, characterized in that The candidate form detection strategy includes at least one of a transmission rule detection strategy, a transmission context detection strategy, a transmission behavior detection strategy, a transmission baseline detection strategy and a transmission prediction detection strategy.

5. The method according to claim 4, characterized in that If the current form detection strategy is a transmission behavior detection strategy, then performing form transmission detection on the form data according to the current form detection strategy to obtain the current form abnormal state of the form data includes: Determining a data transmission context corresponding to the formal data according to a data initiator identifier and a data receiver identifier in the formal data; Acquire historical behavior data in the data transmission scenario from a database, and determine historical access time according to the historical behavior data; The data access time is determined from the form data, and a behavior transmission detection is performed on the form data according to the historical access time and the data access time to obtain a current form abnormal state of the form data.

6. The method according to any one of claims 1 to 5, characterized in that The step of performing a basic abnormal data detection on the content data according to a preset basic detection rule base to determine a basic abnormal state of the content data includes: Determine, according to the rule content table in the preset basic detection rule library, candidate detection rules for performing basic detection of abnormal data on the content data; Determine the rule effective time of the candidate detection rule according to the rule status table in the preset basic detection rule library, and select the target detection rule from the candidate detection rules according to the rule effective time; According to the target detection rule, a basic abnormal data detection is performed on the content data to determine a basic abnormal state of the content data.

7. A data transmission device, characterized in that: include: A content data acquisition module, used to acquire the data flow to be transmitted, and acquire corresponding content data according to the data flow to be transmitted; A basic abnormal state determination module, configured to perform a basic abnormal data detection on the content data according to a preset basic detection rule base, and determine a basic abnormal state of the content data; A data packet abnormal state determination module is used to perform digital asset profiling processing on the content data to obtain a data packet to be transmitted including the content data, and perform abnormal data packet detection on the data packet to be transmitted to obtain a data packet abnormal state of the data packet to be transmitted; A flow sensitivity determination module, used to determine the flow sensitivity of the data flow to be transmitted according to the basic abnormal state and the data packet abnormal state; The data transmission module is used to select a corresponding target desensitization strategy from pre-built candidate desensitization strategies according to the traffic sensitivity, and desensitize the data packet to be transmitted according to the target desensitization strategy to transmit the desensitized content data.

8. The device according to claim 7, characterized in that The data packet abnormal state determination module includes: A transmission data packet generating unit, configured to perform digital asset profiling processing on the content data, obtain form data corresponding to the content data, and generate a data packet to be transmitted including the content data and the form data; a deep abnormal state determining unit, configured to perform abnormal data depth detection on the content data to determine a deep abnormal state of the content data; a form abnormality state determining unit, configured to perform form transmission detection on the form data according to a preset candidate form detection strategy, and determine a form abnormality state of the form data; The data packet abnormal state determining unit is used to generate a data packet abnormal state including the deep abnormal state and the formal abnormal state.

9. An electronic device, characterized in that: include: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement a data transmission method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, a data transmission method according to any one of claims 1 to 6 is implemented.

Citation Information

Cited By

  • Data outbound subject management and control architecture based on tree structure

    CN120851373A

  • Data desensitization method, device and equipment based on large model and storage medium

    CN120874127A