A cloud platform data encryption transmission system and method

By providing a data encryption transmission system with encrypted data identification, encryption, check and transmission feedback modules on the cloud platform, the problem of low encryption transmission efficiency in the prior art is solved, and efficient and secure data transmission is achieved.

CN118300880BActive Publication Date: 2024-11-29BEIJING CHINASOFT LINKAGE TECHNOLOGY CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202410539116.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-30
Publication Date
2024-11-29
Estimated Expiration
2044-04-30

AI Technical Summary

Technical Problem

In the prior art, no encrypted data is identified, resulting in low efficiency in encrypted transmission of network information.

Method used

Provide a cloud platform data encryption transmission system, including an encrypted data identification module, a data encryption module, a data transmission module, a data decryption module and a transmission feedback module. The data is statically and dynamically identified through the encrypted data identification module, generated the A-standard data transmission key, encrypted and verified the data, and adjusted the optimization process through the transmission feedback module to improve the encryption transmission efficiency.

Benefits of technology

By static and dynamic identification of data, the efficiency of encrypted data recognition is improved; through data encryption and verification, the integrity and security of data during transmission is ensured; through adjustment of the transmission feedback module, the efficiency of encrypted transmission is further improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118300880B_ABST
    Figure CN118300880B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of data information transmission, and in particular to a cloud platform data encryption transmission system and method thereof, including an encrypted data identification module for statically identifying data to obtain A-label data, and also for dynamically identifying data and supplementing A-label data, and also for optimizing dynamic identification results and optimizing large language models; a data encryption module for generating A-label data transmission keys, and also for data partitioning and generating A-label data vectors, and also for data encryption, data verification, and data encapsulation; a data transmission module for data transmission; a data decryption module for data integrity verification and decryption; and a transmission feedback module for adjusting large language model optimization processes and data verification processes. The present invention realizes automatic identification and transmission of network information encrypted data, and improves encryption transmission efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of data information transmission, and in particular to a cloud platform data encryption transmission system and method thereof. Background Art

[0002] When users upload data to the cloud platform for sharing through terminals, the encrypted data in the data is identified and securely transmitted to ensure that the data is not obtained or tampered with by unauthorized third parties during the transmission process. This is especially important for data containing sensitive information such as patient information, financial data, etc., and can effectively prevent data leakage and abuse. Many industries have specific regulations and compliance requirements involving data protection and privacy. Encrypting data and transmitting it in a secure environment can help companies meet these requirements and avoid facing fines and legal proceedings. The security of data is directly related to the business continuity of the company. If the data is tampered with or lost during transmission, it may cause business interruption, resulting in significant losses and reputation issues. Through encrypted data identification and secure transmission, the integrity and availability of the data can be ensured, thereby maintaining the business continuity of the company. On the cloud platform, data sharing and collaboration are required between different users or organizations. Through encrypted data identification and secure transmission, the security and privacy of data can be ensured during the sharing process, thereby promoting digital collaboration and win-win results.

[0003] Chinese patent publication number: CN116471007B discloses a network information encryption transmission method based on a cloud platform, including: obtaining data to be encrypted, splitting to obtain split value combinations, obtaining the feasibility of the split value combinations according to the Hamming distance between the values ​​in the sliding window and the split value combinations, counting the character frequencies of the split combinations, constructing a Huffman tree, obtaining an adaptively assigned Huffman tree according to the number of binary tree layers in the Huffman tree, interactively assigning binary tree codes in the adaptively assigned Huffman tree according to a first chaotic sequence, sorting the leaf nodes in the adaptively assigned Huffman tree according to a second chaotic sequence, obtaining a transformed Huffman tree, encoding according to the adaptively assigned Huffman tree and the transformed Huffman tree to obtain encrypted data, and decrypting the encrypted data according to a key. However, the scheme only encrypts and transmits network information by changing the character statistical frequency in the data and only constructing different types of Huffman trees for encoding conversion, without identifying the data to be encrypted, and cannot realize automatic identification and transmission of encrypted data of network information, and the efficiency of encrypted transmission of network information is low. Summary of the invention

[0004] To this end, the present invention provides a cloud platform data encryption transmission system and method thereof, so as to overcome the problem of low efficiency of network information encryption transmission caused by failure to identify the encrypted data in the prior art and inability to realize automatic identification and transmission of network information encryption data.

[0005] To achieve the above objectives, on the one hand, the present invention provides a cloud platform data encryption transmission system, comprising:

[0006] The encrypted data recognition module is used to statically recognize the data and obtain the A-label data, and is also used to dynamically recognize the data according to the large language model and supplement the A-label data according to the dynamic recognition results, and is also used to obtain manually marked A-label data, and optimize the dynamic recognition results according to the manually marked A-label data, and generate a model optimization database to optimize the large language model;

[0007] A data encryption module is used to generate an A-label data transmission key, to divide the A-label data and generate an A-label data vector, to encrypt the divided A-label data, to verify the encrypted A-label data, and to encapsulate the A-label data that has been successfully verified;

[0008] A data transmission module, used for transmitting the A-label data after data encapsulation;

[0009] A data decryption module is used to verify the data integrity of the encrypted A-label data and decrypt it when the encrypted A-label data is complete;

[0010] The transmission feedback module is used to judge the encryption transmission efficiency according to the amount of data that fails to be transmitted, and adjust the large language model optimization process and data verification process according to the judgment result.

[0011] Furthermore, the encrypted data recognition module obtains the character data at the preset position, collects the residence time T of the character data at the preset position, compares the residence time T with the preset residence time T0, and performs static recognition on the data according to the comparison result, wherein:

[0012] When T>T0, the encrypted data recognition module determines that the static recognition of the character data is successful, and marks it as A-labeled data;

[0013] When T≤T0, the encrypted data recognition module determines that static recognition of the character data fails.

[0014] Furthermore, the encrypted data recognition module establishes a large language model based on the encrypted data vocabulary, preprocesses the data, and inputs the preprocessed data into the large language model to obtain a text segment containing the encrypted data, which is supplemented as A-label data.

[0015] Furthermore, the encrypted data identification module obtains manually marked A-marked data, sets the newly added A-marked data as a text segment containing encrypted data, and deletes the deleted A-marked data from the text segment containing encrypted data.

[0016] Furthermore, the encrypted data recognition module obtains the number of manual marking times u, compares the number of manual marking times u with the preset number of manual marking times u0, and optimizes the large language model according to the comparison result, wherein:

[0017] When u≤u0, the encrypted data recognition module does not optimize the large language model;

[0018] When u>u0, the encrypted data recognition module optimizes the large language model, and uses the manually marked A-labeled data as a model optimization database, and adds it to the encrypted data vocabulary to optimize the large language model.

[0019] Furthermore, the data encryption module generates an A-label data transmission key using a gpg command and stores it in an HSM hardware security module;

[0020] The data encryption module uses preset padding characters to fill the A-label data to an integer multiple of 128 bytes, divides the filled A-label data, uses Poly1305 to generate a MAC for each data block obtained after the data division, and uses a strong random number generator to generate an A-label data vector;

[0021] The data encryption module uses the A-label data transmission key and the A-label data vector to encrypt the divided A-label data through the ChaCha20 algorithm, and appends the MAC generated by each data block to the encrypted data block to obtain the MAC appended to the data block.

[0022] Furthermore, the data encryption module presets the sampled data volume p0 at each interval, uses the A-label data transmission key, calculates the MAC of the encrypted A-label data through the Poly1305 algorithm, and compares the calculated MAC of the encrypted A-label data with the MAC attached to the data block, and performs data verification on the encrypted A-label data according to the comparison result, wherein:

[0023] When the calculated MAC of the encrypted A-label data is consistent with the MAC attached to the data block, the data encryption module determines that the encrypted A-label data verification is successful;

[0024] When the calculated MAC of the encrypted A-label data is inconsistent with the MAC attached to the data block, the data encryption module determines that the encrypted A-label data verification fails;

[0025] The data encryption module performs data encapsulation on the encrypted A-label data of the successfully verified A-label data, the MAC attached to the data block, and the A-label data vector to obtain encapsulated A-label data;

[0026] The data transmission module uses the transport layer security protocol TLS to transmit the A-label data after data encapsulation.

[0027] Furthermore, the data decryption module uses the A-label data transmission key to calculate the MAC of the encrypted A-label data through the Poly1305 algorithm, and compares the calculated MAC of the encrypted A-label data with the MAC attached to the data block, and performs data integrity verification on the encrypted A-label data according to the comparison result, wherein:

[0028] When the calculated MAC of the encrypted A-label data is consistent with the MAC attached to the data block, the data decryption module determines that the encrypted A-label data is complete, and uses the A-label data transmission key and the A-label data vector to decrypt the encrypted A-label data through the ChaCha20 algorithm;

[0029] When the calculated MAC of the encrypted A-label data is inconsistent with the MAC attached to the data block, the data decryption module determines that the encrypted A-label data is incomplete.

[0030] Furthermore, the transmission feedback module compares the amount of transmission failure data n within the feedback cycle with the preset amount of transmission failure data n0, and determines the encryption transmission efficiency according to the comparison result, wherein:

[0031] When n≤n0, the transmission feedback module determines that the encryption transmission efficiency is normal;

[0032] When n>n0, the transmission feedback module determines that the encryption transmission efficiency is abnormal;

[0033] When the encryption transmission efficiency is abnormal, the transmission feedback module adjusts the preset number of manual marking times u0 in the large language model optimization process, sets the adjusted preset number of manual marking times as un0, and sets un0=ni×u0;

[0034] When the encryption transmission efficiency is abnormal, the transmission feedback module adjusts the preset sampling data volume p0 in the data verification process, and the adjusted preset sampling data volume is pn0, and pn0=ni×p0 is set.

[0035] On the other hand, the present invention also provides a cloud platform data encryption transmission method, comprising:

[0036] Step S1, statically identify the data through the encrypted data identification module to obtain A-label data;

[0037] Step S2, dynamically identifying the data according to the large language model through the encrypted data identification module, and supplementing the A-label data according to the dynamic identification result;

[0038] Step S3, obtaining manually marked A-label data through the encrypted data recognition module, optimizing the dynamic recognition result according to the manually marked A-label data, and generating a model optimization database to optimize the large language model;

[0039] Step S4, generating a data transmission key for label A through a data encryption module;

[0040] Step S5, dividing the A-label data by a data encryption module and generating an A-label data vector;

[0041] Step S6, encrypting the divided A-label data through a data encryption module, verifying the encrypted A-label data, and encapsulating the A-label data that has been successfully verified;

[0042] Step S7, transmitting the A-label data after data encapsulation through the data transmission module;

[0043] Step S8, verifying the data integrity of the encrypted A-label data through a data decryption module, and decrypting the encrypted A-label data when it is complete;

[0044] Step S9, judging the encryption transmission efficiency according to the amount of transmission failure data through the transmission feedback module, and adjusting the large language model optimization process and the data verification process according to the judgment result.

[0045] Compared with the prior art, the beneficial effects of the present invention are that the system statically identifies the data through the encrypted data identification module to obtain the A-label data to improve the efficiency of encrypted data identification, the system dynamically identifies the data according to the large language model through the encrypted data identification module, and supplements the A-label data according to the dynamic identification result, so as to further improve the accuracy of A-label data identification, the system obtains the manually marked A-label data through the encrypted data identification module, optimizes the dynamic identification result according to the manually marked A-label data, and generates a model optimization database to optimize the large language model, so as to further improve the accuracy of A-label data identification, the system generates an A-label data transmission key through the data encryption module, and the system divides the A-label data through the data encryption module and generates an A-label data vector to prevent replay attacks hit, improve the security of encrypted transmission, the system encrypts the divided A-label data through the data encryption module, and verifies the encrypted A-label data, and encapsulates the A-label data that has successfully been verified to verify the integrity of the A-label data and ensure that the A-label data has not been tampered with during the transmission process. The system transmits the A-label data after data encapsulation through the data transmission module, and the system verifies the data integrity of the encrypted A-label data through the data decryption module, and decrypts the encrypted A-label data when it is complete to ensure the integrity of the data during the encrypted transmission process and improve the efficiency of encrypted transmission. The system judges the encryption transmission efficiency according to the amount of transmission failure data through the transmission feedback module, and adjusts the large language model optimization process and the data verification process according to the judgment result to further improve the encryption transmission efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a schematic diagram of the structure of the cloud platform data encryption transmission system of this embodiment;

[0047] Figure 2 The figure is a flow chart of the cloud platform data encryption transmission method of this embodiment. DETAILED DESCRIPTION

[0048] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0049] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0050] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0051] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0052] See also Figure 1 As shown, it is a schematic diagram of the structure of the cloud platform data encryption transmission system of this embodiment, and the system includes:

[0053] The encrypted data recognition module is used to statically recognize the data and obtain the A-label data, and is also used to dynamically recognize the data according to the large language model and supplement the A-label data according to the dynamic recognition results, and is also used to obtain manually marked A-label data, and optimize the dynamic recognition results according to the manually marked A-label data, and generate a model optimization database to optimize the large language model;

[0054] A data encryption module is used to generate an A-label data transmission key, to divide the A-label data and generate an A-label data vector, to encrypt the divided A-label data, to verify the encrypted A-label data, and to encapsulate the A-label data that has been successfully verified;

[0055] A data transmission module, used for transmitting the A-label data after data encapsulation;

[0056] A data decryption module is used to verify the data integrity of the encrypted A-label data and decrypt it when the encrypted A-label data is complete;

[0057] The transmission feedback module is used to judge the encryption transmission efficiency according to the amount of data that fails to be transmitted, and adjust the large language model optimization process and data verification process according to the judgment result.

[0058] Specifically, the system is applied to a data sharing cloud platform. When a user uploads data to the cloud platform for sharing through a terminal, the encrypted data in the data is identified and transmitted securely to achieve data sharing collaboration while ensuring the security of the encrypted data, thereby improving the efficiency of encrypted transmission of network information. The system statically identifies the data through an encrypted data identification module to obtain A-label data to improve the efficiency of encrypted data identification. The system dynamically identifies the data according to a large language model through an encrypted data identification module, and supplements the A-label data according to the dynamic identification result to further improve the accuracy of A-label data identification. The system obtains manually marked A-label data through an encrypted data identification module, optimizes the dynamic identification result according to the manually marked A-label data, and generates a model optimization database to optimize the large language model to further improve the accuracy of A-label data identification. The system generates A-label data transmission through a data encryption module Key, the system divides the A-label data through the data encryption module and generates the A-label data vector to prevent replay attacks and improve the security of encrypted transmission. The system encrypts the divided A-label data through the data encryption module, and performs data verification on the encrypted A-label data, and encapsulates the A-label data that succeeds in verification to verify the integrity of the A-label data and ensure that the A-label data has not been tampered with during the transmission process. The system transmits the A-label data after data encapsulation through the data transmission module. The system verifies the data integrity of the encrypted A-label data through the data decryption module, and decrypts the encrypted A-label data when it is complete to ensure the integrity of the data during the encrypted transmission process and improve the efficiency of encrypted transmission. The system judges the encryption transmission efficiency according to the amount of transmission failure data through the transmission feedback module, and adjusts the large language model optimization process and the data verification process according to the judgment result to further improve the encryption transmission efficiency.

[0059] Specifically, the encrypted data recognition module obtains the character data at the preset position, collects the residence time T of the character data at the preset position, compares the residence time T with the preset residence time T0, and performs static recognition on the data according to the comparison result, wherein:

[0060] When T>T0, the encrypted data recognition module determines that the static recognition of the character data is successful, and marks it as A-labeled data;

[0061] When T≤T0, the encrypted data recognition module determines that static recognition of the character data fails.

[0062] Specifically, the preset position refers to the preset position of a specific storage bucket, database, file or part of a cloud service where encrypted data can be obtained. This embodiment does not limit the setting method of the preset position. Those skilled in the art can freely set it according to actual conditions, such as by manual marking. The preset position can also be set by big data analysis. Taking big data analysis as an example, the encrypted data recognition module can use the high-frequency occurrence position of the A-labeled data as the preset position. The character data refers to data containing digital or text content. This embodiment does not limit the reading method of the character data. Those skilled in the art can freely set it according to actual conditions, as long as the requirements for the number of characters are met. The reading requirements of the data can be met, such as the SDK software development tool kit and the API application program interface can be set to read the character data, the dwell time refers to the duration of time that the character data remains unchanged at the preset position, the present embodiment does not limit the method for obtaining the dwell time, and technical personnel in this field can freely set it according to actual needs, and only need to meet the acquisition requirements of the dwell time of the character data at the preset position, such as the built-in clock can be set to obtain the dwell time, the preset dwell time refers to the preset value of the duration that reflects the successful static recognition of the character data, such as the preset dwell time T0=5s can be set, and the static recognition refers to the method of recognizing the A-mark data according to the preset position.

[0063] Specifically, the encrypted data recognition module performs static recognition on the data to perform encrypted data recognition on the character data at a preset position, thereby improving recognition efficiency.

[0064] Specifically, the encrypted data recognition module establishes a large language model according to the encrypted data vocabulary, preprocesses the data, and inputs the preprocessed data into the large language model to obtain a text segment containing the encrypted data, and supplements it as A-label data.

[0065] Specifically, the encrypted data vocabulary refers to a private data dictionary including personal information, health data, financial information, passwords, etc. for subsequent data matching and screening. The present embodiment does not limit the matching process between the large language model and the encrypted data vocabulary. Those skilled in the art can set it according to actual conditions, and only need to meet the large language model's text segment recognition requirements for encrypted data. For example, the text verification data set can be set to be input into the large language model that has matched the encrypted data vocabulary, and the amount of data with successful text verification is obtained, the accuracy is calculated, and the large language model with an accuracy greater than the preset accuracy is output, the encrypted database of the large language model with an accuracy less than or equal to the preset accuracy is corrected, the text verification data set that fails the text verification is divided into encrypted data segments, and the encrypted data segments of the divided text verification data set are added to the encrypted database until the accuracy is greater than the preset accuracy. The large language model is output. The present embodiment does not limit the data preprocessing method, and those skilled in the art can freely set it according to actual conditions, and only need to meet the data preprocessing requirements, such as the data can be set to be cleaned and formatted.

[0066] Specifically, the encrypted data recognition module preprocesses the data to remove irrelevant characters and noise. The encrypted data recognition module inputs the preprocessed data into a large language model to obtain a text segment containing encrypted data, and supplements it with A-label data to further improve the accuracy of A-label data recognition.

[0067] Specifically, the encrypted data identification module obtains manually marked A-marked data, sets the newly added A-marked data as a text segment containing encrypted data, and deletes the deleted A-marked data from the text segment containing encrypted data.

[0068] Specifically, the manually marked A-labeled data refers to encrypted data manually marked by a user, including newly added A-labeled data and deleted A-labeled data.

[0069] Specifically, the encrypted data recognition module adds and deletes text segments containing encrypted data according to the manually marked A-label data to optimize the A-label data recognition and further improve the accuracy of the A-label data recognition.

[0070] Specifically, the encrypted data recognition module obtains the number of manual marking times u, compares the number of manual marking times u with the preset number of manual marking times u0, and optimizes the large language model according to the comparison result, wherein:

[0071] When u≤u0, the encrypted data recognition module does not optimize the large language model;

[0072] When u>u0, the encrypted data recognition module optimizes the large language model, and uses the manually marked A-labeled data as a model optimization database, and adds it to the encrypted data vocabulary to optimize the large language model.

[0073] Specifically, the encrypted data recognition module optimizes the large language model, uses the manually marked A-labeled data as a model optimization database, adds it to the encrypted data vocabulary to optimize the large language model, so as to improve the accuracy of the large language model recognition, thereby improving the recognition efficiency.

[0074] Specifically, the data encryption module uses the gpg command to generate an A-label data transmission key and stores it in the HSM hardware security module.

[0075] Specifically, the gpg is the abbreviation of GNU Privacy Guard, which is an encryption software used for data encryption, decryption, signing and verification. The A-standard data transmission key refers to a key used for data encryption and data verification. The HSM hardware security module refers to a hardware device used to securely manage and store keys.

[0076] Specifically, the data encryption module uses the gpg command to generate an A-label data transmission key to generate a random strong password. The data encryption module stores the A-label data transmission key in the HSM hardware security module for secure storage.

[0077] Specifically, the data encryption module uses preset filling characters to fill the A-label data to an integer multiple of 128 bytes, divides the filled A-label data, uses Poly1305 to generate a MAC for each data block obtained after data division, and uses a strong random number generator to generate an A-label data vector.

[0078] Specifically, the preset filling character refers to a specific filling character used to fill the remaining part of the A-label data block, such as 0x00. The Poly1305 refers to a message authentication code algorithm used in combination with the ChaCha20 encryption algorithm to provide data integrity and authenticity verification. MAC refers to a message authentication code of a security mechanism used to ensure the integrity and consistency of information transmission. The generated A-label data vector refers to a random or non-repetitive initialization vector. This embodiment does not limit the type of strong random number generator, and those skilled in the art can freely set it according to actual conditions, such as setting it to CryptGenRandom.

[0079] Specifically, the data encryption module divides the A-label data into blocks or packets of appropriate sizes to facilitate subsequent encryption processing, and the data encryption module generates an A-label data vector to prevent replay attacks.

[0080] Specifically, the data encryption module uses the A-label data transmission key and the A-label data vector to encrypt the divided A-label data through the ChaCha20 algorithm, and appends the MAC generated by each data block to the encrypted data block to obtain the MAC attached to the data block.

[0081] Specifically, the ChaCha20 algorithm refers to a stream cipher algorithm using a one-time counter or an initialization vector IV.

[0082] Specifically, the data encryption module encrypts the divided A-label data through the ChaCha20 algorithm to ensure the security of the A-label data through the round function and permutation operation of ChaCha20.

[0083] Specifically, the data encryption module presets the sampled data volume p0 at each interval, uses the A-label data transmission key, calculates the MAC of the encrypted A-label data through the Poly1305 algorithm, and compares the calculated MAC of the encrypted A-label data with the MAC attached to the data block, and performs data verification on the encrypted A-label data according to the comparison result, wherein:

[0084] When the calculated MAC of the encrypted A-label data is consistent with the MAC attached to the data block, the data encryption module determines that the encrypted A-label data verification is successful;

[0085] When the calculated MAC of the encrypted A-label data is inconsistent with the MAC attached to the data block, the data encryption module determines that the encrypted A-label data verification fails.

[0086] Specifically, the preset sampling data volume refers to the preset sampling interval for data verification of the encrypted A-label data. The preset sampling data volume is set according to the total amount of A-label data. When the total amount of A-label data is small, the preset sampling data volume is reduced. When the total amount of A-label data is large, the preset sampling data volume is increased.

[0087] Specifically, the data encryption module performs data verification on the encrypted A-label data to verify the integrity of the A-label data and ensure that the A-label data has not been tampered with during transmission.

[0088] Specifically, the data encryption module encapsulates the encrypted A-label data of the successfully verified A-label data, the MAC attached to the data block, and the A-label data vector to obtain the encapsulated A-label data.

[0089] Specifically, the data transmission module uses the transport layer security protocol TLS to transmit the A-label data after data encapsulation.

[0090] Specifically, the transport layer security protocol TLS refers to a communication security protocol that provides privacy and data integrity in Internet communications.

[0091] Specifically, the data decryption module uses the A-label data transmission key to calculate the MAC of the encrypted A-label data through the Poly1305 algorithm, and compares the calculated MAC of the encrypted A-label data with the MAC attached to the data block, and performs data integrity verification on the encrypted A-label data according to the comparison result, where:

[0092] When the calculated MAC of the encrypted A-label data is consistent with the MAC attached to the data block, the data decryption module determines that the encrypted A-label data is complete, and uses the A-label data transmission key and the A-label data vector to decrypt the encrypted A-label data through the ChaCha20 algorithm;

[0093] When the calculated MAC of the encrypted A-label data is inconsistent with the MAC attached to the data block, the data decryption module determines that the encrypted A-label data is incomplete.

[0094] Specifically, the transmission feedback module compares the amount of transmission failure data n within the feedback cycle with the preset amount of transmission failure data n0, and determines the encryption transmission efficiency according to the comparison result, wherein:

[0095] When n≤n0, the transmission feedback module determines that the encryption transmission efficiency is normal;

[0096] When n>n0, the transmission feedback module determines that the encryption transmission efficiency is abnormal.

[0097] Specifically, the transmission feedback module determines the encryption transmission efficiency according to the comparison result of the transmission failure data volume and the preset transmission failure data volume, so as to make optimization adjustments, thereby continuously improving the encryption transmission efficiency.

[0098] Specifically, the feedback cycle refers to the preset duration for feedback on the encryption transmission efficiency. For example, the feedback cycle can be set to 7 days. The amount of transmission failure data refers to the sum of the amount of invalid data, missing data and lost data within the feedback cycle. The invalid data refers to the A-label data that is not the encrypted data content after decryption. The missing data refers to the data that is incomplete after encryption as determined by the data decryption module. The lost data refers to the data that cannot be decrypted according to the A-label data transmission key.

[0099] Specifically, when the encryption transmission efficiency is abnormal, the transmission feedback module adjusts the preset number of manual marking times u0 in the large language model optimization process, sets the adjusted preset number of manual marking times to un0, sets un0=ni×u0, 0.85<np<1, and in this embodiment, the preferred value of ni is ni=0.93.

[0100] Specifically, the transmission feedback module adjusts the preset number of manual marking times u0 in the large language model optimization process to improve the optimization strength of the large model, thereby improving the recognition accuracy of the A-label data.

[0101] Specifically, when the encryption transmission efficiency is abnormal, the transmission feedback module adjusts the preset sampling data volume p0 in the data verification process, and the adjusted preset sampling data volume is pn0, and pn0=ni×p0 is set.

[0102] Specifically, the transmission feedback module adjusts the preset sampling data volume p0 in the data verification process to increase the sampling intensity and further improve the data verification efficiency.

[0103] See also Figure 2 As shown, it is a schematic diagram of the process of the cloud platform data encryption transmission method of this embodiment, and the method includes:

[0104] Step S1, statically identify the data through the encrypted data identification module to obtain A-label data;

[0105] Step S2, dynamically identifying the data according to the large language model through the encrypted data identification module, and supplementing the A-label data according to the dynamic identification result;

[0106] Step S3, obtaining manually marked A-label data through the encrypted data recognition module, optimizing the dynamic recognition result according to the manually marked A-label data, and generating a model optimization database to optimize the large language model;

[0107] Step S4, generating a data transmission key for label A through a data encryption module;

[0108] Step S5, dividing the A-label data by a data encryption module and generating an A-label data vector;

[0109] Step S6, encrypting the divided A-label data through a data encryption module, verifying the encrypted A-label data, and encapsulating the A-label data that has been successfully verified;

[0110] Step S7, transmitting the A-label data after data encapsulation through the data transmission module;

[0111] Step S8, verifying the data integrity of the encrypted A-label data through a data decryption module, and decrypting the encrypted A-label data when it is complete;

[0112] Step S9, judging the encryption transmission efficiency according to the amount of transmission failure data through the transmission feedback module, and adjusting the large language model optimization process and the data verification process according to the judgment result.

[0113] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A cloud platform data encryption transmission system, characterized in that: include: The encrypted data recognition module is used to statically recognize the data and obtain the A-label data, and is also used to dynamically recognize the data according to the large language model and supplement the A-label data according to the dynamic recognition results, and is also used to obtain manually marked A-label data, and optimize the dynamic recognition results according to the manually marked A-label data, and generate a model optimization database to optimize the large language model; A data encryption module is used to generate an A-label data transmission key, to divide the A-label data and generate an A-label data vector, to encrypt the divided A-label data, to verify the encrypted A-label data, and to encapsulate the A-label data that has been successfully verified; A data transmission module, used for transmitting the A-label data after data encapsulation; A data decryption module is used to verify the data integrity of the encrypted A-label data and decrypt it when the encrypted A-label data is complete; The transmission feedback module is used to judge the encryption transmission efficiency according to the amount of data that failed to be transmitted, and adjust the large language model optimization process and data verification process according to the judgment result; The encrypted data recognition module obtains the character data at the preset position, collects the residence time T of the character data at the preset position, compares the residence time T with the preset residence time T0, and performs static recognition on the data according to the comparison result, wherein: When T>T0, the encrypted data recognition module determines that the static recognition of the character data is successful, and marks it as A-labeled data; When T≤T0, the encrypted data recognition module determines that the static recognition of the character data fails; The encrypted data recognition module establishes a large language model according to the encrypted data vocabulary, preprocesses the data, and inputs the preprocessed data into the large language model to obtain a text segment containing the encrypted data, and supplements it as A-label data; The encrypted data identification module obtains manually marked A-marked data, sets the newly added A-marked data as a text segment containing encrypted data, and deletes the deleted A-marked data from the text segment containing encrypted data.

2. The cloud platform data encryption transmission system according to claim 1 is characterized in that: The encrypted data recognition module obtains the number of manual marking times u, compares the number of manual marking times u with the preset number of manual marking times u0, and optimizes the large language model according to the comparison result, wherein: When u≤u0, the encrypted data recognition module does not optimize the large language model; When u>u0, the encrypted data recognition module optimizes the large language model, and uses the manually marked A-labeled data as a model optimization database, and adds it to the encrypted data vocabulary to optimize the large language model.

3. The cloud platform data encryption transmission system according to claim 2 is characterized in that: The data encryption module uses the gpg command to generate the A-label data transmission key and stores it in the HSM hardware security module; The data encryption module uses a preset padding character to fill the A-label data to an integer multiple of 128 bytes, divides the filled A-label data, uses Poly1305 to generate a MAC for each data block obtained after the data division, and uses a strong random number generator to generate an A-label data vector; The data encryption module uses the A-label data transmission key and the A-label data vector to encrypt the divided A-label data through the ChaCha20 algorithm, and appends the MAC generated by each data block to the encrypted data block to obtain the MAC appended to the data block.

4. The cloud platform data encryption transmission system according to claim 3 is characterized in that: The data encryption module presets the sampled data volume p0 at each interval, uses the A-label data transmission key, calculates the MAC of the encrypted A-label data through the Poly1305 algorithm, and compares the calculated MAC of the encrypted A-label data with the MAC attached to the data block, and performs data verification on the encrypted A-label data according to the comparison result, wherein: When the calculated MAC of the encrypted A-label data is consistent with the MAC attached to the data block, the data encryption module determines that the encrypted A-label data verification is successful; When the calculated MAC of the encrypted A-label data is inconsistent with the MAC attached to the data block, the data encryption module determines that the encrypted A-label data verification fails; The data encryption module performs data encapsulation on the encrypted A-label data of the successfully verified A-label data, the MAC attached to the data block, and the A-label data vector to obtain encapsulated A-label data; The data transmission module uses the transport layer security protocol TLS to transmit the A-label data after data encapsulation.

5. The cloud platform data encryption transmission system according to claim 4 is characterized in that: The data decryption module uses the A-label data transmission key to calculate the MAC of the encrypted A-label data through the Poly1305 algorithm, and compares the calculated MAC of the encrypted A-label data with the MAC attached to the data block, and performs data integrity verification on the encrypted A-label data according to the comparison result, where: When the calculated MAC of the encrypted A-label data is consistent with the MAC attached to the data block, the data decryption module determines that the encrypted A-label data is complete, and uses the A-label data transmission key and the A-label data vector to decrypt the encrypted A-label data through the ChaCha20 algorithm; When the calculated MAC of the encrypted A-label data is inconsistent with the MAC attached to the data block, the data decryption module determines that the encrypted A-label data is incomplete.

6. The cloud platform data encryption transmission system according to claim 1 is characterized in that: The transmission feedback module compares the amount of transmission failure data n within the feedback cycle with the preset amount of transmission failure data n0, and determines the encryption transmission efficiency according to the comparison result, wherein: When n≤n0, the transmission feedback module determines that the encryption transmission efficiency is normal; When n>n0, the transmission feedback module determines that the encryption transmission efficiency is abnormal; When the encryption transmission efficiency is abnormal, the transmission feedback module adjusts the preset number of manual marking times u0 in the large language model optimization process, sets the adjusted preset number of manual marking times as un0, and sets un0=ni×u0; When the encryption transmission efficiency is abnormal, the transmission feedback module adjusts the preset sampling data volume p0 in the data verification process, and the adjusted preset sampling data volume is pn0, and pn0=ni×p0 is set.

7. A method applied to the cloud platform data encryption transmission system according to any one of claims 1 to 6, characterized in that: include: Step S1, statically identify the data through the encrypted data identification module to obtain A-label data; Step S2, dynamically identifying the data according to the large language model through the encrypted data identification module, and supplementing the A-label data according to the dynamic identification result; Step S3, obtaining manually marked A-label data through the encrypted data recognition module, optimizing the dynamic recognition result according to the manually marked A-label data, and generating a model optimization database to optimize the large language model; Step S4, generating a data transmission key for label A through a data encryption module; Step S5, dividing the A-label data by a data encryption module and generating an A-label data vector; Step S6, encrypting the divided A-label data through a data encryption module, verifying the encrypted A-label data, and encapsulating the A-label data that has been successfully verified; Step S7, transmitting the A-label data after data encapsulation through the data transmission module; Step S8, verifying the data integrity of the encrypted A-label data through a data decryption module, and decrypting the encrypted A-label data when it is complete; Step S9, judging the encryption transmission efficiency according to the amount of transmission failure data through the transmission feedback module, and adjusting the large language model optimization process and the data verification process according to the judgment result.

Citation Information

Patent Citations

  • Cloud-based network information encryption transmission method

    CN116471007B

  • Financial auditing method and system based on homomorphic encryption and classification optimization

    CN116245669A

  • Efficient de-duplication using block-based convergent encryption

    CN116547642A