Data multi-dimensional linkage security encryption and decryption method based on privacy calculation

Abstract

The invention discloses a data multi-dimensional linkage security encryption and decryption method based on privacy calculation, and relates to the technical field of data security, and the method comprises the steps: 1, obtaining data, and extracting the features and dimensions of the data; 2, segmenting the data into a plurality of sub-data blocks according to the characteristics and dimensions of the data; 3, encrypting each sub-data block through an encryption algorithm to generate a plurality of encrypted data blocks; 4, carrying out relevance logic verification on the encrypted data of the encrypted data block, and judging whether the encryption process of the encryption algorithm is correct or not; 5, performing multi-dimensional linkage analysis on the encrypted data of the plurality of encrypted data blocks through privacy calculation to obtain multi-dimensional linkage analysis data; and step 6, decrypting the multi-dimensional linkage analysis data to obtain final analysis result data. The privacy of the data is protected, the accuracy of the data can be effectively judged, and the user experience of a user is improved while the safety is improved.

Description

Technical Field

[0001] The present invention relates to the field of data security technology, and in particular to a data multi-dimensional linkage security encryption and decryption method based on privacy computing. Background Art

[0002] In the modern information society, data has become an important asset. With the rapid development of technologies such as big data, cloud computing, and artificial intelligence, multi-dimensional linkage analysis of data has become an important means of releasing data value. However, multi-dimensional linkage analysis of data often requires in-depth processing and analysis of data, which may involve data privacy and security issues.

[0003] Existing data security technologies usually use data desensitization and data encryption to protect data privacy and security. Data desensitization technology protects data privacy by blurring the data, making the data lose its original identity to a certain extent. Data encryption technology encrypts data through encryption algorithms, so that only authorized parties holding the correct key can decrypt and access the data. However, existing data desensitization technologies often affect the value of data, limiting the analysis and use of data. Secondly, although existing data encryption technologies can protect data privacy, they will also increase the performance overhead of data processing and affect the efficiency of data processing. Moreover, due to the influence of encryption algorithms and processing performance, existing data encryption technologies may have encryption errors during the encryption process, resulting in errors in the results of multi-dimensional linkage analysis of data, and due to the influence of data privacy protection, it is impossible to know whether the encryption process is correct in time, resulting in the output of incorrect decrypted data affecting the user experience. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the data security technology in the prior art that the data needs to be desensitized, which limits the analysis and utilization of the data. At the same time, the data encryption technology increases the performance overhead of data processing, affects the efficiency of data processing, and encryption errors occur during the encryption process, resulting in the inability to correctly decrypt the data, affecting the user experience. A multi-dimensional linkage security encryption and decryption method for data based on privacy computing is provided. Through this method, the data does not need to be desensitized and can have high data security. At the same time, the data encryption process is verified to ensure the output of correct decrypted data.

[0005] The purpose of the present invention is achieved through the following technical solutions: A data multi-dimensional linkage security encryption and decryption method based on privacy computing, comprising the following steps: Step 1: Obtain and preprocess the data, and then extract the features and dimensions of the data; Step 2: divide the data into several sub-data blocks according to the characteristics and dimensions of the data, each sub-data block represents a feature or a dimension; Step 3, encrypting each sub-data block by an encryption algorithm to generate a number of encrypted data blocks; Step 4, perform correlation logic verification on the encrypted data of the encrypted data block to determine whether the encryption process of the encryption algorithm is correct. If it is correct, jump to step 5. If it is not correct, adjust the parameters of the encryption algorithm and re-execute step 3 until the encryption process of the encryption algorithm is correct; Step 5: Perform multi-dimensional linkage analysis on the encrypted data of several encrypted data blocks through privacy computing to obtain multi-dimensional linkage analysis data; Step 6: Decrypt the multi-dimensional linkage analysis data to obtain the final analysis result data.

[0006] In this scheme, the data is divided into several sub-data blocks according to the characteristics and dimensions of the data, which is conducive to easier understanding and interpretation during data analysis, and is also convenient for subsequent multi-dimensional linkage analysis, while also reducing the complexity of the overall calculation. Through privacy computing, the encrypted data of several encrypted data blocks are subjected to multi-dimensional linkage analysis, ensuring that the data always exists in ciphertext during data analysis and processing. Even if the attacker cracks the privacy computing method, the multi-dimensional linkage analysis data is restored to obtain only the encrypted data. This scheme effectively protects the confidentiality and integrity of the data, so there is no need to perform data desensitization to protect data privacy. At the same time, since this scheme first encrypts each sub-data block and then performs multi-dimensional linkage analysis on the encrypted data, it is also necessary to determine whether the encryption process of the encryption algorithm is correct to ensure that the data decrypted by the multi-dimensional linkage analysis is true and valid. Although the data after the encryption algorithm is encrypted, it is still obtained based on the original data conversion. Therefore, there is still a logical correlation between the encrypted data. Therefore, this scheme verifies the correlation logic of the encrypted data to determine whether the encryption process of the encryption algorithm is correct, which can avoid the situation where the encryption algorithm is wrong and the result of decrypting the multi-dimensional linkage analysis data is also wrong. Since the data is still encrypted during the correlation logic verification, the verification process will not leak the real data, thus protecting the privacy of the data while judging the accuracy of the data.

[0007] Preferably, in step 4, the encrypted data of the encrypted data block is subjected to correlation logic verification to determine whether the encryption process of the encryption algorithm is correct, specifically: Obtaining correlation characteristics of encrypted data of at least two encrypted data blocks, the correlation characteristics including positive correlation, negative correlation and mutual exclusivity; For positively correlated encrypted data blocks, determine whether their correlation coefficients are greater than or equal to a set positive correlation threshold. If they are greater than or equal to the set positive correlation threshold, then it is determined that these encrypted data blocks are positively correlated and the encryption process of the encryption algorithm is correct. If they are less than the set positive correlation threshold, then it is determined that these encrypted data blocks are not correlated and the encryption process of the encryption algorithm is wrong. For the negatively correlated encrypted data blocks, determine whether their correlation coefficient is less than or equal to the set negative correlation threshold. If it is less than or equal to the set negative correlation threshold, then it is determined that these encrypted data blocks are negatively correlated and the encryption process of the encryption algorithm is correct. If it is greater than the set positive correlation threshold, then it is determined that these encrypted data blocks do not have correlation and the encryption process of the encryption algorithm is wrong. For mutually exclusive encrypted data blocks, determine whether the values ​​of the encrypted data of the mutually exclusive encrypted data blocks are within the set mutual exclusion threshold range after cumulative calculation. If they are within the mutual exclusion threshold range, it is determined that these encrypted data are mutually exclusive and the encryption process of the encryption algorithm is correct. Otherwise, it is determined that there is no correlation between these encrypted data and the encryption process of the encryption algorithm is wrong.

[0008] There must be correlations between data of different dimensions. The design of this solution takes into account positive correlation, negative correlation and mutual exclusivity. Although the data is encrypted, there is still correlation between data of different dimensions. Therefore, this solution uses positive correlation, negative correlation and mutual exclusivity to determine whether the encryption process of the encryption algorithm is correct. If there are correlations between multiple encrypted data blocks, the number of encrypted data blocks for correlation logic verification will be flexibly selected according to actual needs. If the number of encrypted data blocks is small, the accuracy of the logic verification may also decrease. If the number of encrypted data blocks is large, the amount of calculation will also increase.

[0009] Preferably, in step 1, the data is preprocessed, specifically: Acquire sensitive items of data and anonymize and de-identify sensitive data. Anonymization and de-identification include data perturbation, pseudonymization, generalization, and data suppression. This solution is designed to prevent the identity of individuals or entities from being identified, further protecting the privacy and security of data.

[0010] Preferably, in step 2, the sub-data blocks are also subjected to random processing, and the random processing includes randomly rearranging, replacing or adding noise to the data, and randomly perturbing part or all of the data in the sub-data blocks.

[0011] Preferably, the step 3 is specifically as follows: A lightweight symmetric encryption algorithm is used to encrypt the data of the sub-data block to generate encrypted data; The encrypted data is stored in a local storage device, cloud storage, or distributed storage system, and the stored encrypted data constitutes an encrypted data block.

[0012] Preferably, in step 5, a multi-dimensional linkage analysis is performed on the encrypted data of several encrypted data blocks through privacy computing, including: Use homomorphic encryption technology to perform arithmetic operations on encrypted data, including addition and multiplication; Multi-party secure computing is used to perform homomorphic encryption on multiple encrypted data blocks separately, and then the data homomorphically encrypted by all parties are merged to obtain multi-dimensional linkage analysis data.

[0013] As a preference, after obtaining the final analysis result data, the analysis result data is also shared and exchanged, specifically as follows: The owner of the analysis result data encrypts the analysis result data and publishes it to the data sharing platform, sets the access control policy, and defines the user attribute set that is allowed to access the data; The data requester sends an access request to the data sharing platform and provides its own attribute proof; The data sharing platform verifies whether the user's attributes meet the access control policy set by the data owner based on the attribute-based encryption method. If the user meets the access control policy, he / she is allowed to access the data. The data sharing platform provides the key required for decryption or directly decrypts the analysis result data, and the data requester obtains the decrypted analysis result data.

[0014] Preferably, the attribute-based encryption method specifically adopts a ciphertext policy attribute-based encryption method. The owner of the analysis result data embeds an access control policy when encrypting the data. Only users whose attribute sets meet the access policy are allowed to access the data. The design of this solution enhances the flexibility and security of data access control and supports complex permission management and fine-grained access control.

[0015] As a preferred method, the data multi-dimensional linkage security encryption and decryption method based on privacy computing also performs fine-grained control and fuzzification processing on the analysis result data. The design of this solution ensures that users can only obtain data information within their authority to prevent excessive exposure of data.

[0016] Preferably, in step 6, after decrypting the multi-dimensional linkage analysis data, noise or disturbance is added to the encryption result according to the differential privacy algorithm. The design of this solution can prevent attackers from reversely inferring the original data through the analysis results, improve the security and privacy protection level of data analysis, and also ensure the validity and accuracy of the analysis results.

[0017] The beneficial effect of the present invention is that the present invention can avoid the situation that the error in the multi-dimensional linkage analysis of the encrypted data causes the decryption result of the multi-dimensional linkage analysis data to be wrong by performing correlation logic verification on the encrypted data to determine whether the confidential process data of the confidential algorithm is correct, thereby ensuring the accuracy of the multi-dimensional linkage analysis result. Since the data is still encrypted when the correlation logic verification is performed, the verification process will not leak the real data, thereby protecting the privacy of the data and determining the accuracy of the data.

[0018] The present invention effectively protects the confidentiality and integrity of data, so there is no need to perform data desensitization to protect data privacy. The lightweight symmetric encryption algorithm used in the present invention can reduce the performance overhead in the encryption process and improve the efficiency of data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0020] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be more comprehensive and complete and fully convey the concept of the example embodiments to those skilled in the art.

[0021] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0022] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0023] Embodiment 1: A data multi-dimensional linkage security encryption and decryption method based on privacy computing, such as Figure 1 As shown, the following steps are included: Step 1: Obtain and preprocess the data, and then extract the features and dimensions of the data; Step 2: divide the data into several sub-data blocks according to the characteristics and dimensions of the data, each sub-data block represents a feature or a dimension; Step 3, encrypting each sub-data block by an encryption algorithm to generate a number of encrypted data blocks; Step 4: Perform multi-dimensional linkage analysis on the encrypted data of several encrypted data blocks through privacy computing to obtain multi-dimensional linkage analysis data; Step 5: perform correlation logic verification on the multi-dimensional linkage analysis data to determine whether the multi-dimensional linkage analysis data is correct. If correct, jump to step 6. If incorrect, adjust the parameters of the multi-dimensional linkage analysis and re-execute step 4 until the multi-dimensional linkage analysis data is correct. Step 6: Decrypt the multi-dimensional linkage analysis data to obtain the final analysis result data.

[0024] In this scheme, the data is divided into several sub-data blocks according to the characteristics and dimensions of the data, which is conducive to easier understanding and interpretation during data analysis, and is also convenient for subsequent multi-dimensional linkage analysis, while also reducing the complexity of the overall calculation. Through privacy computing, the encrypted data of several encrypted data blocks are subjected to multi-dimensional linkage analysis, ensuring that the data always exists in ciphertext during data analysis and processing. Even if the attacker cracks the privacy computing method, the multi-dimensional linkage analysis data is restored to obtain only the encrypted data. This scheme effectively protects the confidentiality and integrity of the data, so there is no need to perform data desensitization to protect data privacy. At the same time, since this scheme first encrypts each sub-data block and then performs multi-dimensional linkage analysis on the encrypted data, it is also necessary to determine whether the encryption process of the encryption algorithm is correct to ensure that the data decrypted by the multi-dimensional linkage analysis is true and valid. Although the data after the encryption algorithm is encrypted, it is still obtained based on the original data conversion. Therefore, there is still a logical correlation between the encrypted data. Therefore, this scheme determines whether the encryption process of the encryption algorithm is correct by performing correlation logic verification on the encrypted data. It can be used to determine whether the encryption process of the encryption algorithm is correct by performing an error in the encryption algorithm, which may cause the decryption result of the multi-dimensional linkage analysis data to be wrong. Since the data is still encrypted during the correlation logic verification, the verification process will not leak the real data, thus protecting the privacy of the data while judging the accuracy of the data.

[0025] In step 4, the encrypted data of the encrypted data block is subjected to correlation logic verification to determine whether the encryption algorithm encryption process is correct, specifically: Obtaining correlation characteristics of encrypted data of at least two encrypted data blocks, the correlation characteristics including positive correlation, negative correlation and mutual exclusivity; For positively correlated encrypted data blocks, determine whether their correlation coefficients are greater than or equal to a set positive correlation threshold. If they are greater than or equal to the set positive correlation threshold, then it is determined that these encrypted data blocks are positively correlated and the encryption process of the encryption algorithm is correct. If they are less than the set positive correlation threshold, then it is determined that these encrypted data blocks are not correlated and the encryption process of the encryption algorithm is wrong. For the negatively correlated encrypted data blocks, determine whether their correlation coefficient is less than or equal to the set negative correlation threshold. If it is less than or equal to the set negative correlation threshold, then it is determined that these encrypted data blocks are negatively correlated and the encryption process of the encryption algorithm is correct. If it is greater than the set positive correlation threshold, then it is determined that these encrypted data blocks do not have correlation and the encryption process of the encryption algorithm is wrong. For mutually exclusive encrypted data blocks, determine whether the values ​​of the encrypted data of the mutually exclusive encrypted data blocks are within the set mutual exclusion threshold range after cumulative calculation. If they are within the mutual exclusion threshold range, it is determined that these encrypted data are mutually exclusive and the encryption process of the encryption algorithm is correct. Otherwise, it is determined that there is no correlation between these encrypted data and the encryption process of the encryption algorithm is wrong.

[0026] There must be correlations between data of different dimensions. The design of this solution takes into account positive correlation, negative correlation and mutual exclusivity. Although the multi-dimensional linkage data is encrypted, there is still correlation between data of different dimensions. Therefore, this solution uses positive correlation, negative correlation and mutual exclusivity to determine whether the multi-dimensional linkage analysis data is correct. If there is a correlation between multiple encrypted data blocks, the number of encrypted data blocks for correlation logic verification will be flexibly selected according to actual needs. If the number of encrypted data blocks is small, the accuracy of the logic verification may also decrease. If the number of encrypted data blocks is large, the amount of calculation will also increase.

[0027] In this embodiment, for positively correlated encrypted data blocks, the Pearson Correlation Coefficient is used to measure the correlation between them. Assume that there are two encrypted data blocks A and B, each of which contains n encrypted data points, represented as A1, A2, ..., An and B1, B2, ..., Bn respectively. The calculation formula of the Pearson Correlation Coefficient is as follows: ; in, represents the correlation coefficient, and They represent the mean of encrypted data block A and encrypted data block B respectively, and They represent the value of an item in encrypted data block A and encrypted data block B respectively.

[0028] For the negatively correlated encrypted data blocks, the Pearson correlation coefficient is also used for judgment, but the threshold range of the correlation coefficient is set differently.

[0029] Assume there are three encrypted data blocks A, B and C, verify the correlation between them: The correlation coefficient between A and B is calculated to be r=0.8, and the positive correlation threshold is set to 0.7. Therefore, A and B are positively correlated, and the encryption process of the encryption algorithm is correct; the correlation coefficient between A and C is calculated to be r=-0.9, and the negative correlation threshold is set to -0.7. Therefore, A and C are negatively correlated, and the encryption process of the encryption algorithm is correct. The data accumulation value of B and C is calculated to be S=0.5, and the mutually exclusive threshold range is set to [0, 1]. Therefore, B and C are mutually exclusive, and the encryption process of the encryption algorithm is correct.

[0030] In step 1, the data is preprocessed, specifically: Clean, normalize, reduce dimension and convert format of raw data to ensure data quality, consistency and availability, eliminate noise, missing values ​​and outliers, and lay the foundation for subsequent data analysis; obtain sensitive items of data, anonymize and de-identify sensitive item data, and anonymize and de-identify data, including data perturbation, pseudonymization, generalization and data suppression. The design of this solution prevents the identity of individuals or entities from being identified, further protecting the privacy and security of data.

[0031] In the step 2, the sub-data blocks are also subjected to random processing, and the random processing includes randomly rearranging, replacing or adding noise to the data, and randomly perturbing part or all of the data in the sub-data blocks.

[0032] The step 3 is specifically as follows: A lightweight symmetric encryption algorithm is used to encrypt the data of the sub-data block to generate encrypted data; The encrypted data is stored in a local storage device, cloud storage, or distributed storage system, and the stored encrypted data constitutes an encrypted data block.

[0033] In this embodiment, the lightweight symmetric encryption algorithm may be an improved AES algorithm or a lightweight block cipher algorithm, which can reduce the performance overhead in the encryption process and improve the efficiency of data processing.

[0034] In step 4, a multi-dimensional linkage analysis is performed on the encrypted data of several encrypted data blocks through privacy computing, including: Use homomorphic encryption technology to perform arithmetic operations on encrypted data, including addition and multiplication; Multi-party secure computing is used to perform homomorphic encryption on multiple encrypted data blocks separately, and then the data homomorphically encrypted by all parties are merged to obtain multi-dimensional linkage analysis data.

[0035] After obtaining the final analysis result data, the analysis result data is also shared and exchanged, specifically: The owner of the analysis result data encrypts the analysis result data and publishes it to the data sharing platform, sets the access control policy, and defines the user attribute set that is allowed to access the data; The data requester sends an access request to the data sharing platform and provides its own attribute proof; The data sharing platform verifies whether the user's attributes meet the access control policy set by the data owner based on the attribute-based encryption method. If the user meets the access control policy, he / she is allowed to access the data. The data sharing platform provides the key required for decryption or directly decrypts the analysis result data, and the data requester obtains the decrypted analysis result data.

[0036] The attribute-based encryption method specifically adopts a ciphertext policy attribute-based encryption method. The owner of the analysis result data embeds the access control policy when encrypting the data. Only users whose attribute set meets the access policy are allowed to access the data. The design of this solution enhances the flexibility and security of data access control and supports complex permission management and fine-grained access control.

[0037] The data multi-dimensional linkage security encryption and decryption method based on privacy computing also performs fine-grained control and fuzzy processing on the analysis result data. The design of this solution ensures that users can only obtain data information within their authority to prevent excessive exposure of data.

[0038] In step 6, after decrypting the multi-dimensional linkage analysis data, noise or disturbance is added to the encryption result according to the differential privacy algorithm. The design of this solution can prevent attackers from inferring the original data through the analysis results, improve the security and privacy protection level of data analysis, and also ensure the effectiveness and accuracy of the analysis results.

[0039] During the data encryption process, parallel encryption technology and hardware acceleration methods, such as GPU acceleration, FPGA acceleration or dedicated encryption chips, are used to perform parallel encryption processing on multiple sub-data blocks, which significantly improves the efficiency of data encryption, reduces the performance overhead of data processing, and meets real-time requirements.

[0040] The following is a specific application of obtaining the carbon emission results of an enterprise based on the historical energy consumption data of the enterprise through the method of this embodiment: Obtain the historical energy consumption data of enterprises. This data should include energy consumption records of different enterprises, such as the consumption of coal, oil, gas and other energy sources, as well as the timestamp of consumption; According to the characteristics and dimensions of the data, the data is divided into several sub-data blocks. For example, the data is divided according to the dimensions of energy type (coal, oil, gas, clean energy, etc.), time period (year, month, day), etc., and each sub-data block is encrypted. At the same time, the encrypted data is logically verified for correlation, such as the negative correlation between clean energy and fossil energy, and the positive correlation between energy consumption and carbon emission factors.

[0041] The encrypted data blocks are analyzed in a multi-dimensional linkage manner through privacy computing technology. The encrypted data blocks are analyzed in a multi-dimensional linkage manner through privacy computing technology, and the encrypted results calculated by all parties are merged to obtain the final carbon emission result data.

[0042] Decrypt the encrypted carbon emission results obtained by multi-dimensional linkage analysis. Use the corresponding decryption key and lightweight decryption algorithm to restore the original carbon emission result data.

[0043] Enterprises can use the decrypted carbon emission result data for decision support, business optimization, etc. For example, they can formulate emission reduction plans and optimize energy use structure based on carbon emissions.

[0044] Example 2: A method for secure encryption and decryption of data in multi-dimensional linkage based on privacy computing. Its principle and implementation method are basically the same as those of Example 1. The difference is that in the specific implementation method of performing correlation logic verification on the encrypted data of the encrypted data block to determine whether the encryption process of the encryption algorithm is correct, the minimum value, maximum value, quartile and standard deviation of the encrypted data block are also obtained and calculated: For positively correlated encrypted data blocks and negatively correlated encrypted data blocks, determine whether the minimum value, maximum value, quartile and standard deviation of the encrypted data blocks are within a reasonable range. For mutually exclusive encrypted data blocks, check whether the minimum value and maximum value are complementary, and use the quartiles and standard deviation to determine whether there is an overlapping area between the two mutually exclusive encrypted data blocks.

[0045] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the embodiments disclosed herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary technical means in the art that are not disclosed in the present application.

[0046] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A data multi-dimensional linkage security encryption and decryption method based on privacy computing, characterized in that: The following steps are involved: Step 1: Obtain and preprocess the data, and then extract the features and dimensions of the data; Step 2: divide the data into several sub-data blocks according to the characteristics and dimensions of the data, each sub-data block represents a feature or a dimension; Step 3, encrypting each sub-data block by an encryption algorithm to generate a number of encrypted data blocks; Step 4, perform correlation logic verification on the encrypted data of the encrypted data block to determine whether the encryption process of the encryption algorithm is correct. If it is correct, jump to step 5. If it is not correct, adjust the parameters of the encryption algorithm and re-execute step 3 until the encryption process of the encryption algorithm is correct; Step 5: Perform multi-dimensional linkage analysis on the encrypted data of several encrypted data blocks through privacy computing to obtain multi-dimensional linkage analysis data; Step 6: Decrypt the multi-dimensional linkage analysis data to obtain the final analysis result data.

2. According to the method of claim 1, the method is characterized in that: In step 4, the encrypted data of the encrypted data block is subjected to correlation logic verification to determine whether the encryption algorithm encryption process is correct, specifically: Obtaining correlation characteristics of encrypted data of at least two encrypted data blocks, the correlation characteristics including positive correlation, negative correlation and mutual exclusivity; For positively correlated encrypted data blocks, determine whether their correlation coefficients are greater than or equal to a set positive correlation threshold. If they are greater than or equal to the set positive correlation threshold, then it is determined that these encrypted data blocks are positively correlated and the encryption process of the encryption algorithm is correct. If they are less than the set positive correlation threshold, then it is determined that these encrypted data blocks are not correlated and the encryption process of the encryption algorithm is wrong. For the negatively correlated encrypted data blocks, determine whether their correlation coefficient is less than or equal to the set negative correlation threshold. If it is less than or equal to the set negative correlation threshold, then it is determined that these encrypted data blocks are negatively correlated and the encryption process of the encryption algorithm is correct. If it is greater than the set positive correlation threshold, then it is determined that these encrypted data blocks do not have correlation and the encryption process of the encryption algorithm is wrong. For mutually exclusive encrypted data blocks, determine whether the values ​​of the encrypted data of the mutually exclusive encrypted data blocks are within the set mutual exclusion threshold range after cumulative calculation. If they are within the mutual exclusion threshold range, it is determined that these encrypted data are mutually exclusive and the encryption process of the encryption algorithm is correct. Otherwise, it is determined that there is no correlation between these encrypted data and the encryption process of the encryption algorithm is wrong.

3. According to the method of claim 1, the method is characterized in that: In step 1, the data is preprocessed, specifically: Acquire sensitive items of the data and perform anonymization and de-identification processing on the sensitive item data. The anonymization and de-identification processing includes data perturbation, pseudonymization, generalization and data suppression.

4. According to the method of claim 1, the method is characterized in that: In the step 2, the sub-data blocks are also subjected to random processing, and the random processing includes randomly rearranging, replacing or adding noise to the data, and randomly perturbing part or all of the data in the sub-data blocks.

5. According to the method of claim 1, the method is characterized in that: The step 3 is specifically as follows: A lightweight symmetric encryption algorithm is used to encrypt the data of the sub-data block to generate encrypted data; The encrypted data is stored in a local storage device, cloud storage, or distributed storage system, and the stored encrypted data constitutes an encrypted data block.

6. According to the method of claim 1, the method is characterized in that: In step 5, a multi-dimensional linkage analysis is performed on the encrypted data of several encrypted data blocks through privacy computing, including: Use homomorphic encryption technology to perform arithmetic operations on encrypted data, including addition and multiplication; Multi-party secure computing is used to perform homomorphic encryption on multiple encrypted data blocks separately, and then the data homomorphically encrypted by all parties are merged to obtain multi-dimensional linkage analysis data.

7. According to the method of claim 1, the method is characterized in that: After obtaining the final analysis result data, the analysis result data is also shared and exchanged, specifically: The owner of the analysis result data encrypts the analysis result data and publishes it to the data sharing platform, sets the access control policy, and defines the user attribute set that is allowed to access the data; The data requester sends an access request to the data sharing platform and provides its own attribute proof; The data sharing platform verifies whether the user's attributes meet the access control policy set by the data owner based on the attribute-based encryption method. If the user meets the access control policy, he / she is allowed to access the data. The data sharing platform provides the key required for decryption or directly decrypts the analysis result data, and the data requester obtains the decrypted analysis result data.

8. According to the method of claim 7, the multi-dimensional linkage data security encryption and decryption method based on privacy computing is characterized in that: The attribute-based encryption method specifically adopts a ciphertext strategy attribute-based encryption method. The owner of the analysis result data embeds an access control policy when encrypting the data. Only users whose attribute sets meet the access policy are allowed to access the data.

9. The method for multi-dimensional linkage security encryption and decryption of data based on privacy computing according to claim 7 or 8 is characterized in that: The analysis result data is also subjected to fine-grained control and fuzzy processing.

10. According to the method of claim 1, the method is characterized in that: In step 6, after the multi-dimensional linkage analysis data is decrypted, noise or disturbance is added to the encryption result according to the differential privacy algorithm.

Images