A multi-user ciphertext retrieval method supporting mode leakage regulation and dynamic permission management

By segmenting file content using erasure coding and differential privacy technology and generating a noisy index, combined with trusted authorization authority management permissions, the problem of pattern leakage and dynamic permission management in multi-user scenarios is solved, achieving efficient and secure data retrieval.

CN120811599BActive Publication Date: 2026-11-10NANJING UNIV OF SCI & TECH
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202511184202.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-08-22
Publication Date
2026-11-10
Estimated Expiration
2045-08-22

AI Technical Summary

Technical Problem

Existing multi-user symmetric searchable encryption technologies suffer from pattern leakage during data retrieval, leading to privacy degradation and complex access management, making it difficult to meet the needs of dynamic user access.

Method used

The system uses erasure coding technology to fragment file content, combines differential privacy technology to generate a noise index, and manages user permissions through a trusted authorization authority to achieve dynamic permission updates and pattern leakage control.

Benefits of technology

It effectively mitigates privacy degradation caused by pattern leakage, supports dynamic permission management, improves system security and flexibility, and ensures data confidentiality and query efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120811599B_ABST
    Figure CN120811599B_ABST
Patent Text Reader

Abstract

The application discloses a multi-user secret state retrieval method supporting mode leakage regulation and dynamic permission management, and belongs to the technical field of computer application. S1: method initialization, a trusted authorization mechanism generates an encryption key, private information and auxiliary information of each user by using a key generation algorithm KeyGen; S2: local determination of confusion strength by the user; S3: local analysis of index information by the user; S4: local confusion of index information by the user; S5: uploading of fragment information, user encryption of fragment content and a security index, and uploading to a cloud server for storage; S6: sending of a query request; S7: user search of data; S8: decryption of a search result; S9: user permission update; The application can efficiently realize dynamic access control of user permissions, effectively alleviate the privacy degradation problem caused by user privacy level preferences in the cross-user data search process, and thus meet higher security requirements in complex application scenarios.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention belongs to the field of computer application technology, specifically relating to a multi-user encrypted retrieval method that supports pattern leakage control and dynamic permission management. Background Technology

[0002] Searchable Symmetric Encryption (SSE) is a typical encrypted retrieval technology. Its core idea is to encrypt data using symmetric encryption while simultaneously generating a secure index, allowing users to search encrypted data without decryption, achieving "usable but not visible" data. Currently, SSE technology is widely used in cloud storage, data sharing, healthcare, and finance.

[0003] However, while SSE solutions effectively prevent servers from directly accessing data content and query keywords, ensuring data and query privacy, most solutions inevitably leak some "seemingly acceptable information" in order to improve query performance. Typical forms of leakage include: (a) search patterns, identifying identical queries in a sequence of search requests; (b) access patterns, revealing the set of document identifiers matching the query; and (c) size patterns, reflecting the number of documents contained in the search results. This seemingly harmless leaked information can be maliciously exploited to launch leak abuse attacks, recover query keywords, or even reconstruct database records.

[0004] Currently, there are various protection mechanisms to mitigate the aforementioned pattern leakage. Mainstream methods include: (a) cryptographic-based protection mechanisms, such as ORAM, Fully Homomorphic Encryption (FHE), and Function Secret Sharing (FSS) to achieve pattern protection. However, these mechanisms typically involve high computational and communication overhead, making them difficult to apply directly to large-scale data processing. (b) Differential Privacy (DP)-based protection schemes, which introduce false positive and false negative responses by adding noise to obfuscate query results and efficiently hide pattern leakage. Strong noise injection can significantly affect query accuracy, but its efficiency and flexibility have led to extensive research.

[0005] Existing privacy protection solutions using differential privacy technology primarily focus on pattern leakage issues in single-user scenarios. In real-world applications, it's more common for multiple users to upload locally encrypted data to cloud servers and retrieve data from each other. While these solutions can theoretically be scaled up to multi-user environments through replication, directly migrating single-user solutions in practice faces numerous challenges, such as difficulties in key isolation and decryption, privacy degradation, decreased retrieval accuracy, and complex access control. Existing multi-user solutions mainly assume that users can securely share encryption keys, meaning each user uses the same key to encrypt local data. However, this places high demands on system security and flexibility. Once the key is leaked, it poses a significant threat to the security of all users' data.

[0006] Because users have varying privacy requirements for local data, each user may independently add noise during local data maintenance. While this independent noise addition protects individual privacy, it increases the overall noise level, further reducing query efficiency. Simultaneously, the data correlation between multiple users is not fully considered, potentially leading to uneven privacy protection and even new privacy leaks. Attackers may leak sensitive information such as query keywords based on leakage function information, such as allowed access patterns in symmetric searchable encryption. This poses challenges to data confidentiality and query privacy, thus placing higher demands on encrypted search methods in multi-user scenarios using different keys. For example, among three users, due to varying confidentiality requirements for local data, the degree of data distortion observable by the attacker for the same data will also differ, allowing for different levels of privacy protection based on user preferences. However, the difficulty for attackers to infer the underlying keywords for each user may be reduced to the lowest level. Attackers may use the least private user's leaked pattern information to perform word frequency statistics, thereby inferring the underlying keywords corresponding to that user's search token, and then linking those keywords to search tokens from other users uploaded at the same time.

[0007] Meanwhile, multi-user scenarios also present challenges in managing permissions, such as dynamic user logout and logon. For example, a company's file access permissions typically include the following types: the general manager can access files from all departments, department managers can only access files within their own departments, and regular employees can only view files related to their own departments. Furthermore, employees may be promoted or demoted. When an employee is promoted, their access permissions need to be expanded, increasing the need for dynamic permission management.

[0008] Therefore, it is necessary to design a multi-user encrypted retrieval method that supports pattern leakage control and dynamic permission management to meet the above requirements. Summary of the Invention

[0009] In response to the problems mentioned in the background art, this invention proposes a multi-user encrypted retrieval method that supports pattern leakage control and dynamic permission management. It can determine the privacy protection level of each user from the perspective of the maximum leakage risk during the data retrieval process, alleviate the privacy degradation problem caused by the leakage of allowed pattern information in searchable encryption, improve the reliability and security of the system, and support the dynamic updating of user permissions.

[0010] Technical Solution: To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0011] A multi-user encrypted retrieval method supporting pattern leakage control and dynamic access control, involving trusted authorization authorities, users, and cloud servers, specifically includes the following steps:

[0012] S1: Method initialization, the trusted authorization authority uses the KeyGen key generation algorithm to generate each user's encryption key, private information and auxiliary information;

[0013] S2: The user determines the obfuscation strength locally. The user determines the privacy protection level of the data by statistically analyzing the distribution characteristics of local data and using a Bayesian-based budget allocation mechanism.

[0014] S3: The user parses the index information locally. The user uses erasure coding technology to divide the file content into fragments for storage, and then parses the fragment content into fragment identifiers and keyword matching pairs.

[0015] S4: User-locally obfuscates index information. Based on the obfuscation strength of S2, the user adds noise to the matching pairs of processed fragment identifiers and keywords, and generates a noisy secure index through the indexing algorithm.

[0016] S5: Upload fragment information, the user encrypts the fragment content and security index, and uploads it to the cloud server for storage;

[0017] S6: Send a query request. The user uses the key management mechanism to recover the authorized user key set based on private and auxiliary information, encrypts the key to generate a search token set and uploads it to the cloud server.

[0018] S7: When a user searches for data, the cloud server retrieves the encrypted database based on a set of search tokens and returns the search results.

[0019] S8: Decrypt search results. Users decrypt fragmented content in search results and reconstruct the original file using erasure coding decoding technology.

[0020] S9: User permission update. The trusted authorization authority responds to the user's request, performs key update, permission addition or deletion operations, and updates auxiliary information and corresponding data in the cloud server.

[0021] As a preferred option, the specific implementation process in S1 is as follows:

[0022] Trusted authorization authorities run the KeyGen key generation algorithm to generate keys for each user and maintain private information about data owners' keys. And auxiliary information for recovering the data owner's key. ;Specifically:

[0023] ,

[0024] in, This indicates auxiliary information used to recover the data owner's key. This represents a symmetric encryption algorithm in cryptography. The encryption key representing the data owner. Includes index encryption key and fragment encryption key .

[0025] As a preferred option, the specific implementation process in S2 is as follows:

[0026] First, users need to parse the file content into file identifier-keyword matching pairs; then determine the user data proportion and keyword coverage; each user determines the proportion of local documents... and keyword coverage The appropriate level of privacy protection is calculated as follows:

[0027] ,

[0028] in, The value indicates the level of privacy protection; Obfuscate() is a function that calculates the level of obfuscation.

[0029] As a preferred option, the specific implementation process in S3 is as follows:

[0030] The user uses erasure coding technology to divide the content of a file into m fragments for storage, specifically:

[0031] ;

[0032] Where S represents the fragment content; D represents the file content that needs to be split; m represents the number of fragments; z represents the minimum number of fragments that can be used to recover the original document content; ErasureCode.Encode() represents the erasure coding algorithm;

[0033] Then the segmented content is parsed into Matching pairs with keyword w .

[0034] Preferably, in S4, the specific implementation process is as follows:

[0035] Each user calculates the privacy protection strength for each keyword w , in the matching pair implement obfuscation;

[0036] First, the user converts the privacy protection strength of a certain keyword into the flipping probabilities p and q for implementing differential privacy obfuscation;

[0037] Then, obfuscate the fragments of this keyword, specifically in two ways: for a fragment containing this keyword, the user will randomly generate a random number r1, if the random number r1>p, the fragment is retained, otherwise, it is discarded; for a fragment not containing this keyword, the user will randomly generate a random number r2, if the random number r2<q, the fragment will be added to the matching pair of this keyword, otherwise, there is no matching relationship;

[0038] Finally, each user can obtain the obfuscated noisy index information ;

[0039] Then, each user processes the noisy secure index based on the index algorithm BldInx, specifically as follows:

[0040] ;

[0041] Wherein, I represents the noisy secure index; k represents the key used to encrypt the noisy plaintext index.

[0042] Preferably, in S5, the specific implementation process is as follows:

[0043] The user processes the local fragment content, encrypts it with the fragment key, and uploads it together with the secure index;

[0044] Based on the data encryption algorithm SymEnc, each user locally calculates the fragment ciphertext, specifically as follows:

[0045] ;

[0046] Wherein, c represents the ciphertext of the encrypted fragment content; represents the symmetric key for encrypting and decrypting the fragment content; S represents the fragment content.

[0047] Preferably, in S6, the specific implementation process is as follows:

[0048] Using the key derivation algorithm KeyDer, recover the authorized user key set based on private information and auxiliary information;

[0049] If user u wants to search for other users' data on the server, the user needs to directly use local private information. and auxiliary information Recover the authorized user's key ,in , Indicates user The index encryption key, This represents the encryption key for the fragmented content of user v;

[0050] The user uses the same key recovery algorithm to recover the keys of each legitimate data owner who has access, thus obtaining a user key set. ;

[0051] Truder trapdoor generation algorithm: Client input key The search term is "w". Finally, the client receives a search token. Using a trapdoor generation algorithm The user uses the recovered user key set Each user's key is used to encrypt the keyword of interest, thus obtaining a search credential for accessing the corresponding user's encrypted database. ,in, This represents the generated search credentials, and k represents the key used to encrypt the noisy plaintext index. Keywords indicating interest;

[0052] Users use the same trapdoor generation algorithm to obtain a set of search tokens that can be used to access all legally accessible encrypted databases and upload them.

[0053] As a preferred option, the specific implementation process in S8 is as follows:

[0054] S81: The user uses the fragment decryption key obtained from the recovered authorized user to decrypt and obtain the original fragment content;

[0055] S82: Users put plaintext fragments belonging to the same document into the same set. If the size of the set Greater than the document recovery threshold Then use erasure coding decoding algorithm. To restore the original file content, specifically:

[0056] ;

[0057] in, Indicates by The original file content was successfully recovered from the fragment; This indicates the total number of fragments into which the document content is divided; This indicates the minimum number of fragments that can be used to recover the original document content; This represents the set of all fragments belonging to the same document.

[0058] As a preferred option, in S81, specifically:

[0059] ;

[0060] in, Indicates the data decryption algorithm; This represents the plaintext fragment obtained after decryption; The symmetric key used to encrypt and decrypt the fragment content; Ciphertext representing the encrypted fragment content.

[0061] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0062] (1) This invention takes symmetric searchable encryption technology and differential privacy technology as the core and designs a scenario of independent encryption for multiple users. The method of this invention breaks through the limitations of existing symmetric searchable encryption technology in access control and leakage assessment. It can efficiently realize dynamic access control of user permissions. At the same time, in the process of cross-user data search, it effectively alleviates the privacy degradation problem caused by user privacy level preferences, thereby meeting the higher security requirements in complex application scenarios.

[0063] (2) The method of this invention determines the privacy protection level of each user from the perspective of tolerable maximum leakage risk, alleviates the privacy degradation problem caused by the leakage of allowed pattern information in searchable encryption, and improves the reliability and security of the system. It supports pattern information hiding and dynamic permission updates, and can configure fine-grained obfuscation strategies according to the access needs of different users. At the same time, while protecting data confidentiality, it can also support the dynamic updating of user permissions, control the degree of pattern leakage in the query process, allow legitimate users to access their own data and the data of users with lower permission levels, and achieve secure and efficient data retrieval. Attached Figure Description

[0064] Figure 1 This is a system model diagram of the present invention;

[0065] Figure 2 This is a schematic diagram of the user access control relationship of the present invention;

[0066] Figure 3 This is a user search diagram of the present invention;

[0067] Figure 4 This is a schematic diagram of the key update in this invention;

[0068] Figure 5A schematic diagram is added to illustrate the user permissions of this invention;

[0069] Figure 6 This is a schematic diagram illustrating the deletion of user permissions according to the present invention;

[0070] Figure 7 A flowchart of the overall solution of the present invention. Detailed Implementation

[0071] The present invention will be further illustrated below with reference to specific embodiments. These embodiments are implemented based on the technical solutions of the present invention, and it should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

[0072] like Figure 1 As shown in the figure, the multi-user encrypted retrieval method with support for pattern leakage control and dynamic permission management provided in this embodiment mainly includes trusted authorities (TAs), users, and cloud servers.

[0073] Trusted Authorities (TAs): Develop and maintain access control policies for user data; assign data encryption keys to each user class and maintain private information about the encryption keys and auxiliary information for recovering encryption keys from other users.

[0074] It involves creating permission relationships between users, allocating keys to each node user, and maintaining access control policies.

[0075] Users are the producers and owners of private data. They can encrypt local data and upload it to cloud server storage, generate query tokens and request data queries, and change access permissions and request TA updates. Access permissions include updates to user encryption keys, additions, and deletions of access permissions.

[0076] Users who receive keys distributed by a trusted authorized authority encrypt local data using their private keys and upload the corresponding encrypted database to a cloud server. Authorized users within the system can then request data queries from the cloud server. Access to data in a user's child node depends on their user class's position within the access structure; otherwise, access is denied. Users can also request key updates from the cloud server.

[0077] Cloud servers provide remote data storage services, store user-uploaded data, and offer secure search services; including secure index and fragmented content storage, and fulfill user data query requests upon receiving search credentials.

[0078] In this embodiment, a defense design was implemented for the cloud server.

[0079] This invention provides a multi-user encrypted search method that supports pattern leakage control and dynamic permission management. It supports dynamic user permission changes and aims to achieve efficient and secure data retrieval. It constructs a secure and efficient user permission change mechanism, a pattern leakage controllable mechanism, and a pattern hiding mechanism. Specific functions include:

[0080] 1. User access control: This invention generates and maintains access control policies between users through a trusted authorization authority, assigns a key to each user node, and allows users to access their own data and the data of their child nodes according to their role permissions, thereby achieving fine-grained access management and ensuring that the access permissions of users with different roles are isolated from each other.

[0081] 2. Adjustable Pattern Leakage: This invention, through the data distribution of each user, allows users to calculate a reasonable level of privacy protection locally, based on the principle of the maximum tolerable leakage risk, and implement an obfuscation strategy of corresponding strength. Simultaneously, this method does not require a pre-established threat detection model; it relies solely on offline prior probabilities to provide users with adaptive privacy level guidance values.

[0082] 3. Pattern Information Hiding: This invention provides a pattern hiding mechanism, allowing cloud servers to store secure indexes with hidden pattern information. This solution can effectively defend against various attack methods such as word frequency attacks, while reducing the adversary's ability to guess underlying keywords while ensuring data privacy. It also guarantees system storage and computing efficiency, meeting the practical security application needs under potential adversary semi-honest server monitoring.

[0083] 4. Highly efficient encrypted search: This invention supports efficient retrieval of encrypted data, enabling users to quickly obtain the data they need without decrypting all encrypted content, thus balancing data availability and security.

[0084] 5. User Permission Update: This invention provides a user permission change mechanism. When a legitimate user's key is leaked, they can request a key update. After the key is updated, the data related to that user on the server will be updated, and the legitimate user who previously had access to that user's data will still retain their access rights. When an illegitimate user's data access rights are changed to legitimate user rights (i.e., user rights are increased), granting the user the right to view all data owned by the data owner, this dynamically adds user access permissions. Conversely, when a legitimate user's data access rights are changed to illegitimate user rights (i.e., user rights are deleted), preventing the user from viewing all data owned by the data owner, this dynamically deletes user access permissions. This improves the system's flexibility and reliability, meeting the practical application needs of dynamically changing environments while ensuring data confidentiality.

[0085] Through the above functions, this invention achieves flexible and efficient access control while protecting data confidentiality and hiding pattern information, significantly improving the security, reliability and practicality of the system, and is applicable to various scenarios such as cloud storage and data sharing.

[0086] In this embodiment, a Bayesian-based budget allocation mechanism is used to achieve the goal of controllable pattern leakage, but the specific implementation is not limited to this method.

[0087] This example requires the application of a multi-user searchable encryption method that supports pattern leakage control and user permission updates. This includes various related algorithms, as shown in Table 1 below.

[0088] Table 1 Relevant Algorithm Table

[0089]

[0090] Based on the above components, participating entities, and related technical methods, the specific implementation steps of the method of the present invention are as follows:

[0091] S1: Method initialization, the trusted authorization authority uses the KeyGen key generation algorithm to generate an encryption key for each user. Private information and auxiliary information ;

[0092] like Figure 2 As shown, TA runs KeyGen to generate a key for each user. Maintain the private information of the data owner's key. And auxiliary information for recovering the data owner's key. ;Specifically:

[0093] ,

[0094] in, This indicates auxiliary information used to recover the data owner's key. This represents a symmetric encryption algorithm in cryptography. The encryption key representing the data owner. Includes index encryption key and fragment encryption key .

[0095] Each user receives information generated at the TA. , , .

[0096] S2: User determines the obfuscation strength locally: The user determines the privacy protection level of the data by statistically analyzing the distribution characteristics of local data and using a Bayesian budget allocation mechanism.

[0097] First, users need to parse the file content into matching pairs of file identifiers and keywords to calculate the following two key metrics.

[0098] Then, by statistically analyzing the proportion of local files to global files, the first key indicator, namely the proportion of user data, is determined; and by statistically analyzing the proportion of user content and documents matching each keyword to the total number of local documents, the second key indicator, namely the keyword coverage rate, is determined.

[0099] Finally, using the two key indicators mentioned above as two prior probabilities in Bayesian analysis, different confusion strengths are determined for each key based on Bayes' theorem and user preferences, thereby establishing data associations among multiple users and mitigating pattern leakage issues caused by user privacy preferences. Specifically:

[0100] First, each user determines the proportion of their local documents. and keyword coverage The appropriate level of privacy protection is calculated as follows:

[0101] ,

[0102] in, It is the privacy protection strength, a parameter representing the level of obfuscation. This represents a function for calculating the confusion intensity.

[0103] This determines the level of privacy protection for information matching the keyword w.

[0104] S3: The user parses the index information locally. The user uses erasure coding technology to divide the file content into fragments for storage, and then parses the fragment content into fragment identifiers and keyword matching pairs.

[0105] Users ensure search accuracy by converting the processing objects from matching file identifiers with keywords to matching fragment identifiers with keywords. A false negative file indicates that the file is missing, and this is done to prevent a decrease in search accuracy.

[0106] First, the user uses erasure coding technology to split the content of a file into m... Fragmented storage. This technology requires only sufficient storage space. The number of fragments is enough to reconstruct the original file content, allowing for false negative fragments. Finally, the user needs to parse the fragmented content into matching pairs of fragment identifiers and keywords.

[0107] The data uploaded by users here is obfuscated index information. The obfuscation technique uses differential privacy (DP) to prevent DP from directly affecting the document ID. This leads to serious search accuracy problems, therefore erasure coding is used to remove the obfuscated objects from... Divide into fragment IDs ( ).

[0108] The segmentation technique requires the use of erasure coding algorithms. Implementation. Erasure coding divides the original document content into fragments and stores them, resulting in... The fragments are as follows:

[0109] ;

[0110] in, Indicates fragmented content; This indicates the file content that needs to be split; Indicates the number of fragments; This indicates the minimum number of fragments that can be used to recover the original document content; This represents the erasure coding algorithm.

[0111] Even if false negative fragments occur after DP obfuscation, the client can still rely on the retained fragments. The original file's content is reconstructed from fragments. Finally, the fragmented content is parsed into... With keywords Matching pairs .

[0112] S4: User-locally obfuscates index information. Based on the obfuscation strength of S2, the user adds noise to the matching pairs of processed fragment identifiers and keywords, and generates a noisy secure index through the indexing algorithm.

[0113] Legitimate users process their local index by adding noise to hide pattern information, then upload the noisy, secure index after encryption. Legitimate users then apply the determined obfuscation strength to the matching pairs of processed fragment identifiers and keywords to obtain a noisy fragment index, which is then encrypted using the user's key and uploaded.

[0114] Specifically: Each user, based on each keyword Calculated privacy protection strength In the matching pair To obfuscate information; firstly, users will perceive the privacy protection strength of a certain keyword. Transformed into the flip probability of implementing differential privacy obfuscation and (The conversion formula here can depend on the user's optimization goals and can be designed to support objectives such as minimizing storage overhead or maximizing search performance); then, the fragments containing the keyword are obfuscated, specifically in two ways: for each fragment containing the keyword, the user randomly generates a random number. If the random number If the keyword is present, the fragment is retained; otherwise, it is discarded. For fragments that do not contain the keyword, the user will randomly generate a random number. If the random number If the match is found, the fragment will be added to the matching pair with that keyword; otherwise, there will be no matching relationship. Finally, each user will receive the obfuscated, noisy index information. .

[0115] Then, each user calculates and processes the noisy, safe index locally based on the BldInx indexing algorithm, specifically as follows:

[0116] ;

[0117] in, This indicates a noisy safety index; This represents the key used to encrypt the noisy plaintext index.

[0118] S5: Authorized users upload fragment information: Users encrypt fragment content and security index, and upload it to the cloud server for storage;

[0119] Authorized users process local fragmented content, encrypt it using the fragment key, and then upload it along with the security index.

[0120] Each user computes and processes the noisy security index locally. .

[0121] Then each user uses a data encryption algorithm. The fragment ciphertext is calculated locally, specifically as follows:

[0122] ;

[0123] in, Ciphertext representing the encrypted fragment content; The symmetric key used to encrypt and decrypt the fragment content; This indicates fragmented content of the file.

[0124] Finally, each user will generate two encrypted messages locally. All data is uploaded to the server and stored in EDB.

[0125] S6: A legitimate user sends a query request. The user uses the key management mechanism to recover the authorized user key set based on private and auxiliary information, encrypts the key to generate a search token set, and uploads it to the cloud server.

[0126] Legitimate users can use the key management mechanism, along with their private and auxiliary information, to locally recover a set of user keys that are below their own privileges.

[0127] Using key derivation algorithm Enter user Private information Auxiliary information Low-privilege target users Finally, the client obtains the user's... encryption key .

[0128] If user Want to search for other users on the server (such as user) For the data, users need to directly use local private information. and auxiliary information Recover the authorized user's key ,in , and in among them Indicates user The index encryption key, Indicates user The fragmented content encryption key. Here, the user uses the same key recovery algorithm to recover the keys of the legitimate data owners one by one, obtaining the user's key set. .

[0129] Trapdoor generation algorithm Client enters key Keywords for the search Finally, the client obtains the search token. Using a trapdoor generation algorithm The user uses the recovered user key set Each user's key is used to encrypt the keyword of interest, thus obtaining a search credential for accessing the corresponding user's encrypted database. ,in, It is the generated search credentials. This represents the key used to encrypt the noisy plaintext index. This indicates keywords of interest. Users use the same trapdoor generation algorithm to obtain a set of search tokens that can be used in all legally accessible encrypted databases and upload them.

[0130] like Figure 3As shown, if user 3 wants to search for data from other users (user 5 and user 6) on the server, the user needs to directly use private information. and auxiliary information Recover the authorized user's key ,in, , In this way, legitimate users can use auxiliary information This will give you the key set of all authorized users. .

[0131] User 3 uses the index encryption key obtained from the recovered authorized user to encrypt the keywords of interest. Generate search token To facilitate querying other users' encrypted data on the server side, users will use all authorized keys. Encrypt the keyword Generate a set of search tokens for searching all authorized users. Upload it to the server.

[0132] S7: Authorized users can search data: Authorized users can query the data they have uploaded, submit query requests to the cloud server, and also query the data of user nodes with access permissions.

[0133] like Figure 3 As shown, the server received the set of search tokens. Then, it will search the corresponding user's encrypted dataset using a data search algorithm. Finally, the server obtains the matching index and matching documents corresponding to the search token. This is then returned to the client.

[0134] in, This indicates the matching result returned by the server; This indicates an encrypted database stored on the server side; This represents the set of search tokens.

[0135] S8: Legitimate User Decrypts Search Results: Legitimate users receive the noisy search results returned by the server and decrypt the fragmented ciphertext locally using the fragment key. Then, from the retained fragments (≥z) belonging to the same file, the user reconstructs the original file content using erasure coding decoding technology.

[0136] Decrypting each user's data:

[0137] S81: The user uses the recovered authorized user's fragment decryption key (the same as the encryption key) to decrypt and obtain the original fragment content, specifically:

[0138] ;

[0139] in, Indicates the data decryption algorithm; This represents the plaintext fragment obtained after decryption; The symmetric key used to encrypt and decrypt the fragment content; Ciphertext representing the encrypted fragment content.

[0140] S82: Users put plaintext fragments belonging to the same document into the same set. If the size of the set Greater than the document recovery threshold Then use erasure coding decoding algorithm. To restore the original file content, specifically:

[0141] ;

[0142] in, Indicates by The original file content was successfully recovered from the fragment; This indicates the total number of fragments into which the document content is divided; This indicates the minimum number of fragments that can be used to recover the original document content; This represents the set of all fragments belonging to the same document.

[0143] S9: User Permission Update: Authorized users can use the key management mechanism to provide dynamic permission management functions for each user in a multi-user scenario, such as updating user keys and adding or deleting user access permissions.

[0144] First, a user can submit a key update request, sending the updated key to the TA, or the TA can randomly generate an updated key for the user. Simultaneously, the TA updates all auxiliary information related to that user, maintaining the original user permission relationships while the cloud server updates the corresponding encrypted data.

[0145] Second, when a legitimate user is added to access the data, access permissions need to be added to the new user. The new user needs to use the new user's private information to maintain the data owner's encryption key and update it in the auxiliary information so that the legitimate user can restore the data owner's key before accessing the data.

[0146] Third, when deleting a legitimate user who has access to data, the original legitimate user's access permissions must also be deleted. First, the user updates the encryption key for the data owner and removes some auxiliary information related to the old key from the original auxiliary information. While maintaining the permission relationships of other users, the cloud server updates the corresponding encrypted data.

[0147] Key update and permission update:

[0148] like Figure 4 As shown, if user 2 wants to update the key TA will generate a random key. Replace the original key Legitimate users with permissions related to User 2 can access private information. Get updated auxiliary information Finally, TA updated the update assistance information for all legitimate users related to User 2. .

[0149] like Figure 5 As shown, if user 6 wants to add more information to user... Authorization relationship TA will use users Private information Maintain the encryption key of authorized user 6 Get updated auxiliary information Finally, the TA update includes permissions associated with the user. Auxiliary information .

[0150] like Figure 6 As shown, if user 6 wants to delete the user Authorization relationship First, it will update the authorized user. key Then delete the old key from the original auxiliary information. Related auxiliary information such as Finally, TA updated the accessibility information that user 6 did not have permission to associate with. .

[0151] The above embodiment takes a single keyword as an example. To expand to multiple keywords, simply add an encryption step for cross-items during the update and a judgment step for cross-items during the query.

[0152] This invention, based on symmetric searchable encryption methods, adds privacy protection features for pattern information and introduces a dynamic access control mechanism. While ensuring data confidentiality and supporting efficient data retrieval, it effectively mitigates the problem of attackers speculating on query keywords. This invention allows legitimate users to access their own data and data of users with lower privileges. This invention allows semi-honest servers to observe obfuscated noisy pattern information, preventing the leakage of true pattern information during data retrieval and improving system security and reliability.

[0153] This invention, while protecting the privacy of sensitive data, determines a reasonable privacy level allocation strategy, effectively mitigating the problem of sensitive information leakage caused by pattern information leakage during the query process, thereby significantly improving the system's security and flexibility. To achieve flexible data management, in the multi-user scenario described in this application, users can use private keys to encrypt local data.

[0154] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A multi-user encrypted retrieval method supporting pattern leakage control and dynamic permission management, characterized in that: Participating entities include trusted authorization authorities, users, and cloud servers, and the specific steps include: S1: Method initialization, the trusted authorization authority uses the KeyGen key generation algorithm to generate each user's encryption key, private information and auxiliary information; S2: The user determines the obfuscation strength locally. The user determines the privacy protection level of the data by statistically analyzing the distribution characteristics of local data and using a Bayesian-based budget allocation mechanism. S3: The user parses the index information locally. The user uses erasure coding technology to divide the file content into fragments for storage, and then parses the fragment content into fragment identifiers and keyword matching pairs. S4: User-locally obfuscates index information. Based on the obfuscation strength of S2, the user adds noise to the matching pairs of processed fragment identifiers and keywords, and generates a noisy, secure index using an indexing algorithm. The specific implementation process is as follows: The privacy protection strength calculated for each user based on each keyword w In the matching pair Implement obfuscation; First, the user will specify the privacy protection level of a certain keyword. Transformed into the flip probability of implementing differential privacy obfuscation and ; Then, the fragments containing the keyword are obfuscated, specifically in two ways: For fragments containing the keyword, the user will randomly generate a random number. If the random number If the keyword is present, the fragment is retained; otherwise, it is discarded. For fragments that do not contain the keyword, the user will randomly generate a random number. If the random number If the match is true, the fragment will be added to the matching pair with the keyword; otherwise, there will be no matching relationship. Finally, each user will receive the obfuscated, noisy index information. ; Then, each user processes the noisy security index based on the BldInx indexing algorithm, specifically as follows: ; Where I represents a noisy secure index; k represents the key used to encrypt the noisy plaintext index; S5: Upload fragment information, the user encrypts the fragment content and security index, and uploads it to the cloud server for storage; S6: Send a query request. The user uses the key management mechanism to recover the authorized user key set based on private and auxiliary information, encrypts the key to generate a search token set and uploads it to the cloud server. S7: When a user searches for data, the cloud server retrieves the encrypted database based on a set of search tokens and returns the search results; the specific implementation process is as follows: Using the KeyDer key derivation algorithm, the authorized user key set is recovered based on private and auxiliary information; If user u wants to search for other users' data on the server, the user needs to directly use local private information. and auxiliary information Recover the authorized user's key ,in , Indicates user The index encryption key, Indicates user Encryption key for fragmented content; The user uses the same key recovery algorithm to recover the keys of each legitimate data owner who has access, thus obtaining a user key set. ; Truder trapdoor generation algorithm: Client input key The search term is "w". Finally, the client receives a search token. Using a trapdoor generation algorithm The user uses the recovered user key set Each user's key is used to encrypt the keyword of interest, thus obtaining a search credential for accessing the corresponding user's encrypted database. ,in, This represents the generated search credentials, and k represents the key used to encrypt the noisy plaintext index. Keywords indicating interest; Users use the same trapdoor generation algorithm to obtain a set of search tokens that can be used to access all legally accessible encrypted databases and upload them; S8: Decrypt search results. Users decrypt fragmented content in search results and reconstruct the original file using erasure coding decoding technology. S9: User permission update. The trusted authorization authority responds to the user's request, performs key update, permission addition or deletion operations, and updates auxiliary information and corresponding data in the cloud server.

2. The multi-user encrypted retrieval method supporting pattern leakage control and dynamic permission management according to claim 1, characterized in that: In S1, the specific implementation process is as follows: Trusted authorization authorities run the KeyGen key generation algorithm to generate keys for each user and maintain private information about the data owner's keys. And auxiliary information for recovering the data owner's key. ;Specifically: ; in, This indicates auxiliary information used to recover the data owner's key. This represents a symmetric encryption algorithm in cryptography. The encryption key representing the data owner. Includes index encryption key and fragment encryption key .

3. The multi-user encrypted retrieval method supporting pattern leakage control and dynamic permission management according to claim 1, characterized in that: In S2, the specific implementation process is as follows: First, users parse the file content into matching pairs of file identifiers and keywords; then, they determine the user data percentage and keyword coverage; each user then bases their analysis on the percentage of local documents. and keyword coverage The appropriate level of privacy protection is calculated as follows: ; in, The value indicates the level of privacy protection; Obfuscate() is a function that calculates the level of obfuscation.

4. The multi-user encrypted retrieval method supporting pattern leakage control and dynamic permission management according to claim 1, characterized in that: In S3, the specific implementation process is as follows: The user uses erasure coding technology to divide the content of a file into m fragments for storage, specifically: ; Where S represents the fragment content; D represents the file content that needs to be split; m represents the number of fragments; z represents the minimum number of fragments that can be used to recover the original document content; ErasureCode.Encode() represents the erasure coding algorithm; Then the segmented content is parsed into Matching pairs with keyword w , For fragment ID, This is a set of keywords.

5. The multi-user encrypted retrieval method supporting pattern leakage control and dynamic permission management according to claim 1, characterized in that: In S5, the specific implementation process is as follows: Users process local fragmented content, encrypt it using the fragment key, and then upload it along with the secure index. Based on the SymEnc data encryption algorithm, each user calculates the fragmented ciphertext locally, specifically as follows: ; Where c represents the ciphertext of the encrypted fragment content; S represents the symmetric key used to encrypt and decrypt the fragment content; S represents the fragment content.

6. The multi-user encrypted retrieval method supporting pattern leakage control and dynamic permission management according to claim 1, characterized in that: In S8, the specific implementation process is as follows: S81: The user uses the fragment decryption key obtained from the recovered authorized user to decrypt and obtain the original fragment content; S82: Users put plaintext fragments belonging to the same document into the same set. If the size of the set Greater than the document recovery threshold Then use erasure coding decoding algorithm. To restore the original file content, specifically: ; in, Indicates by The original file content was successfully recovered from the fragment; This indicates the total number of fragments into which the document content is divided; This indicates the minimum number of fragments that can be used to recover the original document content; This represents the set of all fragments belonging to the same document.

7. The multi-user encrypted retrieval method supporting pattern leakage control and dynamic permission management according to claim 6, characterized in that: In S81, specifically: ; in, Indicates the data decryption algorithm; This represents the plaintext fragment obtained after decryption; The symmetric key used to encrypt and decrypt the fragment content; Ciphertext representing the encrypted fragment content.

Citation Information

Patent Citations

  • Safe and efficient cross-user ciphertext retrieval and customized data sharing method

    CN115879158A

  • Safe and efficient sorting searchable encryption method and system in cloud environment

    CN117914603A